US20050289089A1 - Methods for multi-class cost-sensitive learning - Google Patents
Methods for multi-class cost-sensitive learning Download PDFInfo
- Publication number
- US20050289089A1 US20050289089A1 US10/876,533 US87653304A US2005289089A1 US 20050289089 A1 US20050289089 A1 US 20050289089A1 US 87653304 A US87653304 A US 87653304A US 2005289089 A1 US2005289089 A1 US 2005289089A1
- Authority
- US
- United States
- Prior art keywords
- cost
- instance
- learning
- class
- sensitive
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10—TECHNICAL SUBJECTS COVERED BY FORMER USPC
- Y10S—TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10S706/00—Data processing: artificial intelligence
- Y10S706/902—Application using ai with detail of the ai system
- Y10S706/932—Mathematics, science, or engineering
Definitions
- the present invention generally relates to the field of cost-sensitive learning in the areas of machine learning and data mining and, more particularly, to methods for solving multi-class cost-sensitive learning problems using a binary classification algorithm.
- This algorithm is based on techniques of data space expansion and gradient boosting with stochastic ensembles.
- Classification in the presence of varying costs associated with different types of misclassification is important for practical applications, including many data mining applications, such as targeted marketing, fraud and intrusion detection, among others. Classification is often idealized as a problem where every example is equally important, and the cost of misclassification is always the same. The real world is messier. Typically, some examples are much more important than others, and the cost of misclassifying in one way differs from the cost of misclassifying in another way. A body of work on this subject has become known as cost-sensitive learning, in the areas of machine learning and data mining.
- the first category is concerned with making particular classifier learners cost-sensitive, including methods specific for decision trees (see, for example., U. Knoll, G. Nakhaeizadeh, and B. Tausend, “Cost-sensitive pruning of decision trees”, Proceedings of the Eight European Conference on Machine Learning, pp. 383-386, 1994, and J. Bradford, C. Kunz, R. Kohavi, C. Brunk, and C. Brodley, “Pruning decision trees with misclassification costs”, Proceedings of the European Conference on Machine Learning, pp. 131-136, 1998), neural networks (see, for example, P. Geibel and F.
- the methods are based on example weighting schemes that are derived using two key ideas: 1) data space expansion and 2) gradient boosting with stochastic ensembles.
- the latter is a formal framework that give rise to a coherent body of methods.
- One of the methods of invention which is based on the idea 1) above, works by repeatedly sampling from the expanded data set, which is obtained by enhancing each example in the original data set with as many data points as there are possible labels for any single instance. It then repeatedly draws sub-sample from this expanded data set using weighted sampling according to a certain example weighting scheme, in which each labeled example is given the weight specified as the difference between the maximum possible misclassification cost for the instance in question and the misclassification associated with the label in the particular labeled example. The example weighting remains constant throughout the iterative sampling procedure. It then finally outputs a classifier hypothesis which is the average of all the hypotheses output in the respective iterations.
- Another one of the methods of invention which is based on the idea 2) above, works by iteratively applying weighted sampling from the same expanded data set, using a different weighting scheme.
- the weighting scheme of this method gives each labeled example the weight specified as the difference between the average cost on that instance by the averaged hypotheses from the iterations so far and the misclassification cost associated with the label in the labeled example in question.
- the weighting changes in every iteration, since it depends on the performance of the averaged hypothesis obtained up to the current iteration.
- the example weights used in this method can be both positive and negative, since the label given in any labeled example does not necessarily correspond to the best label for the given instance, i.e.
- the results of the methods of invention are obtained by outputting all of the classifier representations obtained through the iterations, and represent the average over them.
- These representations can be arbitrary representations of classifiers, such as decision trees, neural networks and support vector machines, for the problem at hand, such as network intrusion detection, fraud detection, targeted marketing, credit risk rating, among other things.
- each one of these representations could be a decision tree that specifies a set of conditions on various attributes of a network connection event, which together signal certain types of network intrusion.
- Such representations can be further applied on a new network connection to output judgment whether or not the connection is to be suspected to be some type of an intrusion attempt with reasonable likelihood, and decisions can be based on this judgment to determine the appropriate course of action, such as denial of service or probing.
- FIG. 1 is a block diagram showing the architecture of the system implementing one of the methods according to the invention
- FIG. 2 is a flow chart showing the logic of the method for multi-class cost-sensitive learning implemented on the system shown in FIG. 1 ;
- FIG. 3 is a block diagram showing the architecture of the system implementing another one of the methods according to the invention.
- FIG. 4 is a flow chart showing the logic of the method for multi-class cost-sensitive learning implemented on the system shown in FIG. 3 ;
- FIG. 5 is an example of a decision tree to illustrate the process implemented by the invention.
- a popular formulation of the cost-sensitive learning problem is via the use of a cost matrix.
- a cost matrix, C(y 1 , y 2 ) specifies how much cost is incurred when misclassifying an example labeled y 2 as y 1 , and the goal of a cost-sensitive learning method is to minimize the expected cost.
- Zadrozny and Elkan B. Zadrozny and C. Elkan, “Learning and making decisions when costs and probabilities are both unknown”, Proceedings of the seventh International Conference on Knowledge Discovery and Data Mining, pp.
- examples of the form (x, y, c) are drawn from a distribution D over a domain X ⁇ Y ⁇ R + .
- the goal is to find a classifier h:X ⁇ Y having minimum importance-weighted misclassificaton error: arg ⁇ ⁇ min h ⁇ E ( x , y , c ) - D ⁇ c ⁇ I ⁇ ( h ⁇ ( x ) ⁇ y )
- a learning method attempts to meet this goal by minimizing the empirical weighted error in some hypothesis class : arg ⁇ ⁇ min h ⁇ H ⁇ ⁇ ( x , y , c ) ⁇ S ⁇ c ⁇ I ⁇ ( h ⁇ ( x ) ⁇ y ) ( 3 )
- importance weighted classification can be solved very well with a classification method, by use of weighte
- stochastic cost-sensitive learning is defined as that of minimizing the following expected cost: arg ⁇ ⁇ min h ⁇ E D ⁇ ⁇ y ⁇ Y ⁇ C ⁇ ( x , y ) ⁇ h ⁇ ( x , y ) Note that in the special case that h is deterministic, this formulation is equivalent to the definition given in Equation (1).
- Our methodology can be interpreted as a reduction, which translates a multi-class cost-sensitive learning problem to a classifier learning problem. That is, it allows us to solve the cost-sensitive learning problem using an arbitrary classifier learning method as a component algorithm.
- This methodology is derived using two key ideas: 1) expanding data space and 2) gradient boosting with stochastic ensembles.
- Theoretical performance guarantee on a particular variant of the invented methodology is derived using a convexification of the objective function by the expected cost function. Below we will explain these two key ideas by exhibiting a prototypical method based on each.
- a representative method in the prior art of iterative methods for cost-sensitive learning is the method proposed in Zadrozny, Langford and Abe, ibid., called costing.
- the weighting scheme of this method exploits the following observation:
- C(x, y 2 ) can be further reduced to a formulation in terms of a single importance number per example. This is possible by associating a number indicating the importance of an example (x, y 2 ), given by
- This conversion allows us to reduce the cost-sensitive learning problem to a weighted classifier learning problem, but it has not been known how that would be done for the multi-class scenario. It is therefore natural to consider iterative weighting schemes, in which example weights are iteratively modified in search for the optimal weighting.
- H t denote the mixture hypothesis of the learning procedure at round t.
- ⁇ C ⁇ 1 W ⁇ ⁇ x ⁇ S ⁇ ⁇ y ⁇ Y ⁇ ( C ⁇ ( x , H t - 1 ⁇ ( x ) ) - C ⁇ ( x , y ) ) ⁇ h ⁇ ( x , y ) ( 9 )
- the weak learner now receives both positive and negative weights. While the minimization of weighted misclassification with positive and negative weights makes perfect sense as an optimization problem, its interpretation as a classification problem is not immediately clear. In particular, it prohibits the use of weighted sampling as a means of realizing the weighted classification problem.
- GBSE-T has a boosting property given a version of weak learning condition on the component classifier.
- This weak learning condition which we make precise below, is one that is sensitive to class imbalance.
- ) m where l I(w ⁇ 0).
- the expected empirical cost of H t equals the following, where we drop the subscript t from ⁇ t .
- the binary classifier used as the component learner is likely to be given weighted sample with balanced positive and negative examples. As the number of iterations increases and progress is made, however, it will receive samples that are increasingly more negative. (This is because the positive examples correspond to labels that can further improve the current performance.) It therefore becomes easier to attain high weighted accuracy by simply classifying all examples to be negative.
- the weak learning condition of Equation 12 appropriately deals with this issue, as it requires that the weak learner achieve better weighted accuracy than that attainable by assigning all examples to the negative class.
- the weighted sampling can be done in two steps; the instance is sampled in the first step according to a probability proportional to max y ⁇ w x,y and then choosing the label y with a probability proportional to
- the weighted sampling can be done in two steps; the instance is sampled in the first step according to the same probability as above, and for the chosen instance, examples are deterministically added for all possible labels.
- FIG. 1 there is shown a system on which a method for multi-class, cost-sensitive learning according to the invention may be implemented.
- This system comprises a multi-class cost-sensitive learning top control module 1 which controls the overall control flow, making use of various sub-components of the system.
- a learning algorithm storage module 2 stores a representation of an algorithm for classification learning.
- An arbitrary algorithm for classification can be used here.
- the learning algorithm can be a decision tree learning algorithm, a na ⁇ ve Bayes method, a logistic regression method or neural networks.
- the model output module 3 stores the models obtained as a result of applying the learning algorithm stored in module 2 to training data given by weighted sampling module 4 and outputs a final model by aggregating these models.
- the weighted sampling module 4 accesses the data stored in data storage module 7 , samples a relatively small subset of the data with acceptance probability determined using the example weights, and passes the obtained sub-sample to module 1 .
- the weight update module 5 updates the example weights for sampling using a particular function determined by the current weights and current models.
- the model update module 6 updates the current model using the model's output in the previous iterations stored in the current model storage module 8 and the output model of the current iteration output by module 3 and stores the resulting updated model in module 8 .
- FIG. 2 shows a flow diagram of the process implemented in the system of FIG. 1 .
- the first three steps initialize the process.
- expanded data T is initialized using the input data S.
- H 0 is initialized by setting for all (x,y) in T.
- the decision block in Step 26 determines if there is more data in T or a STOP condition has been met. If not, in Step 27 , (x,y) is sampled from T and accepted with a probability proportional to
- Step 26 A return is then made to the decision block in Step 26 .
- the process goes to Step 29 where the learning algorithm is run on T t to obtain model h t .
- Step 30 f t is set equal to stoch(h t ).
- Step 31 ⁇ t is chosen and H t is set equal to (1 ⁇ t )H t-1 + ⁇ t f t .
- FIG. 3 shows a system on which another method for multi-class, cost-sensitive learning according to the invention may be implemented.
- This system is similar to that shown in FIG. 1 and comprises a multi-class cost-sensitive learning top control module 1 which controls the overall control flow, making use of various sub-components of the system, a learning algorithm storage module 2 , which stores a representation of an algorithm for classification learning, a model output module 3 , which stores the models obtained as a result of applying the learning algorithm stored in module 2 to training data given by weighted sampling module 4 and outputs a final model by aggregating these models, and a weighted sampling module 4 , which accesses the data stored in data storage module 7 , samples a relatively small subset of the data with acceptance probability determined using the example weights, and passes the obtained sub-sample to module 1 .
- a learning algorithm storage module 2 which stores a representation of an algorithm for classification learning
- a model output module 3 which stores the models obtained as a result of applying the learning algorithm stored in module 2 to
- the weight calculation module 5 ′ replaces the weight upadate module 5 , which updates the example weights for sampling using a dynamically changing weighting scheme.
- the model update module 6 updates the current model using the model's output in the previous iterations stored in the current model storage module 8 and the output model of the current iteration output by module 3 and stores the resulting updated model in module 8 .
- FIG. 4 shows a flow diagram of the process implemented in the system of FIG. 3 .
- the first step initializes the process.
- expanded data T is initialized using the input data S.
- Step 42 the weights for all (x,y) in Tare set.
- the iteration begins in the decision block of Step 43 .
- Step 44 The process then loops back to decision block in Step 44 until there is either no more data in T or a stop condition has been met.
- the learning algorithm is run in Step 47 on T′ to obtain a model h t .
- Network intrusion detection has recently become a proto-typical application problem for multi-class, cost-sensitive learning.
- the multi-class aspect is essential because in this application there are typically more than one level of intrusion detection, such as probing and denial of service.
- the cost-sensitive aspect is important because vastly different costs are associated with different types of misclassification (e.g., false negatives are usually a magnitude more costly than false positives) and it is absolutely critical that any learning method used to derive an intrusion detection rule is sensitive to this cost structure.
- a network intrusion detection system based on the method and system of the invention for multi-class, cost-sensitive learning consists of the following steps:
- a typical set of features used to transform connection data into a well-defined feature vector is that used in the network intrusion data set known as “KDD CUP 99” data, which is publically available.
- KDD CUP 99 network intrusion data set
- Basic Features of Individual TCP Connections feature name description type duration length (number of seconds) of the continuous connection protocol_type type of protocol, e.g., TCP, UDP, etc. discrete service network service on the destination, e.g., discrete http, telnet, etc.
- hot number of “hot” indicators continuous num_failed_logins number of failed login attempts continuous logged_in 1 if successfully logged in; 0 discrete otherwise num_compromised number of “compromised” conditions continuous root_shell 1 if root shell is obtained; 0 discrete otherwise su_attempted 1 if “su-root” command attempted; 0 discrete otherwise num_root number of “root” accesses continuous num_file_creations number of file creation operations continuous num_shells number of shell prompts continuous num_access_files number of operations on access continuous control files num_outbound_cmds number of outbound commands in an continuous ftp session is_hot_login 1 if the login belongs to the “hot” discrete list; 0 otherwise is_guest_login 1 if the login is a “guest” login; 0 discrete otherwise
- FIG. 1 and the flow chart of FIG. 2 illustrate a preferred embodiment of the invention, which corresponds to the method “GBSE” described herein.
- the method “DSE”, also described herein may be used in the alternative.
- DSE the sampling weights remain unchanged throughout all iterations. Consequently, the modules and funcationalities that are related to weight updating are unnecessary.
- MetaCost Another deviation is that we use a variant of MetaCost that skips the last step of learning a classifier on a relabeled training data set. It has been observed that this variant performs at least as well as MetaCost, in terms of cost minimization. (This variant has been called BagCost by D. Margineantu in Methods for Cost-Sensitive Learning, PhD thesis, Department of Computer Science, Oregon State University, Corvallis, Oreg., 2001.) Also, in our implementation of AvgCost, we perform weighted sampling multiple times to obtain an emsemble of hypotheses, then output their average as the final hypothesis.
- ⁇ circumflex over (P) ⁇ (y 1 ) and ⁇ circumflex over (P) ⁇ (y 2 ) be the empirical probabilities of occurrence of classes y 1 and y 2 in the training data.
- the diagonal entries were then chosen from the interval [0,1000], which often leads to cost matrices in which the correct label is not the least costly one. Besides being unreasonable (see C.
- Table 2 lists the average costs attained by each of these methods on the 6 data sets, and their stand errors. These results were obtained by averaging over 20 runs, each run consisting of 30 iterations of the respective learning method. These results appear quite convincing: GBSE out-performs all comparison methods on all data sets, except on Splice, for which it ranks second after MetaCost. Also, GBSE is the best performing among the proposed methods, confirming our claim that the combination of various techniques involved is indeed necessary to attain this level of performance.
- Table 3 lists the average total data size used by each of the methods in 30 iterations. Examining these results in conjunction with the data characteristics in Table 1 reveals a definite trend.
- the data sets are divided into to groups: those having very large skews, or very low class ratios (Annealing, KDD-99 and Solar flare), and those having moderate skews (Satellite, Splice and Letter).
- the methods based on example weighting (AvgCost, GBSE, DSE) use magnitudes smaller data sizes for the three data sets in the first group (i.e., with large skews), as compared to other methods, Bagging and MetaCost.
- the performance of GBSE is especially spectacular on this group, achieving much lower costs while requiring very small data sizes. It is worth mentioning that it is these data sets in the first group with large skews that require cost-sensitive learning the most.
Abstract
Methods for multi-class cost-sensitive learning are based on iterative example weighting schemes and solve multi-class cost-sensitive learning problems using a binary classification algorithm. One of the methods works by iteratively applying weighted sampling from an expanded data set, which is obtained by enhancing each example in the original data set with as many data points as there are possible labels for any single instance, using a weighting scheme which gives each labeled example the weight specified as the difference between the average cost on that instance by the averaged hypotheses from the iterations so far and the misclassification cost associated with the label in the labeled example in question. It then calls the component classification algorithm on a modified binary classification problem in which each example is itself already a labeled pair, and its (meta) label is 1 or 0 depending on whether the example weight in the above weighting scheme is positive or negative, respectively. It then finally outputs a classifier hypothesis which is the average of all the hypotheses output in the respective iterations.
Description
- 1. Field of the Invention
- The present invention generally relates to the field of cost-sensitive learning in the areas of machine learning and data mining and, more particularly, to methods for solving multi-class cost-sensitive learning problems using a binary classification algorithm. This algorithm is based on techniques of data space expansion and gradient boosting with stochastic ensembles.
- 2. Background Description
- Classification in the presence of varying costs associated with different types of misclassification is important for practical applications, including many data mining applications, such as targeted marketing, fraud and intrusion detection, among others. Classification is often idealized as a problem where every example is equally important, and the cost of misclassification is always the same. The real world is messier. Typically, some examples are much more important than others, and the cost of misclassifying in one way differs from the cost of misclassifying in another way. A body of work on this subject has become known as cost-sensitive learning, in the areas of machine learning and data mining.
- Research in cost-sensitive learning falls into three main categories. The first category is concerned with making particular classifier learners cost-sensitive, including methods specific for decision trees (see, for example., U. Knoll, G. Nakhaeizadeh, and B. Tausend, “Cost-sensitive pruning of decision trees”, Proceedings of the Eight European Conference on Machine Learning, pp. 383-386, 1994, and J. Bradford, C. Kunz, R. Kohavi, C. Brunk, and C. Brodley, “Pruning decision trees with misclassification costs”, Proceedings of the European Conference on Machine Learning, pp. 131-136, 1998), neural networks (see, for example, P. Geibel and F. Wysotzki, “Perceptron based learning with example dependent and noisy costs”, Proceedings of the Twentieth International Conference on Machine Learning, 2003), and support vector machines (see, for example, G. Fumera and F. Roli, “Cost-sensitive learning in support vector machines”, VIII Convegno Associazione Italiana per L'Intelligenza Artificiale, 2002). The second category uses Bayes risk theory to assign each example to its lowest expected cost class (see, for example, P. Domingos, “MetaCost: A general method for making classifiers cost sensitive”, Proceedings of the Fifth International Conference on Knowledge Discovery and Data Mining, pp. 144-164, ACM Press, 1999, and D. Margineantu, Methods for Cost-Sensitive Learning, PhD thesis, Department of Computer Science, Oregon State University, Corvallis, 2001). This requires classifiers to output class membership probabilities and sometimes requires estimating costs (see, B. Zadrozny and C. Elkan, “Learning and making decisions when costs and probabilities are both unknown”, Proceedings of the Seventh International Confernece on Knowledge Discovery and Data Mining, pp. 204-213, ACM Press, 2001) (when the costs are unknown at classification time). The third category concerns methods that modify the distribution of training examples before applying the classifier learning method, so that the classifier learned from the modified distribution is cost-sensitive. We call this approach cost-sensitive learning by example weighting. Work in this area includes stratification methods (see, for example, P. Chan and S. Stolfo, “Toward scalable learning with non-uniform class and cost distributions”, Proceedings of the Fourth International Conference on Knowledge Discovery and Data Mining, pp. 164-168, 1998, and L. Breiman, J. H. Friedman, R. A. Olsen, and C. J. Stone, Classification and Regression Trees, Wadsworth International Group, 1984) and the costing algorithm (see, for example, B. Zadrozny, J. Langford, and N. Abe, “Cost-sensitive learning by cost-proportionate example weighting”, Proceedings of the Third IEEE International Conference on Data Mining, pp. 435-442, 2003). This approach is very general since it reuses arbitrary classifier learners and does not require accurate class probability estimates from the classifier. Empirically this approach attains similar or better cost-minimization performance.
- Unfortunately, current methods in this category suffer from a major limitation: they are well-understood only for two-class problems. In the two-class case, it is easy to show that each example should be weighted proportionally to the difference in cost between predicting correctly or incorrectly (see, again, Zadrozny et al., ibid.). However, in the multi-class case there is more than one way in which a classifier can make a mistake, breaking the application of this simple formula. Heuristics, such as weighting examples by the average misclassification cost, have been proposed (see, again, Breiman et al., ibid., and the Margineantu thesis, ibid.), but they are not well-motivated theoretically and do not seem to work very well in practice when compared to methods that use Bayes risk minimization (see, again, Domingos, ibid.).
- It is therefore an object of the present invention to provide a method for multi-class cost-sensitive learning based on an example weighting scheme.
- According to the invention, the methods are based on example weighting schemes that are derived using two key ideas: 1) data space expansion and 2) gradient boosting with stochastic ensembles. The latter is a formal framework that give rise to a coherent body of methods.
- One of the methods of invention, which is based on the idea 1) above, works by repeatedly sampling from the expanded data set, which is obtained by enhancing each example in the original data set with as many data points as there are possible labels for any single instance. It then repeatedly draws sub-sample from this expanded data set using weighted sampling according to a certain example weighting scheme, in which each labeled example is given the weight specified as the difference between the maximum possible misclassification cost for the instance in question and the misclassification associated with the label in the particular labeled example. The example weighting remains constant throughout the iterative sampling procedure. It then finally outputs a classifier hypothesis which is the average of all the hypotheses output in the respective iterations.
- Another one of the methods of invention, which is based on the idea 2) above, works by iteratively applying weighted sampling from the same expanded data set, using a different weighting scheme. The weighting scheme of this method gives each labeled example the weight specified as the difference between the average cost on that instance by the averaged hypotheses from the iterations so far and the misclassification cost associated with the label in the labeled example in question. Emphatically, the weighting changes in every iteration, since it depends on the performance of the averaged hypothesis obtained up to the current iteration. Additionally, the example weights used in this method can be both positive and negative, since the label given in any labeled example does not necessarily correspond to the best label for the given instance, i.e. the label with the minimum cost, due to the use of data space expansion. Negative weights do not admit the use of weighted sampling. The method deals with this problem by calling the component classification algorithm on a modified binary classification problem in which each example is itself already a labeled pair, and its (meta) label is 1 or 0 depending on whether the example weight in the above weighting scheme is positive or negative, respectively.
- The results of the methods of invention are obtained by outputting all of the classifier representations obtained through the iterations, and represent the average over them. These representations can be arbitrary representations of classifiers, such as decision trees, neural networks and support vector machines, for the problem at hand, such as network intrusion detection, fraud detection, targeted marketing, credit risk rating, among other things. For example, in the application to network intrusion detection, each one of these representations could be a decision tree that specifies a set of conditions on various attributes of a network connection event, which together signal certain types of network intrusion. Such representations can be further applied on a new network connection to output judgment whether or not the connection is to be suspected to be some type of an intrusion attempt with reasonable likelihood, and decisions can be based on this judgment to determine the appropriate course of action, such as denial of service or probing.
- The foregoing and other objects, aspects and advantages will be better understood from the following detailed description of a preferred embodiment of the invention with reference to the drawings, in which:
-
FIG. 1 is a block diagram showing the architecture of the system implementing one of the methods according to the invention; -
FIG. 2 is a flow chart showing the logic of the method for multi-class cost-sensitive learning implemented on the system shown inFIG. 1 ; -
FIG. 3 is a block diagram showing the architecture of the system implementing another one of the methods according to the invention; -
FIG. 4 is a flow chart showing the logic of the method for multi-class cost-sensitive learning implemented on the system shown inFIG. 3 ; and -
FIG. 5 is an example of a decision tree to illustrate the process implemented by the invention. - We begin by introducing some general concepts and notation we use in the rest of the description.
- A popular formulation of the cost-sensitive learning problem is via the use of a cost matrix. A cost matrix, C(y1, y2), specifies how much cost is incurred when misclassifying an example labeled y2 as y1, and the goal of a cost-sensitive learning method is to minimize the expected cost. Zadrozny and Elkan (B. Zadrozny and C. Elkan, “Learning and making decisions when costs and probabilities are both unknown”, Proceedings of the seventh International Conference on Knowledge Discovery and Data Mining, pp. 204-213, ACM Press, 2001) noted that this formulation is not applicable in situations in which misclassification costs depend on particular instances, and proposed a more general form of cost function, C(x, y1 , y2), that allows dependence on the instance x. Here we adopt this general formulation, but note that in the reasonable case in which the cost is minimized by the true label, we can drop the redundant information y2 and write C(x, y1) for C(x, y1, y2).
- Once we allow the costs to depend on each example, it is natural to assume that the costs are generated according to some distribution, along with the examples, which leads to the following formulation. In (multi-class) cost sensitive classification, examples of the form (x, (C(x, y1), . . . , C(x, yk)) are drawn from a distribution D over a domain X×R+
k . (Throughout, we will let k denote |Y|.) Given a set of examples, S=(xi, (C(xi, y))yεY)m, the goal is to find a classifier h:X→{1, . . . , k} which minimizes the expected cost of the classifier:
We can assume without loss of generality that the costs are normalized so that
Note that with this normalization, the above formulation of cost is equivalent to the common fonnulation in terms of misclassification cost, i.e.,
Nonmally a learning method attempts to do this by minimizing the empirical cost in the given training data, given some hypothesis class :
We note that we sometimes use the empirical expectation notation, Ê, to refer to the averaged empirical cost, namely
As a building block of our method, we make use of methods for solving importance weighted classification problems, which we define below. In importance weighted classification, examples of the form (x, y, c) are drawn from a distribution D over a domain X×Y×R+. Given a set of examples S=(x, y, c)m, the goal is to find a classifier h:X→Y having minimum importance-weighted misclassificaton error:
Again, usually, a learning method attempts to meet this goal by minimizing the empirical weighted error in some hypothesis class :
We note that importance weighted classification can be solved very well with a classification method, by use of weighted rejection sampling techniques (see,again, Zadronzny, Langford, and Abe, ibid.). - In the above, we assumed that the hypotheses output by a cost-sensitive learner is a functional hypothesis h, i.e., h:X→Y. It is also possible to allow hypotheses that are stochastic, namely
h:X×Y→[0,1]
subject to the stochastic condition:
With stochastic hypotheses, stochastic cost-sensitive learning is defined as that of minimizing the following expected cost:
Note that in the special case that h is deterministic, this formulation is equivalent to the definition given in Equation (1). Also, this is a convexification of the standard objective function that we usually expect a stochastic cost-sensitive learner to minimize, i.e.,
We also consider a variant of cost-sensitive learning in which relational hypotheses are allowed. Here relational hypotheses h are relations over X×Y, i.e., h:X×Y→{0, 1}. In general h is neither functional nor stochastic, and in particular it may violate the stochastic condition, - We often use the more general notation of h(x,y), meant for stochastic and relational hypotheses, even when h is a deterministic function from X to Y. As notational shorthand, for a stochastic hypothesis h, we write h(x) to denote h(x, ·):Y→[0,1], and C(x, h(x)) to denote the expected cost of its predictions, i.e.,
Finally, we note that we often write “x ε S” as a shorthand for “y ε Y(x,y) ε S”. - Our methodology can be interpreted as a reduction, which translates a multi-class cost-sensitive learning problem to a classifier learning problem. That is, it allows us to solve the cost-sensitive learning problem using an arbitrary classifier learning method as a component algorithm. This methodology is derived using two key ideas: 1) expanding data space and 2) gradient boosting with stochastic ensembles. Theoretical performance guarantee on a particular variant of the invented methodology is derived using a convexification of the objective function by the expected cost function. Below we will explain these two key ideas by exhibiting a prototypical method based on each.
- A representative method in the prior art of iterative methods for cost-sensitive learning is the method proposed in Zadrozny, Langford and Abe, ibid., called costing. The weighting scheme of this method exploits the following observation: For the binary class case, the above formulation in terms of cost per example, C(x, y2), can be further reduced to a formulation in terms of a single importance number per example. This is possible by associating a number indicating the importance of an example (x, y2), given by |C(x, 0)−C(x, 1)|. This conversion allows us to reduce the cost-sensitive learning problem to a weighted classifier learning problem, but it has not been known how that would be done for the multi-class scenario. It is therefore natural to consider iterative weighting schemes, in which example weights are iteratively modified in search for the optimal weighting.
- A straightforward application of iterative weighting suffers from an inability to directly take into account the different costs associated with multiple ways of misclassifying examples. This translates to non-convergence of the method in practice. We address this issue by the technique of expanding data space, the first of the two key ideas.
- Data Space Expansion
- The objective of minimizing the empirical cost on the original training sample is equivalent to minimization on the following expanded sample. Given a labeled sample S consisting of (x, (C(x,y))yεY) of size m, we define a sample S′ of size mk for classification, where k is the size of the label set, i.e., k=|Y|, as follows.
Minimizing the importance weighted loss,
on this new dataset also minimizes the cost on our original sample. The algorithm DSE (Data Space Expansion) takes advantage of this observation, which is summarized below as a theorem. -
THEOREM 1. With the definitions given inFIG. 3 , a hypothesis IL minimizing the weighted classification error on the expanded weighted sample S′,
also minimizes the cost on the original sample S,
Gradient Boosting with Stochastic Ensembles - Having described the idea of data space expansion, we now cmbine it with the gradient boosting framework to arrive at our main method. In particular, we cast the stochastic multiclass cost-sensitive learning in the framework of gradient boosting (see L. Mason, J. Baxter, P. Barlett, and M. Frean, “Boosting algorithms as gradient descent”, Advances in Neural Information Processing Systems 12, pp. 512-518, 2000), with the objective function defined as the expected cost of the stochastic ensemble, obtained as a mixture of individual hypotheses, on the expanded data set. As we stated above, a functional hypothesis of the form h:X→Y can be viewed as a special case of a stochastic hypothesis. We then define a stochastic ensemble hypothesis H, given multiple functional hypotheses, ht, t=1, . . . , T, as the conditional distribution defined as the mixture of the component hypotheses, namely,
Let Ht denote the mixture hypothesis of the learning procedure at round t. The procedure is to update its current combined hypothesis by the mixture of the previous combined hypothesis and a new hypothesis, i.e., by setting
H t(x,y)=(1−β)H t-1(x,y)+βh(x,y)
Thus, the expected cost of Ht on x is
C(x, H t(x))=(1−β)C(x, H t-1(x))+βC(x, h t(x))
Now, suppose that h predicts a particular label y for x, i.e., h(x,y)=1, then
C(x, H t(x))=(1−β)C(x, H t-1(x))+βC(x,y)
If we now take a derivative of this function with respect to β, we get
Note that this is the difference between the average cost of the current ensemble hypothesis and the new weak hypothesis assigning probability one to the specified label. - We then take this derivative with respect to all data points (x,y) in the expanded data set S′, and thus the gradient is mk-dimensional. We then expect the weak learner to find a hypothesis h whose inner-product with the negative gradient is large. That is, the output h of the weak learner seeks to maximize the following sum.
where W denotes the sum of absolute values of the weights, i.e.,
Note that unlike the weights typically used in existing hosting methods, the weights wx,y:=C(x, Ht-1(x))−C(x,y) can be negative, since y is not necessarily the best (least cost) label. This means that the weak learner now receives both positive and negative weights. While the minimization of weighted misclassification with positive and negative weights makes perfect sense as an optimization problem, its interpretation as a classification problem is not immediately clear. In particular, it prohibits the use of weighted sampling as a means of realizing the weighted classification problem. - We deal with this problem by converting a relational version of the weighted multi-class classification problem (i.e., of finding h to maximize Equation 9) in each iteration to a weighted binary classification problem. Specifically, we convert each example pair (x,y) to ((x,y), l), and set l=1 if the weight on (x,y) is positive, and l=0 if the weight is negative. The output hypothesis of the binary classifier is in general relational, so it is converted to a stochastic hypothesis by the procedure Stochastic. (The particular way this procedure is defined is motivated by the theoretical guarantee, which will be shown in the next subsection.) The overall process, consisting of multiple iterations of such a reduction, constitutes a reduction of the stochastic multi-class cost-sensitive classification to binary weighted classification.
- With the foregoing definitions, we can now state our main method, GBSE (Gradient Boosting with Stochastic Ensembles).
- It turns out that a strong theoretical performance guarantee can be proved on a variant of this method, which we describe below. We define the per label average cost, {tilde over (C)}(x, H(x)), of a stochastic hypothesis H, in general, as follows.
Note that, with this definition, the empirical loss (cost) of H on the original sample S, C(H, S), can be expressed as the sum of this per label cost over the expanded data set S′={(x,y)|x ε S, y ε Y}.
The variant, for which we prove our theoretical performance guarantee is obtained by simply replacing the weight updating rule of GBSE by the following:
w x,y ={tilde over (C)}(x, H t-1(x))−C(x,y)
The resulting variant, which we call GBSE-T (Gradient Boosting with Stochastic Ensembles—Theoretical version), is summarized inFIG. 5 . - We can show that GBSE-T has a boosting property given a version of weak learning condition on the component classifier. This weak learning condition, which we make precise below, is one that is sensitive to class imbalance.
-
DEFINITION 1. We say that an algorithm A for the binary importance weighted classification problem, as defined above, satisfies the weak learning condition for a given classification sample S=(x,y)m, if for arbitrary distribution over S, (w)m, Σw=1, when it is given S′=(x,y,w)m as input, its output h satisfies the following, for some fixed γ>0: -
THEOREM 2. Suppose that the component leaner A satisfies the weak learning condition for the input sample S. Then, the output of GBSE-T will converge to a stochastic ensemble hypothesis achieving minimum expected cost on the (original) sample S. In particular, if we set αt=α for all t,
Proof - We first establish the following simple correspondence between the weak learning conditions on the relational multi-class classification problem that we wish to solve in each iteration, and the binary classification problem that is given to the component algorithm to solve it.
-
DEFINITION 2. Let S be a weighted sample of the form S=(x,y,w)m, where weights w can be both positive and negative. Then define a transformed sample S′ from S by S′=((x,y), l, |w|)m where l=I(w≧0). -
- 1. The relational weighted multi-class classification problem for S is to find a relational hypothesis h:X×Y→{0, 1} that maximizes the following sum:
- 2. The weighted binary classification problem for S′ is to find a hypothesis h′:X×Y→{0, 1} that maximizes the following weighted classification accuracy:
- 1. The relational weighted multi-class classification problem for S is to find a relational hypothesis h:X×Y→{0, 1} that maximizes the following sum:
-
LEMMA 1. Assume the notation ofDefinition 2. Then, for arbitrary ε>0, h satisfies the following condition on the relational multi-class classification problem for S:
a(h,S)≧ε
if and only if (the same) h satisfies the corresponding condition on the transformed binary classification problem for S′:
Proof of Lemma 1
Hence the lemma follows.
Proof ofTheorem 2 - First, note that applying Stochastic to h, can increase the expected cost only for x's such that |[y|ht(x,y)=1}|=0, and for such x's the cost of f, equals that of H{t-1} by the definition of Stochastic. Hence, the empirical cost of f, on the original sample S, C(ft, S), satisfies the following:
Now recall that the expected empirical cost of Ht equals the following, where we drop the subscript t from αt.
Hence, by combining Equation 13 and Equation 14, we can show the following bound on the decrease in empirical cost in each iteration:
In the above derivation, the second to last inequality follows from the weak learning condition and applyingLemma 1 with weights {tilde over (C)}(x, Ht-1(x))−C(x,y). The last inequality follows from the fact that the weights are nonnalized so that the minimum achievable cost is zero for all x. Noting that the sum of these weights is positive whenever the current ensemble hypothesis is sub-optimal, this guarantees a positive progress in each iteration unless optimality is achieved. Since the expected empirical cost function as defined by ΣxΣy F(x,y) C(x,y) is convex (in fact linear), this implies convergence to the global optimum. Noting that in each iteration, the empirical cost is reduced at least by a factor of
and the theorem follows. - Note that at earlier iterations, the binary classifier used as the component learner is likely to be given weighted sample with balanced positive and negative examples. As the number of iterations increases and progress is made, however, it will receive samples that are increasingly more negative. (This is because the positive examples correspond to labels that can further improve the current performance.) It therefore becomes easier to attain high weighted accuracy by simply classifying all examples to be negative. The weak learning condition of Equation 12 appropriately deals with this issue, as it requires that the weak learner achieve better weighted accuracy than that attainable by assigning all examples to the negative class.
- In addition to the two variants of the Gradient Boosting with StochastiEnsembles method presented above, namely GBSE and GBSE-T, other related variations are possible. For example, in one variant, the weighted sampling can be done in two steps; the instance is sampled in the first step according to a probability proportional to
maxy×wx,y
and then choosing the label y with a probability proportional to
|wx,y|
In a yet another variant, the weighted sampling can be done in two steps; the instance is sampled in the first step according to the same probability as above, and for the chosen instance, examples are deterministically added for all possible labels. - Referring now to
FIG. 1 , there is shown a system on which a method for multi-class, cost-sensitive learning according to the invention may be implemented. This system comprises a multi-class cost-sensitive learningtop control module 1 which controls the overall control flow, making use of various sub-components of the system. A learningalgorithm storage module 2 stores a representation of an algorithm for classification learning. An arbitrary algorithm for classification can be used here. Alternatively the learning algorithm can be a decision tree learning algorithm, a naïve Bayes method, a logistic regression method or neural networks. Themodel output module 3 stores the models obtained as a result of applying the learning algorithm stored inmodule 2 to training data given byweighted sampling module 4 and outputs a final model by aggregating these models. Theweighted sampling module 4 accesses the data stored indata storage module 7, samples a relatively small subset of the data with acceptance probability determined using the example weights, and passes the obtained sub-sample tomodule 1. Theweight update module 5 updates the example weights for sampling using a particular function determined by the current weights and current models. Themodel update module 6 updates the current model using the model's output in the previous iterations stored in the currentmodel storage module 8 and the output model of the current iteration output bymodule 3 and stores the resulting updated model inmodule 8. -
FIG. 2 shows a flow diagram of the process implemented in the system ofFIG. 1 . The first three steps initialize the process. InStep 21, expanded data T is initialized using the input data S. InStep 22, H0 is initialized by setting for all (x,y) in T. Finally, inStep 23, the weights for all (x,y) in Tare initialized. The iteration begins in the decision block ofStep 4. A test is made to determine if i=t. If not, Step 25 performs the computation for all (x,y)
The decision block inStep 26 determines if there is more data in T or a STOP condition has been met. If not, inStep 27, (x,y) is sampled from T and accepted with a probability proportional to |w(x,y)|. Next, inStep 28, if accepted, ((x,y), |(w(x,y)>0)) is added to sub-sample Tt. A return is then made to the decision block inStep 26. When there is no more data in T or a STOP condition has been met, the process goes to Step 29 where the learning algorithm is run on Tt to obtain model ht. Next, inStep 30, ft is set equal to stoch(ht). Then, inStep 31, αt is chosen and Ht is set equal to (1−αt)Ht-1+αtft. The index i is incremented atStep 31, and a return is then made to the decision block inStep 24. If i=t, then inStep 33 the final model Ht is output. -
FIG. 3 shows a system on which another method for multi-class, cost-sensitive learning according to the invention may be implemented. This system is similar to that shown inFIG. 1 and comprises a multi-class cost-sensitive learningtop control module 1 which controls the overall control flow, making use of various sub-components of the system, a learningalgorithm storage module 2, which stores a representation of an algorithm for classification learning, amodel output module 3, which stores the models obtained as a result of applying the learning algorithm stored inmodule 2 to training data given byweighted sampling module 4 and outputs a final model by aggregating these models, and aweighted sampling module 4, which accesses the data stored indata storage module 7, samples a relatively small subset of the data with acceptance probability determined using the example weights, and passes the obtained sub-sample tomodule 1. Theweight calculation module 5′ replaces theweight upadate module 5, which updates the example weights for sampling using a dynamically changing weighting scheme. Themodel update module 6 updates the current model using the model's output in the previous iterations stored in the currentmodel storage module 8 and the output model of the current iteration output bymodule 3 and stores the resulting updated model inmodule 8. -
FIG. 4 shows a flow diagram of the process implemented in the system ofFIG. 3 . The first step initializes the process. InStep 41, expanded data T is initialized using the input data S. InStep 42, the weights for all (x,y) in Tare set. The iteration begins in the decision block ofStep 43. A test is made to determine if i=t. If not, a test is made inStep 44 to determine if there is no more data in T or a stop condition has been met. If not, Step 45 samples (x,y) from T and accepts (x,y) with probability proportional to w(x,y). If accepted, (x,y) is added to sub-sample T′ inStep 46. The process then loops back to decision block inStep 44 until there is either no more data in T or a stop condition has been met. At this point, the learning algorithm is run inStep 47 on T′ to obtain a model ht. InStep 48, α is chosen so that when i=1, α=0 and Ht=(1−αt)Ht-1+αtft. The index i is incremented atStep 49, and a return is then made to the decision block inStep 43. If i=t, then inStep 50 the final model Ht is output. - As a concrete example of applying the method of the invention to a real world problem, we describe an application to network intrusion detection. Network intrusion detection has recently become a proto-typical application problem for multi-class, cost-sensitive learning. The multi-class aspect is essential because in this application there are typically more than one level of intrusion detection, such as probing and denial of service. The cost-sensitive aspect is important because vastly different costs are associated with different types of misclassification (e.g., false negatives are usually a magnitude more costly than false positives) and it is absolutely critical that any learning method used to derive an intrusion detection rule is sensitive to this cost structure.
- A network intrusion detection system based on the method and system of the invention for multi-class, cost-sensitive learning consists of the following steps:
-
- 1) Convert past network connection data to a set of feature vectors, by mapping information on a network connection to a feature vector.
- 2) Label each of these vectors with known labels, such as “normal”, “probe”, “denial of service”, or specific types of intrusions.
- 3) Apply the method of the invention on the above data set, and obtain a classification rule.
- 4) Convert new network connection data to feature vectors, apply the above classification rule to them, and flag those connections corresponding to feature vectors that are classified as different types of “intrusions” as such.
- A typical set of features used to transform connection data into a well-defined feature vector is that used in the network intrusion data set known as “KDD CUP 99” data, which is publically available. Here is the list of features in this data set (given in three separate tables).
Basic Features of Individual TCP Connections feature name description type duration length (number of seconds) of the continuous connection protocol_type type of protocol, e.g., TCP, UDP, etc. discrete service network service on the destination, e.g., discrete http, telnet, etc. src_bytes number of data bytes from source to continuous desitination dst_bytes number of data bytes from destination to continuous source flag normal or error status of the connection discrete land 1 if connection is from/to the same discrete host/port; 0 otherwise wrong_fragment number of “wrong” fragments continuous urgent number of urgent packets continuous -
Content Features Within a Connection Suggested by Domain Knowledge feature name description type hot number of “hot” indicators continuous num_failed_logins number of failed login attempts continuous logged_in 1 if successfully logged in; 0 discrete otherwise num_compromised number of “compromised” conditions continuous root_shell 1 if root shell is obtained; 0 discrete otherwise su_attempted 1 if “su-root” command attempted; 0 discrete otherwise num_root number of “root” accesses continuous num_file_creations number of file creation operations continuous num_shells number of shell prompts continuous num_access_files number of operations on access continuous control files num_outbound_cmds number of outbound commands in an continuous ftp session is_hot_login 1 if the login belongs to the “hot” discrete list; 0 otherwise is_guest_login 1 if the login is a “guest” login; 0 discrete otherwise -
Traffic Features Computed Using a Two-Second Time Window feature name description type count number of connections to the same host continuous as the current connection in the past two seconds Note: The following features refer to these same host connections. serror_rate % of connections that have “SYN” continuous errors rerror_rate % of connections that have “REJ” continuous errors same_srv_rate % of connections of the same service continuous diff_srv_rate % of connections of different services continuous srv_count number of connections to the same continuous service as the current connection in the past two seconds Note: The following features refer to these same-service connections. srv_server_rate % of connections that have “SYN” continuous errors srv_rerror_rate % of connections that have “REJ” continuous errors srv_diff_host_rate % of connections to different hosts continuous - As a result of applying the multi-class, cost-sensitive learning method of the invention to a data set consisting of these features and the corresponding labels, using a decision tree algorithm as the “classification learning algorithm” stored in
Module 2 ofFIG. 1 , one obtains, as the classification rule, a voting function over a number of decision trees, such as the tree shown inFIG. 5 . - The system diagram of
FIG. 1 and the flow chart ofFIG. 2 illustrate a preferred embodiment of the invention, which corresponds to the method “GBSE” described herein. However, it will be understood by those skilled in the art that the method “DSE”, also described herein, may be used in the alternative. The main difference between DSE and GBSE is that in DSE, the sampling weights remain unchanged throughout all iterations. Consequently, the modules and funcationalities that are related to weight updating are unnecessary. - We use the C4.5 decision tree learner described by J. Quinlan in C4.5: Programs for Machine Learning, Morgan Kaufmann (1993), as the base classifier learning method, because it is a standard for empirical comparisons and it was used as the base learner by Domingos for the MetaCost method (see, P. Domingos, “MetaCost: A general method for making classifiers cost sensitive”, Proceedings of the Fifth International Conference on Knowledge Discovery and Data Mining, pp. 155-164, ACM Press, 1999).
- We compare our methods against three representative methods: Bagging (see L. Breiman, “Bagging predictors”, Machine Learning, 24(2):123-140, 1996), Averaging cost (see, P. Chan and S. Stolfo, “Toward scalable learning with non-uniform class and cost distributions”, Proceedings of the Fourth International Conference on Knowledge Discovery and Data Mining, pp. 164-168, 1998), and MetaCost (see, Domingos, ibid.). The Averaging cost method was also used for comparison in Domingos, ibid. Note that Bagging is a cost-insensitive learning method. Here we give a brief description of these methods, and refer the reader to Breiman, ibid., and Domingos, ibid., for the details.
-
- Bagging obtains multiple sub-samples by sampling with replacement, feeds them to the base learner (C4.5), and takes the average over the ensemble of output hypotheses.
- Averaging Cost (AvgCost) obtains a subsample by weighted sampling with weights defined as the average cost for each x, and then feeds it to the base learner (C4.5).
- MetaCost uses bagging to obtain an ensemble of hypotheses, uses the ensemble to estimate the class probabilities, and then outputs a hypothesis that minimizes the expected risk with respect to these estimates.
- There are some deviations from these methods in our implementation, which we clarify below. The main deviation is that we use rejection sampling for all methods, while other sampling schemes such as resampling with replacement are used in the original methods. We do this for two reasons: (1) inadequacy of resampling with replacement, especially for C4.5, has been noted by various authors (see, for example, B. Zadrozny, J. Langford, and N. Abe, “Cost-sensitive learning by cost-proportionate example weighting”, Proceedings of the Third IEEE International Conference on Data Mining, pp. 435-442, 2003); and (2) since our methods use rejection sampling, we do the same for the other methods for fairness of comparison. We stress that this deviation should only improve their performance. Another deviation is that we use a variant of MetaCost that skips the last step of learning a classifier on a relabeled training data set. It has been observed that this variant performs at least as well as MetaCost, in terms of cost minimization. (This variant has been called BagCost by D. Margineantu in Methods for Cost-Sensitive Learning, PhD thesis, Department of Computer Science, Oregon State University, Corvallis, Oreg., 2001.) Also, in our implementation of AvgCost, we perform weighted sampling multiple times to obtain an emsemble of hypotheses, then output their average as the final hypothesis. We note that, due to our normalization assumption that the minimum cost for each instance x is always zero, our version of AvgCost is identical to a more sophisticated variant in which the difference between the average cost and the minimum cost is used for sampling weights. Our experience shows that this variant of AvgCost performs better than the original method.
- The methods were applied to five benchmark datasets available from the UCI machine learning repository (C. L. Blake and C. J. Merz, “UCI repository of machine learning databases”, Department of Information and Computer Sciences, University of California, Irvine, Calif., 1998) and one dataset from the UCI KDD archive (S. D. Bay, “UCI archive”, Department of Information and Computer Sciences, University of California, 2000). These datasets were selected by the criteria of having approximately 1,000 data or more, besides being multiclass problems. A summary of these datasets is given in Table 1.
TABLE 1 Data set characteristics: data size, number of classes, and the ratio between the frequency of the most common class to the least common. Dataset # of examples # of classes Class ratio Annealing 898 5 0.01316 KDD-99 197710 5 0.0001278 Letter 20000 26 0.9028 Satellite 6435 6 0.4083 Solar flare 1389 7 0.002562 Splice 3190 3 0.4634
Except for the KDD-99 dataset, these datasets do not have standard misclassification costs associated with them. For this reason, we follow Domingos and generate cost matrices according to a model that gives higher costs for misclassifying a rare class as a frequent one, and inversely for lowest. (Note therefore that our experiments do not exploit the full generality of the instance-dependent cost formulation presented above.) This reflects a situation that is found in many practical data mining applications, including direct marketing and fraud detection, where the rare classes are the most valuable to identify correctly. - Our cost model is as follows: Let {circumflex over (P)}(y1) and {circumflex over (P)}(y2) be the empirical probabilities of occurrence of classes y1and y2 in the training data. We choose the non-diagonal entries of the cost matrix C(y1, y2), y1≠y2 with uniform probability from the interval [0,2000 {circumflex over (P)}(y1)/{circumflex over (P)}(y2)]. In Domingos, ibid., the diagonal entries were then chosen from the interval [0,1000], which often leads to cost matrices in which the correct label is not the least costly one. Besides being unreasonable (see C. Elkan, “Magical thinking in data mining: Lessons from coil challenge 2000”, Proceedings of the Seventh International Conference on Knowledge Discovery and Data Mining, pp. 426-43 1, ACM Press, 1999), these cost matrices can give an unfair advantage to cost-sensitive methods over cost-insensitive ones. We therefore set the diagonal entries to be identically zero, which is consistent with our normalization assumption.
- In all experiments, we randomly select ⅔ of the examples in the dataset for training and use the remaining ⅓ for testing. Also, for each training/test split we generate a different cost matrix according to the rules above. Thus, the standard deviations that we report reflect both variations in the data and in the misclassification costs.
- We remark on certain implementation details of the proposed learning methods in our experimentation. First, we note that in all of the methods used for comparison, C4.5 was used as the component algorithm, and the final hypothesis is expressed as an ensemble of output decision tress of C4.5. Its output hypothesis is therefore also an ensemble of decision trees. Next, the choice of the mixture weight αt was unspecified in the algorithm descriptions. The choice of αt was set at 1/t for most methods.
- The results of these experiments are summarized in Tables 2 and 3.
TABLE 2 Experimental results: the average cost and standard error. Dataset Bagging AvgCost MetaCost DSE GBSE Annealing 1059 ± 174 127.4 ± 12.2 206.8 ± 42.8 127.1 ± 14.9 33.72 ± 4.29 Solar 5403 ± 397 237.8 ± 37.5 5317 ± 390 110.9 ± 28.7 48.17 ± 9.52 KDD-99 319.4 ± 42.2 42.43 ± 7.95 49.39 ± 9.34 46.68 ± 10.16 1.69 ± 0.78 letter 151.0 ± 2.58 91.90 ± 1.36 129.6 ± 2.44 114.0 ± 1.43 84.63 ± 2.24 Splice 64.19 ± 5.25 60.78 ± 3.65 49.95 ± 3.05 135.5 ± 14 57.50 ± 4.38 Satellite 189.9 ± 9.57 107.8 ± 5.95 104.4 ± 6.43 116.8 ± 6.28 93.05 ± 5.57 -
TABLE 3 Experimental results: the average data size used by each method in 30 iterations, and standard error. Dataset Bagging AvgCost MetaCost DSE GBSE Annealing 11991 ± 13.1 1002.8 ± 183 11987 ± 9.84 3795.5 ± 688 1260.2 ± 224 Solar 18499 ± 20.4 334.80 ± 37.5 18510 ± 14.4 2112.8 ± 276 486.45 ± 53.3 KDD-99 395310 ± 143 2551.9 ± 428.6 395580 ± 143 12512 ± 2450 4181 ± 783.6 letter 40037 ± 44.3 159720 ± 2028 40052 ± 41 479130 ± 2710 363001 ± 5557 Splice 42515 ± 26.6 33658 ± 1697 42501 ± 21 52123 ± 592 50284 ± 3659 Satellite 86136 ± 123 60876 ± 1641 85984 ± 127 218870 ± 6516 140810 ± 3335 - Table 2 lists the average costs attained by each of these methods on the 6 data sets, and their stand errors. These results were obtained by averaging over 20 runs, each run consisting of 30 iterations of the respective learning method. These results appear quite convincing: GBSE out-performs all comparison methods on all data sets, except on Splice, for which it ranks second after MetaCost. Also, GBSE is the best performing among the proposed methods, confirming our claim that the combination of various techniques involved is indeed necessary to attain this level of performance.
- Table 3 lists the average total data size used by each of the methods in 30 iterations. Examining these results in conjunction with the data characteristics in Table 1 reveals a definite trend. First, note that the data sets are divided into to groups: those having very large skews, or very low class ratios (Annealing, KDD-99 and Solar flare), and those having moderate skews (Satellite, Splice and Letter). It is evident that the methods based on example weighting (AvgCost, GBSE, DSE) use magnitudes smaller data sizes for the three data sets in the first group (i.e., with large skews), as compared to other methods, Bagging and MetaCost. The performance of GBSE is especially impressive on this group, achieving much lower costs while requiring very small data sizes. It is worth mentioning that it is these data sets in the first group with large skews that require cost-sensitive learning the most.
- We have provided a novel method for multiclass cost-sensitive learning based on gradient boosting with stochastic ensembles. It is not the first time that the issue of incorporating cost-sensitivity to boosting has been addressed. For example, AdaCost (see W. Fan, S. J. Stolfo, J. Zhang, and P. K. Chan, “AdaCost: Misclassification cost-sensitive boosting”, Proceedings of the Sixteenth International Conference on Machine Learning, pp. 97-105, 1999) suggested a way of modifying AdaBoost's exponential loss using a function (called cost adjustment function) of the cost and confidence. The rational choice of this cost adjustment function, however, appears not to be well-understood. The stochastic ensemble that we employ in this method provides a straightforward but reasonable way of incorporating cost and confidence; i.e., in terms of expected cost.
- While the invention has been described in terms of a single preferred embodiment, those skilled in the art will recognize that the invention can be practiced with modification within the spirit and scope of the appended claims.
Claims (20)
1. A method for multi-class, cost-sensitive learning based on iterative example weighting schemes applied to a chosen data set comprising the steps of:
a) obtaining an expanded data set, which is defined by enhancing each example in an original data set with as many data points as there are possible labels for any single instance;
b) repeatedly drawing sub-samples from the expanded data set using weighted sampling according to a certain example weighting scheme that remains constant throughout the iterations, in which each labeled example is given the weight specified as the difference between the maximum possible misclassification cost for the instance in question and the misclassification associated with the label in the particular labeled example;
c) calling a component classification learning algorithm to the sub-sample obtained in step b) and obtaining a hypothesis representing a classifier;
d) outputting all classifier representations obtained through the iterations and representing an average over them, each of which can be an arbitrary representation of classifier for a problem at hand; and
e) outputting all of the representations obtained through the iterations representing an average over them, each of which can be an arbitrary representation of classifier for the problem at hand.
2. The method for multi-class, cost-sensitive learning recited in claim 1 , wherein the learning algorithm is an arbitrary algorithm for classification.
3. The method for multi-class, cost-sensitive learning recited in claim 1 , wherein the learning algorithm is selected from the group consisting of decision tree algorithms, naïve Bayes method, logistic regression method and neural networks.
4. A method for multi-class, cost-sensitive learning based on an example weighting scheme applied to a chosen data set comprising the steps of:
a) obtaining an expanded data set, which is defined by enhancing each example in an original data set with as many data points as there are possibles for any single instance;
b) iteratively applying weighted sampling from the expanded data set, using a dynamically changing weighting scheme involving both positive and negative weights;
c) calling a component classification algorithm on a modified binary classification problem in which each example is itself already a labeled pair, and its (meta) label is 1 or 0 depending on whether the example weight in the above weighting scheme is positive or negative, respectively, and obtains a hypothesis representing a classifier;
d) optionally modifying the obtained classifier, which is in general a relation on the original classification (mapping elements of domain to the labels, so that it is stochastic, namely a conditional probability distribution so that its probabilities over the set of labels sum to one for each instance;
e) outputting all representations obtained through the iterations and representing an average over them, each of which can be an arbitrary representation of classifier for the problem at hand.
5. The method for multi-class, cost-sensitive learning recited in claim 4 , wherein the learning algorithm is an arbitrary algorithm for classification.
6. The method for multi-class, cost-sensitive learning recited in claim 4 , wherein the learning algorithm is selected from the group consisting of decision tree algorithms, naïve Bayes method, logistic regression method and neural networks.
7. The method for multi-class, cost-sensitive learning recited in claim 4 , wherein the dynamically changing weighting of step b) gives each labeled example a weight specified as a difference between an average cost on that instance by the averaged hypotheses from iterations so far and a misclassification cost associated with the label in the labeled example in question
8. The method for multi-class, cost-sensitive learning recited in claim 4 , wherein the dynamically changing weighting of step b) gives each labeled example a weight specified as a difference between an average cost on that instance by an averaged hypotheses from iterations so far divided by a number of labels per instance, and a misclassification cost associated with the label in the labeled example in question.
9. The method for multi-class, cost-sensitive learning recited in claim 4 , wherein the dynamically changing weighting of step b) gives each labeled example a weight specified as a difference between an average cost on that instance by an averaged hypotheses from iterations so far divided by a number of labels per instance, and a misclassification cost associated with the label in the labeled example in question, and the weighted sampling comprises the steps of:
sampling the instance in step a) according to a probability proportional to a maximum of weights for that instance and any of the labels, and
choosing a label with a probability proportional to the absolute value of a weight for that instance and the label in question.
10. The method for multi-class, cost-sensitive learning recited in claim 4 , wherein the dynamically changing weighting of step b) gives each labeled example a weight specified as a difference between an average cost on that instance by an averaged hypotheses from iterations so far divided by a number of labels per instance, and a misclassification cost associated with the label in the labeled example in question, and the weighted sampling comprises the steps of:
sampling the instance step a) according to a probability proportional to a maximum of weights for that instance and any of the labels, and
for the chosen instance, deterministically added examples for all possible labels.
11. A system implementing a method for multi-class, cost-sensitive learning based on iterative example weighting schemes applied to a chose data set comprising:
a multi-class cost-sensitive learning top control module controlling the overall control flow;
a learning algorithm storage module storing a representation of a learning algorithm for classification learning;
a model output module storing models obtained as a result of applying the learning algorithm to training data given by a weighted sampling module and outputting a final model by aggregating these models, said weighted sampling module accessing data stored in a data storage module, sampling a relatively small subset of the data with acceptance probability determined using the example weights, and passing the obtained sub-sample to said top control module;
a weight calculation module updating the example weights for sampling using weighted sampling according to a weighting scheme that remains constant throughout iterations, in which each labeled example is given a weight specified as a difference between a maximum possible misclassification cost for the instance in question and a misclassification cost associated with the label in the particular labeled example; and
a model update module updating current models using a model's output in previous iterations stored in a current model storage module and an output model of a current iteration output and storing a resulting updated model in said current storage module.
12. The system for implementing a method for multi-class, cost-sensitive learning recited in claim 11 , wherein the learning algorithm is an arbitrary algorithm for classification.
13. The system for implementing a method for multi-class, cost-sensitive learning recited in claim 11 , wherein the learning algorithm is selected from the group consisting of decision tree algorithms, naïve Bayes method, logistic regression method and neural networks.
14. A system implementing a method for multi-class, cost-sensitive learning based on an example weighting scheme applied to a chosen data set comprising:
a multi-class cost-sensitive learning top control module controlling the overall control flow;
a learning algorithm storage module storing a representation of a learning algorithm for classification learning;
a model output module storing models obtained as a result of applying the learning algorithm to training data given by a weighted sampling module and outputting a final model by aggregating these models, said weighted sampling module accessing data stored in a data storage module, sampling a relatively small subset of the data with acceptance probability determined using the example weights, and passing the obtained sub-sample to said top control module;
a weight calculation module calculating the example weights for sampling using a dynamically changing weighting scheme involving both positive and negative weights; and
a model update module updating the current models using the models output in the previous iterations stored in a current model storage module and the output model of the current iteration output and storing the resulting updated model in said current storage module.
15. The system for implementing a method for multi-class, cost-sensitive learning recited in claim 14 , wherein the learning algorithm is an arbitrary algorithm for classification.
16. The system for implementing a method for multi-class, cost-sensitive learning recited in claim 14 , wherein the learning algorithm is selected from the group consisting of decision tree algorithms, naïve Bayes method, logistic regression method and neural networks.
17. The system implementing a method for multi-class, cost-sensitive learning recited in claim 14 , wherein the dynamically changing weighting scheme used by the weight calculation module gives each labeled example a weight specified as a difference between an average cost on that instance by the averaged hypotheses from iterations so far and a misclassification cost associated with the label in the labeled example in question
18. The system implementing a method for multi-class, cost-sensitive learning recited in claim 14 , wherein the dynamically changing weighting scheme used by the weight calculation module gives each labeled example a weight specified as a difference between an average cost on that instance by an averaged hypotheses from iterations so far divided by a number of labels per instance, and a misclassification cost associated with the label in the labeled example in question.
19. The system implementing a method for multi-class, cost-sensitive learning recited in claim 14 , wherein the dynamically changing weighting scheme used by the weight calculation module gives each labeled example a weight specified as a difference between an average cost on that instance by an averaged hypotheses from iterations so far divided by a number of labels per instance, and a misclassification cost associated with the label in the labeled example in question, and the weighted sampling samples the instance according to a probability proportional to a maximum of weights for that instance and any of the labels, and a label is chosen with a probability proportional to the absolute value of a weight for that instance and the label in question.
20. The system implementing a method for multi-class, cost-sensitive learning recited in claim 14 , wherein the dynamically changing weighting scheme used by the weight calculation module gives each labeled example a weight specified as a difference between an average cost on that instance by an averaged hypotheses from iterations so far divided by a number of labels per instance, and a misclassification cost associated with the label in the labeled example in question, and the weighted sampling samples the instance according to a probability proportional to a maximum of weights for that instance and any of the labels, and for the chosen instance, deterministically adds examples for all possible labels.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/876,533 US20050289089A1 (en) | 2004-06-28 | 2004-06-28 | Methods for multi-class cost-sensitive learning |
US11/937,629 US7558764B2 (en) | 2004-06-28 | 2007-11-09 | Methods for multi-class cost-sensitive learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/876,533 US20050289089A1 (en) | 2004-06-28 | 2004-06-28 | Methods for multi-class cost-sensitive learning |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/937,629 Division US7558764B2 (en) | 2004-06-28 | 2007-11-09 | Methods for multi-class cost-sensitive learning |
Publications (1)
Publication Number | Publication Date |
---|---|
US20050289089A1 true US20050289089A1 (en) | 2005-12-29 |
Family
ID=35507280
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/876,533 Abandoned US20050289089A1 (en) | 2004-06-28 | 2004-06-28 | Methods for multi-class cost-sensitive learning |
US11/937,629 Active US7558764B2 (en) | 2004-06-28 | 2007-11-09 | Methods for multi-class cost-sensitive learning |
Family Applications After (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/937,629 Active US7558764B2 (en) | 2004-06-28 | 2007-11-09 | Methods for multi-class cost-sensitive learning |
Country Status (1)
Country | Link |
---|---|
US (2) | US20050289089A1 (en) |
Cited By (26)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070159481A1 (en) * | 2006-01-11 | 2007-07-12 | Naoki Abe | Method and apparatus for presenting feature importance in predictive modeling |
US20090228499A1 (en) * | 2008-03-05 | 2009-09-10 | Schmidtler Mauritius A R | Systems and methods for organizing data sets |
US20100042561A1 (en) * | 2008-08-12 | 2010-02-18 | International Business Machines Corporation | Methods and systems for cost-sensitive boosting |
US7689873B1 (en) * | 2005-09-19 | 2010-03-30 | Google Inc. | Systems and methods for prioritizing error notification |
CN103049759A (en) * | 2012-12-14 | 2013-04-17 | 上海邮政科学研究院 | Postal code recognition method for postal sorting system |
CN105305826A (en) * | 2014-07-28 | 2016-02-03 | 苏州弗莱希智能科技有限公司 | Multi-output switching power supply circuit |
US20160070986A1 (en) * | 2014-09-04 | 2016-03-10 | Xerox Corporation | Domain adaptation for image classification with class priors |
CN105260746B (en) * | 2015-10-09 | 2019-02-22 | 济南大学 | A kind of integrated Multi-label learning system of expansible multilayer |
CN109446393A (en) * | 2018-09-12 | 2019-03-08 | 北京邮电大学 | A kind of Web Community's topic classification method and device |
CN109831774A (en) * | 2019-01-08 | 2019-05-31 | 中国联合网络通信集团有限公司 | A kind of big data expands quadrat method and device |
CN110059756A (en) * | 2019-04-23 | 2019-07-26 | 东华大学 | A kind of multi-tag categorizing system based on multiple-objection optimization |
US10572877B2 (en) * | 2014-10-14 | 2020-02-25 | Jpmorgan Chase Bank, N.A. | Identifying potentially risky transactions |
US10614361B2 (en) * | 2015-09-09 | 2020-04-07 | Intel Corporation | Cost-sensitive classification with deep learning using cost-aware pre-training |
CN111652268A (en) * | 2020-04-22 | 2020-09-11 | 浙江盈狐云数据科技有限公司 | Unbalanced stream data classification method based on resampling mechanism |
CN111860638A (en) * | 2020-07-17 | 2020-10-30 | 湖南大学 | Parallel intrusion detection method and system based on unbalanced data deep belief network |
CN111967343A (en) * | 2020-07-27 | 2020-11-20 | 广东工业大学 | Detection method based on simple neural network and extreme gradient lifting model fusion |
EP3701351A4 (en) * | 2018-01-30 | 2021-01-27 | D5Ai Llc | Self-organizing partially ordered networks |
CN112733913A (en) * | 2020-12-31 | 2021-04-30 | 浙江禾连网络科技有限公司 | Child and old person cooperative property safety detection method based on cost Adaboost algorithm |
US11042891B2 (en) | 2017-06-05 | 2021-06-22 | International Business Machines Corporation | Optimizing revenue savings for actionable predictions of revenue change |
US11055730B2 (en) * | 2017-06-05 | 2021-07-06 | International Business Machines Corporation | Optimizing predictive precision for actionable forecasts of revenue change |
US11062334B2 (en) * | 2017-06-05 | 2021-07-13 | International Business Machines Corporation | Predicting ledger revenue change behavior of clients receiving services |
US11093830B2 (en) | 2018-01-30 | 2021-08-17 | D5Ai Llc | Stacking multiple nodal networks |
US20210304039A1 (en) * | 2020-03-24 | 2021-09-30 | Hitachi, Ltd. | Method for calculating the importance of features in iterative multi-label models to improve explainability |
CN113746707A (en) * | 2021-11-08 | 2021-12-03 | 南京信息工程大学 | Encrypted traffic classification method based on classifier and network structure |
CN114638336A (en) * | 2021-12-26 | 2022-06-17 | 海南大学 | Unbalanced learning focusing on strange samples |
US11636379B2 (en) * | 2016-03-26 | 2023-04-25 | Alibaba Group Holding Limited | Distributed cluster training method and apparatus |
Families Citing this family (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080275775A1 (en) * | 2007-05-04 | 2008-11-06 | Yahoo! Inc. | System and method for using sampling for scheduling advertisements in an online auction |
AU2010223925A1 (en) * | 2009-03-13 | 2011-11-03 | Rutgers, The State University Of New Jersey | Systems and methods for the detection of malware |
US8572011B1 (en) * | 2010-07-16 | 2013-10-29 | Google Inc. | Outcome estimation models trained using regression and ranking techniques |
US8738549B2 (en) | 2010-12-21 | 2014-05-27 | International Business Machines Corporation | Predictive modeling |
DE102011076780B4 (en) | 2011-05-31 | 2021-12-09 | Airbus Operations Gmbh | Method and device for condition monitoring, computer program product |
US8825570B2 (en) * | 2012-07-31 | 2014-09-02 | Hewlett-Packard Development Company, L.P. | Active learning with per-case symmetrical importance scores |
US9224104B2 (en) | 2013-09-24 | 2015-12-29 | International Business Machines Corporation | Generating data from imbalanced training data sets |
US9875301B2 (en) | 2014-04-30 | 2018-01-23 | Microsoft Technology Licensing, Llc | Learning multimedia semantics from large-scale unstructured data |
US10013637B2 (en) | 2015-01-22 | 2018-07-03 | Microsoft Technology Licensing, Llc | Optimizing multi-class image classification using patch features |
US9785866B2 (en) | 2015-01-22 | 2017-10-10 | Microsoft Technology Licensing, Llc | Optimizing multi-class multimedia data classification using negative data |
CN105373606A (en) * | 2015-11-11 | 2016-03-02 | 重庆邮电大学 | Unbalanced data sampling method in improved C4.5 decision tree algorithm |
CN106611181A (en) * | 2016-05-30 | 2017-05-03 | 四川用联信息技术有限公司 | Method for constructing cost-sensitive two-dimensional decision tree |
CN106385693B (en) * | 2016-09-22 | 2020-02-07 | 长沙创客软件有限公司 | Telecommunication fraud judgment method aiming at virtual number segment |
CN108073986B (en) * | 2016-11-16 | 2020-05-12 | 北京搜狗科技发展有限公司 | Neural network model training method and device and electronic equipment |
US11321612B2 (en) | 2018-01-30 | 2022-05-03 | D5Ai Llc | Self-organizing partially ordered networks and soft-tying learned parameters, such as connection weights |
CN110072017A (en) * | 2019-04-28 | 2019-07-30 | 济南大学 | Abnormal phone recognition methods and system based on feature selecting and integrated study |
CN110598243B (en) * | 2019-07-26 | 2021-04-30 | 浙江大学 | Virtual sample capacity expansion method based on historical data of mechanical product |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3483512A (en) * | 1965-11-30 | 1969-12-09 | Gen Dynamics Corp | Pattern recognition system |
US6678548B1 (en) * | 2000-10-20 | 2004-01-13 | The Trustees Of The University Of Pennsylvania | Unified probabilistic framework for predicting and detecting seizure onsets in the brain and multitherapeutic device |
-
2004
- 2004-06-28 US US10/876,533 patent/US20050289089A1/en not_active Abandoned
-
2007
- 2007-11-09 US US11/937,629 patent/US7558764B2/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3483512A (en) * | 1965-11-30 | 1969-12-09 | Gen Dynamics Corp | Pattern recognition system |
US6678548B1 (en) * | 2000-10-20 | 2004-01-13 | The Trustees Of The University Of Pennsylvania | Unified probabilistic framework for predicting and detecting seizure onsets in the brain and multitherapeutic device |
Cited By (33)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7689873B1 (en) * | 2005-09-19 | 2010-03-30 | Google Inc. | Systems and methods for prioritizing error notification |
US7561158B2 (en) * | 2006-01-11 | 2009-07-14 | International Business Machines Corporation | Method and apparatus for presenting feature importance in predictive modeling |
US20070159481A1 (en) * | 2006-01-11 | 2007-07-12 | Naoki Abe | Method and apparatus for presenting feature importance in predictive modeling |
US9082080B2 (en) | 2008-03-05 | 2015-07-14 | Kofax, Inc. | Systems and methods for organizing data sets |
US20090228499A1 (en) * | 2008-03-05 | 2009-09-10 | Schmidtler Mauritius A R | Systems and methods for organizing data sets |
US20100262571A1 (en) * | 2008-03-05 | 2010-10-14 | Schmidtler Mauritius A R | Systems and methods for organizing data sets |
US8321477B2 (en) | 2008-03-05 | 2012-11-27 | Kofax, Inc. | Systems and methods for organizing data sets |
US20100042561A1 (en) * | 2008-08-12 | 2010-02-18 | International Business Machines Corporation | Methods and systems for cost-sensitive boosting |
US8275721B2 (en) * | 2008-08-12 | 2012-09-25 | International Business Machines Corporation | Methods and systems for cost-sensitive boosting |
CN103049759A (en) * | 2012-12-14 | 2013-04-17 | 上海邮政科学研究院 | Postal code recognition method for postal sorting system |
CN105305826A (en) * | 2014-07-28 | 2016-02-03 | 苏州弗莱希智能科技有限公司 | Multi-output switching power supply circuit |
US20160070986A1 (en) * | 2014-09-04 | 2016-03-10 | Xerox Corporation | Domain adaptation for image classification with class priors |
US9710729B2 (en) * | 2014-09-04 | 2017-07-18 | Xerox Corporation | Domain adaptation for image classification with class priors |
US10572877B2 (en) * | 2014-10-14 | 2020-02-25 | Jpmorgan Chase Bank, N.A. | Identifying potentially risky transactions |
US10614361B2 (en) * | 2015-09-09 | 2020-04-07 | Intel Corporation | Cost-sensitive classification with deep learning using cost-aware pre-training |
CN105260746B (en) * | 2015-10-09 | 2019-02-22 | 济南大学 | A kind of integrated Multi-label learning system of expansible multilayer |
US11636379B2 (en) * | 2016-03-26 | 2023-04-25 | Alibaba Group Holding Limited | Distributed cluster training method and apparatus |
US11042891B2 (en) | 2017-06-05 | 2021-06-22 | International Business Machines Corporation | Optimizing revenue savings for actionable predictions of revenue change |
US11055730B2 (en) * | 2017-06-05 | 2021-07-06 | International Business Machines Corporation | Optimizing predictive precision for actionable forecasts of revenue change |
US11062334B2 (en) * | 2017-06-05 | 2021-07-13 | International Business Machines Corporation | Predicting ledger revenue change behavior of clients receiving services |
US11461655B2 (en) | 2018-01-30 | 2022-10-04 | D5Ai Llc | Self-organizing partially ordered networks |
US11093830B2 (en) | 2018-01-30 | 2021-08-17 | D5Ai Llc | Stacking multiple nodal networks |
EP3701351A4 (en) * | 2018-01-30 | 2021-01-27 | D5Ai Llc | Self-organizing partially ordered networks |
CN109446393A (en) * | 2018-09-12 | 2019-03-08 | 北京邮电大学 | A kind of Web Community's topic classification method and device |
CN109831774A (en) * | 2019-01-08 | 2019-05-31 | 中国联合网络通信集团有限公司 | A kind of big data expands quadrat method and device |
CN110059756A (en) * | 2019-04-23 | 2019-07-26 | 东华大学 | A kind of multi-tag categorizing system based on multiple-objection optimization |
US20210304039A1 (en) * | 2020-03-24 | 2021-09-30 | Hitachi, Ltd. | Method for calculating the importance of features in iterative multi-label models to improve explainability |
CN111652268A (en) * | 2020-04-22 | 2020-09-11 | 浙江盈狐云数据科技有限公司 | Unbalanced stream data classification method based on resampling mechanism |
CN111860638A (en) * | 2020-07-17 | 2020-10-30 | 湖南大学 | Parallel intrusion detection method and system based on unbalanced data deep belief network |
CN111967343A (en) * | 2020-07-27 | 2020-11-20 | 广东工业大学 | Detection method based on simple neural network and extreme gradient lifting model fusion |
CN112733913A (en) * | 2020-12-31 | 2021-04-30 | 浙江禾连网络科技有限公司 | Child and old person cooperative property safety detection method based on cost Adaboost algorithm |
CN113746707A (en) * | 2021-11-08 | 2021-12-03 | 南京信息工程大学 | Encrypted traffic classification method based on classifier and network structure |
CN114638336A (en) * | 2021-12-26 | 2022-06-17 | 海南大学 | Unbalanced learning focusing on strange samples |
Also Published As
Publication number | Publication date |
---|---|
US7558764B2 (en) | 2009-07-07 |
US20080065572A1 (en) | 2008-03-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7558764B2 (en) | Methods for multi-class cost-sensitive learning | |
Abe et al. | An iterative method for multi-class cost-sensitive learning | |
US8719197B2 (en) | Data classification using machine learning techniques | |
US8239335B2 (en) | Data classification using machine learning techniques | |
US7761391B2 (en) | Methods and systems for improved transductive maximum entropy discrimination classification | |
Zhu et al. | A second-order approach to learning with instance-dependent label noise | |
Lei et al. | GCN-GAN: A non-linear temporal link prediction model for weighted dynamic networks | |
US20080086432A1 (en) | Data classification methods using machine learning techniques | |
US11836615B2 (en) | Bayesian nonparametric learning of neural networks | |
US7493346B2 (en) | System and method for load shedding in data mining and knowledge discovery from stream data | |
US20080071721A1 (en) | System and method for learning models from scarce and skewed training data | |
US8275721B2 (en) | Methods and systems for cost-sensitive boosting | |
EP3754549A1 (en) | A computer vision method for recognizing an object category in a digital image | |
WO2008008142A2 (en) | Machine learning techniques and transductive data classification | |
Han et al. | Local uncertainty sampling for large-scale multiclass logistic regression | |
US11928583B2 (en) | Adaptation of deep learning models to resource constrained edge devices | |
WO2020024444A1 (en) | Group performance grade recognition method and apparatus, and storage medium and computer device | |
Lozano et al. | Multi-class cost-sensitive boosting with p-norm loss functions | |
US20100161534A1 (en) | Predictive gaussian process classification with reduced complexity | |
Valizadegan et al. | Learning to trade off between exploration and exploitation in multiclass bandit prediction | |
US20210192032A1 (en) | Dual-factor identification system and method with adaptive enrollment | |
US20040254930A1 (en) | Construction and selection of a finite mixture model for use in clustering and vector quantization | |
WO2020024448A1 (en) | Group performance grade identification method, device, storage medium, and computer apparatus | |
US20210326760A1 (en) | Learning device, learning method, and prediction system | |
Hasan et al. | Calibrated one round federated learning with bayesian inference in the predictive space |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: INTERNATIONAL BUSINESS MACHINES CORPORATION, NEW Y Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:ABE, NAOKI;ZADROZNY, BIANCA;REEL/FRAME:015723/0455 Effective date: 20040628 |
|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION |