DE10313420A1 - Research system and method for determining information from a database, in particular from the World Wide Web - Google Patents

Abstract

A research system and a method for determining information from a database, in particular from the World Wide Web, are said to be particularly suitable for achieving consistently high search qualities. For this purpose, according to the invention, the joint and / or simultaneous use of a plurality of search engines (2) is provided, with an integration module (6) which contains the result data comprising a plurality of result data records (E) delivered by the search engines (2) into a common one Converted data format, using a prioritization parameter (P) stored in a memory module (14), the assignment of a relevance parameter to each result data record (E) is provided.

Description

The Invention relates to a search system for determining Information from a database using a plurality of Search engines. It also relates to a method of investigation information from a database, especially from the World Wide Web, using such a research system.

In complex databases or even in the global computer network ("World Wide Web") will be enormous Amount of information held by a user for research purposes more or less targeted. To get information from the big To be able to use the amount of information in a targeted manner so-called search engines are used, which are particularly useful for gathering information experience considerable spread from the World Wide Web to have. In such search engines, the user is usually informed about a An input / output module provides an inquiry window via which targeted search or research terms can be specified. Then searches the Search engine the information base of the database or the World Wide Web for suitable key or keywords. The response data records found thereupon are usually from the respective Search engine with regard to its relevance for the given search order categorized and the user in the manner of a hit list in order according to their relevance.

In the meantime, a large number of such search engines have established themselves themselves regarding their performance, reliability and / or the considered Information base in the manner of a thematic specialization can distinguish from each other. Precisely because of the large number of, for example, on the World Wide Web offered information when operating such a search engine certain focus or specialization of individual search engines The quality of the search results may depend on the given search query for individual search engines vary in certain ranges.

The The invention is therefore based on the object of a research system to determine information from a database, in particular from the world wide web, Specify that to achieve a consistently high quality of research in particular Dimensions for shared and / or simultaneous use of a plurality of search engines is particularly suitable. Furthermore, a procedure for the determination is to be carried out information from a database, especially from the World Wide Web, be specified using such a research system.

Regarding the Research system this problem is solved according to the invention with a Integration module, which is supplied by the search engines, respectively a plurality of result records comprehensive results data converted into a common data format and subsequently under recourse to a prioritization parameter stored in the memory module assigns a relevance parameter to each result data set.

The invention is based on the consideration that, in order to achieve consistently high research qualities, a plurality of alternative search engines should in principle be used, so that quality defects dependent on search questions can be compensated for alternately. However, especially when using a plurality of search engines for a search query, it can be expected that a large number of search hits or result data sets will be generated by the whole of the search engines. For a user-friendly and targeted evaluation of these results, a relevance identifier common to all results should be assigned, so that it is understandable for the user that all search results can be systematically arranged according to predetermined relevance criteria. In particular, it should be possible to present the results that are considered the most relevant according to the specified criteria first, so that the user can concentrate on the essential research results. In order to make this possible, the conversion of the incoming data records into a common data format is initially provided, so that a common data evaluation and structuring is made possible. The structure of the result data records found is then provided in the sense of a common relevance structure. This function of merging and consolidating result data sentences from different sources are fulfilled by the integration module.

The hierarchical categorization of the found data records Assignment of a specific relevance characteristic can be done from User or an operator predetermined prioritization criteria respectively. In particular, could intended to be determined when evaluating the results delivered Search engines in general and standardized an increased relevance assign, for example because one for a certain type Search query is known to certain search engines usually deliver comparatively high research quality. In order to make this possible, the prioritization parameters are advantageously dependent on chosen by the respective search engine, so, for example, based on that for a particular search engine stored priority value can be determined that a result data set delivered by this search engine with a comparatively high relevance value.

The Specification of a specific prioritization when considering the search engine can be specified by the user according to type and scope his. Within the result data sets supplied by a search engine the so - called internal prioritization based on the Prioritization already delivered by the search engine. This is done in a further or alternative advantageous embodiment the relevance index for the respective result data record taking into account one of the respective search engine provided output relevance parameter determined by the integration module.

there can normalization in particular in the manner of a "dynamic prioritization" output relevance parameters already supplied by the respective search engines be provided in which the search engine as Most relevant viewed result data record with a preliminary relevance characteristic is provided by, for example, 100%, with that of the respective Search engine already classified as comparatively less relevant further result data sets be provided with a correspondingly graduated relevance characteristic. With this type of evaluation and prioritization, Art equal rights for all used search engines a direct cross-comparison of individual Result records allows. additionally or alternatively it can also be provided that already at normalization, that is to say the formation of the reference variable for the of the respective search engine considered the most relevant result data set, a search engine-specific characteristic is specified by which the relevance characteristic of the most relevant from the respective search engine viewed result data set specified for the search engine becomes. Such a search engine-specific specification thus relative weightings between individual search engines possible in which in particular also experience based on general reliability of the respective search engine and / or also subject-specific or search query-dependent knowledge about the respective search engine can be included.

The Integration module is for the joint evaluation of the different Search engine delivered result records designed, initially one Conversion of the individual result data sets into a common one thus jointly evaluable data format is provided. Is to the integration module for it designed, the specific, used by the respective search engine Evaluate syntax or its data format and convert if necessary to be able to. In order to make this possible, is the integration module advantageously with a second memory module connected in which syntax data for the individual search engines are stored. Using this syntax data the search engine-specific data conversion can then be carried out become.

In Another advantageous embodiment is in addition to the integration module also a distribution module is provided, which one in common Data format incoming search order suitable for the individual Split intended use of search engines. This is the distribution module, the expediently via a common buffer memory module communicates with the integration module, advantageously for that designed, the respective search order suitable in a for the respective Search engine processable syntax or a corresponding data format convert.

In order to enable a reliable assignment of the delivered results to the respective search order even if a comparatively large number of search orders has to be processed in close chronological order, which is particularly the case with different search engine response times, different waiting times for the results and thus overlap in time between different ones Carry out search requests, the integration module and the distribution module are advantageously assigned a common buffer memory module. Characteristic identification data, for example an identification number, are expediently stored in the buffer memory module, with the individual specified search orders output directly to the search engines with a corresponding one Identification number are provided. In the case of the answers provided by the search engines, a clear assignment of the result data records to the respective search request is then made possible in the integration module by using the characteristic identification data stored in the buffer memory, so that a consolidated and consistent evaluation of the incoming results is made possible.

The Distribution module is advantageously specifically designed to the incoming search order on the data side for the respective Search engine convertible data format or syntax to convert. For this purpose, the dispersion module is also advantageous with the second memory module connected in which syntax data for the search engines are deposited. Through targeted maintenance and care of the second memory module is thus made possible in a special way, updates if necessary the syntax used by the search engines or also the addition or deletion complete Search engines in a particularly simple way.

In Another advantageous embodiment is the distribution module on the input side a filter module connected to an input / output module via which individual search requests by the user can be specified. The filter module can in particular be designed for this be, based on adjustable or individually definable filter criteria suppress certain search queries completely or, for example, country or add customer-specific individual specifications. In particular, also be provided, based on the type of request, individual, Search engines recognized as particularly suitable for the subsequent processing of the Specify search order. The filter module can thus in particular for that too be designed depending on the respective search query or its thematic focus to make a preliminary or rough selection of the search engines to be used and therefore a rough strategy for specify the strategy below.

Basically, it can Distribution module and the integration module as separate units, for example on a software or hardware basis. A particularly compact one and therefore cheap Construction can be achieved by using the distribution module advantageously integrated with the integration module in a common unit is.

It is widespread practice, especially when performing Research on the Internet or World-Wide-Web that the user has in common with the hit lists generated for his search query, so-called "banners" are also displayed. In response to a search query, an advertising material database also referred to as an “ad server” together with the results randomly from those stored there Banners selected records with transmitted, which then be displayed on the input / output module. At a such a system, the individual banners are added however usually unspecified and especially without reference to that of the user topic specified by the search query, so that the user interest is only comparatively low on the respective advertising medium. Especially are advantageous further development of the above system now the distribution and integration module on the data side connected to an advertising material database, which acts as a response data record to an incoming search request for an identification signal delivers data record adapted to the search query. In other words: Because of the already planned, comparatively qualified Evaluation of the specified search query using the corresponding conversion enabled in the distribution module is now the generation of a content or thematic identifier for the search query enables when the search query is forwarded to the advertising material database a targeted selection of with respect to the identification signal, ie for example with regard to the topic being worked on, particularly suitable Advertising data record enables. Through such a targeted assignment of the given as the response signal Advertising media records is therefore a significantly increased Attention to the user regarding the advertising material used reachable.

Regarding the The aforementioned task is carried out using the aforementioned method Research system solved, by a majority of those supplied by the search engines of result data sets Result data are converted into a common data format, being subsequently under recourse to prioritization parameters stored in a memory module for everyone Result data record is assigned a relevance parameter. About the Integration module is thus provided for the delivered result data sets generate a consolidated and consistent relevance system in which the individual result records relative to each other with regard to given relevance criteria are positioned appropriately.

The prioritization parameters are advantageously selected as a function of the respective search engine, the relevance parameter being provided in an alternative or additional advantageous development for the respective result data record, taking into account one from the respective search engine output relevance characteristic value is determined.

advantageously, becomes a search request in common data format for each selected search engine converted to their specific data format and then to spent this. In an alternative or additional advantageous embodiment becomes an over an input / output module specified search order before its further Processing subjected to filtering.

advantageously, a search request is added to the search engines to an advertising material database, where an incoming response data record by means of an identification signal is assigned to the respective search query. By the identification signal is particularly possible the response data record with regard to predefinable criteria, for example with regard to the content or topics used, to the respective search query adjust so that, as a result, the user shares with his Search results are offered a thematic matching the search query becomes.

The Advantages achieved with the invention are in particular that through the use of the integration module that incoming Result records converted to a common data format and then one relative weighting and prioritization of the result data sets among themselves by assigning a "global" relevance value consistent sharing across multiple search engines allows is. Through the distribution module communicating with the integration module is still guaranteed that falling back to search engine-specific data stored in the memory module, such as For example, syntax data can be used to address various types Search engines with a common search query is enabled with a consolidated joint evaluation of the data from the the search engines addressed, taking into account the results provided of their specific syntax is made possible.

On embodiment the invention is explained in more detail with reference to a drawing. In it show:

1 schematically a research system for determining information from a database,

2 a conceptual representation of the research system 1 .

3 a schematic listing of the in the research system 1 ongoing processes,

4 an alternative representation for the overall system 1 .

5 schematically the structure of the administration of the research system 1 .

6 a scheme for a submission to a plurality of search engines,

7 a scheme for a distributed query and merging of the results,

8th a scheme for different search modes in the research system 1 .

9 a scheme for filtering search queries,

10 is a schematic representation of the in the search system 1 modules used,

11 schematic representation of the structure of a filter / interpreter module,

12 schematically a structure of an integration module,

13 schematically the structure of a server engine module,

14 schematically a dynamic model for a search process,

15 a flow chart for an initialization of a server engine,

16 a flow chart for processing a search query,

17 a flow chart for the preparation of adapters for a search query, and

18 a scheme for an admin application of the research system according to 1 ,

The research system 1 according to 1 is designed to determine information from a complex database, in particular from the Internet, the World Wide Web or another index, with particularly high research quality. This is the research system 1 data side with a plurality of search engines 2 connected, as are currently widely used for targeted searches of the Internet.

To ensure the simultaneous and joint use of a number of search engines 2 to enable, includes the research system 1 in a central evaluation unit 4 an integration module 6 which is from the search engines 2 hit lists or result data supplied in the form of result data records E can be supplied. The integration module 6 ensures that a common, consistently prioritized evaluation of the different search engines 2 delivered result data records E is enabled.

This is the integration module 6 designed to first convert the incoming result data records E into a common data format, so that a consolidated evaluation is possible. In order to enable the user to evaluate the possibly large amount of information in a targeted manner, the consolidated prioritization and relevance identification of the individual result data records E is provided, whereby according to predefined relevance criteria, all result data records E are assigned comparable relevance characteristic values. To make this possible, the integration module 6 with a memory module 8th connected in which relevant prioritization parameters P are stored for the assignment of the relevance parameters.

When prioritizing, it can in particular be provided that the search engines selected according to a user or an operator 2 to be provided with a comparatively high prioritization and thus a comparatively high relevance value. In this case, the prioritization parameters are selected depending on the respective search engine 2 , Alternatively or additionally, it can be provided in the manner of a normalization by a search engine 2 to standardize the result data records E delivered with respect to their relevance characteristic value to the result data record E that is already used by the respective search engine 2 has been assigned the highest relevance characteristic. In the form of a percentage, this could mean that the search engine 2 Result data record E recognized as the most relevant is provided with a preliminary relevance characteristic value of 100%, the other result data records E of this search engine 2 according to their from the search engine 2 assigned initial relevance parameter receive a correspondingly graduated relevance parameter. If this is for all search engines 2 is carried out equally, there is thus a relative prioritization of the result data records E of different search engines 2 essentially based on their relative relevance with regard to that of the respective search engine 2 the most relevant result data set E.

In addition, a prioritization parameter P can also be used to store a search engine-specific characteristic value which is general or dependent on the particular search query of the respective search engine 2 assigns a reliability parameter.

The evaluation unit 4 also includes a distribution module 10 that has a buffer memory 12 with the integration module 6 communicated. The distribution module 10 is still with a second memory module 14 connected in that for the search engines 2 characteristic data records, in particular characteristic for their data format or syntax, are stored. The distribution module is on the input side 10 via a filter module 16 with an input / output module 18 connected, which in turn is on the input side to the integration module 6 connected is. The input / output module 18 can be configured in particular as a conventional control unit, comprising a keyboard and an output means such as a screen.

When editing one by the user via the input / output module 18 Given the specified search query, this query is first filtered in the filter module 16 , For example, the elimination of search queries for a specific criterion can be provided, or country or region-specific specifications for processing the search query can be taken into account. Furthermore, the filter module 16 a pre-selection of the search engines recognized as suitable for the respective search queries 2 so the filter module 16 in the form of a rough strategy, the search engines to be entrusted with the search query 2 with pretends.

The search query filtered in this way 2 then becomes the distribution module 10 supplied, which uses the search engine-specific information stored in the second memory module and, in particular, syntax parameters, the respective search query in a large number of individualized search queries in the search engines selected 2 implement suitable data formats. Then the distribution module gives 10 the individualized search queries to the selected search engines 2 further, being in the manner of logging and to facilitate administration in the buffer memory 12 a profile of the search query is stored together with an identifier that identifies it.

The search engines started by feeding the individualized search query 2 then complete their search program on the World Wide Web and provide their response signals, for example in the form of hit lists, in which the found result data are found as result data records E according to the search engine 2 recognized relevance are listed in order. Based on the characteristic identification data for the respective search query, the result data records E are then in the integration module 6 reassigned to the respective search query, so that a targeted evaluation is possible even with overlapping searches.

In the integration module 6 is also carried out, possibly also using the second memory module 14 stored syntax information, a conversion of the result data records E into the common data format. Then takes place using the in the memory module 8th stored prioritization parameters P the generation of a relevance characteristic for each result data set E, the relevance characteristics generated in this way being used to produce a consolidated and consistent result hierarchy between the result data sets E. As a result, a homogenized result or hit list is generated for the user, in which the result data records E identified as most relevant are listed first. The assignment of the relevant relevance parameter using the prioritization parameters P can in particular with regard to the respective search engine 2 take place, which has delivered the relevant result data record E. Alternatively or additionally, when assigning the relevance characteristic value, one from the respective search engine 2 provided output relevance parameter are taken into account.

In addition to the search engines 2 is the research system 1 on the data side with an advertising material database 20 connected. In the advertising material database, also referred to as "ad server", advertising elements are stored in the form of so-called "banners", which are provided by the advertising material database 20 in response to an incoming search query to the research system 1 be returned. The research system 1 is designed in such a way that the advertising material database 20 sends selected and specified advertising elements back as a response data record to an incoming search query, the selection and specification being adapted to the respective search query in terms of content or subject. The selection of the returned advertising elements takes place in such a way that the respective advertising element is related to the search query in terms of content or topic, so that a comparatively increased interest of the user in the respective advertising element can be expected. For this purpose, the research system transmits 1 the respective search request together with an identification signal to the advertising material database 20 , the selection signal possibly suitable for the response data record being used on the basis of the supplied identification signal.

2 illustrates the components and actuators involved in relation to the (sub) system limits.

role Description

customer

"Customer" is any user who can start a search.

TSC filter manager

The Role “TSC Filter Manager "describes the employee group, in whose area of responsibility the filter definitions lie. This group of employees also evaluates the statistics logging of the research system.

TSC admin

The role of "admin" describes the group of employees responsible for operating the research system responsible for.

He administrates the configuration file (tsc.cfg) of the research system and monitored the error logs (tsc_error.log).

Description of the external systems

content

Summary

Here search results are provided on private and business homepages demonstrate.

interface

The Interface is analogous to the WWW search with Fast.

P4P server

Summary

The Search results from the Pay4Performance (P4P) server are provided here.

interface

The The GoTo search center interface is included via HTTP request (included Parameters) addressed. The results are delivered in XML format.

FAST search center

Summary

Here Internet-wide search results are provided, which are by means of the search engine "AllTheWeb" were found.

interface

The The interface of the FAST search center is included via HTTP request (included Parameters) addressed. The results are delivered in XML format.

Ad Server

Summary

The According to the search query, ad server delivers an advertising component, which flows into the overall presentation of the search results.

search system processes

In 3 the processes of the research system are shown in the overview and the process run in the "good" case is described.

TSC configuration

The Configuration of the research system is controlled via parameters that are stored in a separate file. This configuration file is saved in XML format in the file system of the search center. she can can be edited using a standard editor.

The modified Configuration can only be achieved with the restart of the "TSC search" processes become.

The process "TSC configuration" describes the necessary steps for the administration of the configuration file of the research system.

Maintain TSC exclusion filter

The TSC exclusion filters define words and phrases for the one Search string, if contained in this, blocked for certain channels becomes. These closures are country-specific and can be defined depending on the time become. Usually the filters are checked using a separate application (the TSC Admin Application). Ultimately, the file is the TSC exclusion filter the product of an export from the "Filter DB" and is available in CSV format If necessary, the file can be edited with a standard editor.

The Activation of the changed Filters can only be reached by restarting the "TSC search" processes.

The following steps are defined for the process "Maintain TSC exclusion filter":

TSC restart

The "TSC Restart" process can be started manually or automatically by other scripts, such as in the context of filter activation.

The "fast-cgi" environment of Apache Web server automatically takes care of restarting the preconfigured number of "TSC search" processes.

Edit TSC search query

The Process "TSC Edit search query " the central process that is realized through the research system becomes.

The following procedure outlines this process at a high level. This process will be detailed later in the document.

Evaluation of TSC statistics

The Statistics file can be viewed using a text editor or imported into other programs because the file is in CSV format is present.

Evaluation of TSC error log

through the error log can be viewed in a text editor.

TSC platform analysis model

Use case diagram, rough analysis

4 describes the rough relationships between the actors and use cases involved in the research system (TSC).

Use case: TSC configuration

The "TSC Admin" is given the opportunity view the configuration of the search center stored in an XML file and change. To do this, he uses the admin tool, which is on a separate, not for Productive environment Computer is running. So that changed Configuration takes effect is a restart of the individual processes necessary.

Business process diagram

In the 5 Administration of the research system shown schematically contains various actions. This includes changing the exclusion filter and activating the changes. To activate it, a script is started that copies the current configuration to the computers of the production system. The computers of the production system can be seen in the diagram at the lower right corner. The administrator himself appears as a "worker" and appears here in the top left of the diagram.

Use case: TSC exclusion filter care for

The "TSC filter manager" maintains the words that should be excluded from the search. These words can be per Country (portal) as well as limited time.

After change the exclusion filter is to restart the search system (TSC), hence the changes can become active.

Use case: configuration and activate filter

The Configuration is opened using the admin tool from the admin computer transfer all computers involved in the search center and then restart of the system. By restarting u. a. new processes started during their Initialization of the current configuration and the filter tables load.

Use case: TSC restart

The Search Center is restarted using the admin tool. In doing so All processes are organized on all computers involved in the search center ended and restarted. A restart is necessary to make changes the exclusion filter or the configuration take effect.

Use case: search query to edit

The "customer" gives in the search mask the input / output system its search expression. This can come from multiple words associated with AND, OR, and NOT operators. This search expression is sent to the server engine from this interprets the requirements of the various (e.g. T. external) adapted search engines and sent to them themselves. Considering The different search results are made of configurable timeouts then integrated into an overall view and to the customer's browser returned.

Use case: evaluation of statistics

The "TSC filter manager" can be operated via a Text editor to view the search terms and processing times. In addition, should the evaluator possibilities are provided, which have an ad hoc standard evaluation according to criteria to be defined.

Use case: evaluation of error log

The "TSC Admin" can be via a Text editor to view the error messages stored in a text file.

Use case diagrams, fine analysis

In 6 the process of a query to the search engines is shown schematically. "Send interpreted request" processes the incoming search request, adjusts it to the syntax of the external search engines involved and sends it on to them.

"Results Integrate "receives the Search results of the search engines and presents them to the user in the desired Format.

7 shows schematically the flow of a distributed query with subsequent merging of the results. "Collect results" accumulates the search results arriving from the search engines. A search result must arrive within a configurable time limit (timeout). If this is not the case, this is interpreted as an error.

"Error handling" takes over the logging of fault conditions into an error file.

"Prioritize" means that of the various search engines delivered results according to certain Business rules.

error Treated and prioritized search results are "integrated into the results page" converted into the desired output format.

The user / customer has, as in 8th shown two ways of searching. With "simple searches" the user can enter one or more search terms and possibly search operators that the FAST-CGI Must match syntax.

With "extended Search "can the Users first select the type of web content they are looking for (website, Picture, music etc.). Depending on the content type, there are customized search options to disposal.

The in 9 In addition to the configuration data, the schematically represented filter or interpreter receives the current "query string". This character string contains parameters and the search query of the user. The search query is modified according to the filter rules of the respective system (search engine). A further step builds on the respective "query strings "and some parameters a complete URL for the search engine. This URL can be identified in the above diagram as "target_URL".

Module overview

The overall system is divided as in 10 shown, essentially in three modules. These are described in more detail below. Adapter classes exist in the interpreter and integrator, each of which is tailored to exactly one target system. All adapter classes have a static and a dynamic part. The static part is represented by configuration data. The dynamic part depends on the search query.

Interpreter module

The module BRISBANE interpreter contains, as in 11 shown classes that serve the filtering and other processing of the search queries. In general it can be said that the interpreter provides the basic data for the integrator. The functionality of the filtering is strongly dependent on the target system (search engine) and is hidden in the corresponding adapter classes.

Module integrator

In the 12 Classes shown in the Integrator module are responsible for forwarding the modified search queries to the target systems. Further tasks of the integrator are the compilation of the search results and their conversion from XML to HTML, the display of the results page and the provision of the statistical data.

For every request An instance of the Request class is created on a target system. These objects represent the communication interface between search centers and target systems.

The Communication with the target systems is carried out by the RequestManager managed. However, it is not for responsible for presenting the results.

Server Engine module

This in 13 Module shown by its classes represents the main process of the search center. It has an instance of the "Configuration" class, which enables access to all configuration data. Within the process, configuration data are stored in different structures, such as AdapterMap.

The Server engine is also for the instantiation of interpreter and integrator as well as the orderly termination of the process (when a signal arrives).

The Class logging is for logging errors and statistical data.

dynamic model

In 14 shows the dynamic behavior of the individual components and the flow of information between them.

After starting the search center software, the configuration is read out. Depending on the configuration, appropriate exclusion filters and a country-specific XSL file are then loaded. The process is now waiting for a search request. If this arrives, the search expression is interpreted. Words that appear in exclusion filters are removed from the search query. In addition, the request to the syntax of betei adjusted search engines. The modified search terms are now sent to the search engines. Depending on the format of the returned results, a transformation to HTML format may be necessary (for example, if the results are in XML format). The results are sent to the customer's browser. The process waits again until the next request arrives. The individual search queries are saved with the search expression, time stamp and duration of processing in the statistics file "tsc_statistics". In addition, error states are entered in the file "tsc_error_log" throughout the entire process.

Server module Engine

initialization

During initialization, as in 15 shown, generated depending on active channels, configuration objects and instances of all adapters. The channels correspond to the previously mentioned channels and represent a very specific target system.

Processing a search query

When a request arrives, as in 16 shown, instances of interpreter and integrator created. All necessary information for the instances is provided by the server engine. The interpreter now prepares the modifications to the search query, the integrator sends them to the target systems and returns the formatted search results. The interpreter and integrator instances are then destroyed.

Interpreter module

The interpreter causes, as in 17 shown to generate all adapters required for the request, one adapter result each. Each individual adapter modifies the original search query using the corresponding filter rules. This query and other parameters are necessary to create an AdapterResult object. The generated AdapterResult object is returned to the interpreter, who adds it to a list.

Module integrator

To Generation of the integrator instance is in an initial one Status. The requests are processed by the call the method "do requests ()".

Main of the fast program

The Fast-CGI program contains a function "main ()" Web server called. Prerequisite for proper work the program is the correct initialization and entry on hold. The configuration is kept in a file. The complete path of this file is specified using a Environment variable.

configuration of the research system

The Search center is about configured a single file (tsc.cfg). This file is located on a separate computer that is not a production computer is used. The configuration file serves as a template for all configuration files per computer in the production environment. When the configuration changes (e.g .: modification of the directory for the statistics files) the configuration file on the outside of the productive system modified computer and scripted on all computers copied in the production system.

Format of the filter table

On The request in the admin tool turns the database (mysql) into a current one Filter file created over transfer the distribution mechanisms to the search center computers becomes.

This Filter file has line by line entries that are within a line the values separated by semicolons.

A line is structured as follows (values in ⌷ are optional):
Country code; word or phrase to be filtered; [Start date]; Time of start; [End date]; Time end; Channel 1; Channel 2; ...; Channel n
z. B.
DE; Neu; 2000-10-10; 12: 00: 00; 2001-12-31; 23: 35: 00; WWW ;;;;;
DE; prohibited; 2001-10-20 01: 00: 00; 2001-12-31; 20: 00: 00; WWW; P4P; BRISBANE; PHP; SP; AD
UK; beer ;; 00: 00: 00 ;; 00: 00: 00; WWW ;; BRISBANE ;; SP;
CH; alcohol; 2001-09-30; 15: 00: 00 ;; 00: 00: 00; WWW ;; BRISBANE; PHP ;;

Activation of configuration and filter table

The Configuration and filter table are activated by the Start scripts.

Format of the log file

The The name of the statistics file is in the configuration file (tsc.cfg) specified, e.g. B. tsc_statistics_ <PID>. <PID> will then replaced by the process ID during operation. Because several at the same time Processes are running, there are accordingly many log files.

All Statistics files are in a central directory.

This Statistics files are saved in CSV format. To the results there is a separate process due to the large volume of data on a specially for that configured server should run. The search processes write either their data directly to the evaluation server or the evaluation process collects the at regular intervals Statistics data of all search processes.

Structure of the file

The Statistics file consists of 4 basic columns and per real channel there are another 4 columns.

Basic columns:

• 1. Entered search terms
• 2. Time of the search
• 3. Search mode
• 4. Total time in ms or timeout

Columns per channel:

• 1. Name
• 2. Search terms used
• 3. Duration in ms or timeout
• 4. Number of results

Quantity structure

• Basic columns approx. 50 bytes
• Channel columns approx. 50 bytes each

Maximum load 500 requests per second, approx. 250 bytes per request:
→ 125 KB per second → 450 MB per hour → 10.8 GB per day

Error log format

The Name of the error log file is in the configuration file (tsc.cfg) specified, e.g. B. tsc_err_ <PID>. <PID> will then replaced by the process ID during operation. Because several at the same time Processes are running, there are accordingly many error log files.

Entries in the error log are made line by line. Within an entry, the values are comma-separated and have the following sematical meaning: time, original URL, adapted URL, time for the trip NEN request, number
Hit, description of the error that occurred
z. B .:

If a general error occurs, it is logged in the form, time, description of the error that occurred, e.g. B.

description the style sheets

The XSL stylesheets are used within the search center to collect the Answers in XML form to an HTML page according to the current one Implement layout specifications. The data for a website is the Style sheet in XML, or passed through style sheet parameters. There is a style sheet for each language in the search center. Within the Style sheets, the search results are prepared in HTML as well the links for the page forward and backward generated. According to the default settings, the corresponding one is then Tab structure over the search results and the advertising banners or the form for the new search created.

Administration of the filter table

context

The context of the admin application is in 18 shown.

1

search system

2

search engines

4

central evaluation

6

integration module

8th

memory module

10

distribution module

12

buffer memory

14

memory module

16

filter module

18

Input / Output Module

20

Advertising media database

e

Result records

P

prioritization parameters

Claims (15)

Research system ( 1 ) to determine information from a database using a plurality of search engines ( 2 ) with an integration module ( 6 ) which is used by the search engines ( 2 ) delivered result data, each comprising a plurality of result data records (E), converted into a common data format and then using a memory module ( 8th ) stored prioritization parameter (P) assigns a relevance parameter to each result data record (E). Research system ( 1 ) according to Claim 1, in which the prioritization parameters (P) depend on the respective search engine ( 2 ) are selected. Research system ( 1 ) according to claim 1 or 2, the integration module ( 6 ) the relevance parameter for the respective result data record (E) taking into account one from the respective search engine ( 2 ) determined output relevance characteristic value. Research system ( 1 ) according to one of claims 1 to 3, the integration module ( 6 ) via a common buffer memory module ( 12 ) with a distribution module ( 10 ) that communicates a search request for each selected search engine (in common data format) 2 ) converted into their specific data format and then output to it. Research system ( 1 ) according to claim 4, the distribution module ( 10 ) with a second memory module ( 14 ) in which syntax data for the search engines ( 2 ) are stored. Research system ( 1 ) according to claim 4 or 5, in the buffer memory module ( 12 ) characteristic identification data can be stored for individual search requests. Research system ( 1 ) according to one of claims 4 to 6, the distribution module ( 10 ) on the input side via a filter module ( 16 ) with an input / output module ( 18 ) connected is. Research system ( 1 ) according to one of claims 4 to 7, the distribution module ( 10 ) with the integration module ( 6 ) is integrated into a common unit. Research system ( 1 ) according to one of claims 4 to 8, wherein the distribution and the integration module ( 6 ) on the data side with an advertising material database ( 20 ) are connected, which as a response data record to an incoming search query delivers a data record adapted to the search query with regard to an identification signal. Method for determining information from a database, in particular from the World Wide Web, using a research system ( 1 ) according to one of Claims 1 to 9, in which the search engines ( 2 ) delivered, each comprising a plurality of result data records (E), are converted into a common data format, and then using a memory module ( 8th ) stored prioritization parameters (P) each result data set (E) is assigned a relevance parameter. Method according to Claim 10, in which the prioritization parameters (P) depend on the respective search engine ( 2 ) to get voted. Method according to Claim 10 or 11, in which the relevance parameter for the respective result data record (E), taking into account one of the respective search engines ( 2 ) provided output relevance characteristic value is determined. Method according to one of Claims 10 to 12, in which a search request arriving in the common data format for each selected search engine ( 2 ) converted into their specific data format and then output to it. Method according to one of claims 10 to 13, in which a via an input / output module ( 18 ) the specified search request is subjected to filtering before it is processed further. Method according to one of Claims 10 to 14, in which a search request is additionally sent to an advertising material database ( 20 ) is output, an incoming response data record being assigned to the respective search query by means of an identification signal.