Suche Bilder Maps Play YouTube News Gmail Drive Mehr »
Anmelden
Nutzer von Screenreadern: Klicke auf diesen Link, um die Bedienungshilfen zu aktivieren. Dieser Modus bietet die gleichen Grundfunktionen, funktioniert aber besser mit deinem Reader.

Patentsuche

  1. Erweiterte Patentsuche
VeröffentlichungsnummerCN104679784 A
PublikationstypAnmeldung
AnmeldenummerCN 201310634240
Veröffentlichungsdatum3. Juni 2015
Eingetragen3. Dez. 2013
Prioritätsdatum3. Dez. 2013
Veröffentlichungsnummer201310634240.2, CN 104679784 A, CN 104679784A, CN 201310634240, CN-A-104679784, CN104679784 A, CN104679784A, CN201310634240, CN201310634240.2
Erfinder周小伟
Antragsteller上海博科资讯股份有限公司
Zitat exportierenBiBTeX, EndNote, RefMan
Externe Links:  SIPO, Espacenet
O2B intelligent searching method and system
CN 104679784 A
Zusammenfassung
The invention discloses an O2B intelligent searching method and system and belongs to the technical field of computer. The O2B intelligent searching method includes that carrying out logic association on words, and building a mapping lexicon; acquiring search content input by a user; carrying out semantic analysis on the search content input by the user; carrying out quadratic search on the semantic analysis result to obtain associated words, and using the associated words as search words; reasoning the search words in a rule base, and obtaining the search result; outputting the research result. The O2B intelligent searching method and system have beneficial effects that the improved O2B intelligent searching method is provided for overcoming the disadvantage that an existing search engine cannot meet the real demand of a user, through semantic analysis and logic mapping, the complicated and changing language is simplified, and the real demand of the user is translated precisely to provide precise search service.
Ansprüche(10)  übersetzt aus folgender Sprache: Chinesisch
1.一种02B智能搜索方法,其特征在于,所述02B智能搜索方法包括以下步骤: a)词库建构步骤:将词汇进行逻辑关联,建立映射词库; b)信息录入步骤:获取用户输入的搜索内容; c)语义分析步骤:对用户输入的搜索内容进行语义分析; d) 二次检索步骤:将语义分析结果,二次检索得出关联词,作为搜索词; e)信息检索步骤:将所述搜索词在规则库中进行推理运算,并得到搜索结果; f)信息反馈步骤:输出所述搜索结果。 A 02B intelligent search method, wherein said 02B intelligent search method comprising the steps of: a) Construction thesaurus step: the vocabulary logical association, the mapping thesaurus; b) information input step: get user input search content; c) semantic analysis step: search the contents of the user input semantic analysis; d) the second retrieval step: semantic analysis results obtained secondary retrieve related words, as the search term; e) information retrieval steps: The search word reasoning operation in the rule base, and get search results; f) Feedback step: output of the search results.
2.根据权利要求1所述的02B智能搜索方法,其特征在于,所述词库建构步骤包括: 词汇逻辑关联步骤:运用图模型处理特征词权重,将词汇进行逻辑关联,并以此建立映身寸关系; 保存映射信息步骤:将所述词汇映射关系存入所述词库。 The 02B intelligent search method according to claim 1, wherein said step includes Thesaurus Construction: Vocabulary logically associated step: using graphical models feature weighting process, the vocabulary logical association, and to establish Ying body inch relations; mapping information stored step: mapping relationship is stored in the vocabulary of the thesaurus.
3.根据权利要求1所述的02B智能搜索方法,其特征在于,所述语义分析步骤包括: 分词处理步骤:利用分词技术对所述用户输入的搜索内容进行分词处理; 提取关键词步骤:根据分词主谓宾等中文语法特点查找关键词。 02B according to claim 1 wherein the intelligent search method claims, characterized in that said semantic analysis steps include: word processing step: use the word technology to search the contents of the user input word processing; extracting keywords steps of: Word SVO and other Chinese characteristic Find keyword syntax.
4.根据权利要求1所述的02B智能搜索方法,其特征在于,所述二次检索步骤为对所述关键词进行相关度分析,根据特征词权重判断词汇关联度,获取与其在所述词库中关联度最高的信息,作为搜索词。 02B according to one of the intelligent search method claims, characterized in that the secondary search step is in the word of the keyword correlation analysis, based on vocabulary term weight determine the degree of association, access to its repository highest correlation, as the search term.
5.根据权利要求1所述的02B智能搜索方法,其特征在于,所述信息检索步骤包括: 维度分析步骤:将所述用户输入的搜索内容进行维度分析,提取多维信息; 规则运算步骤:使用规则引擎将所述搜索词结合所述多维信息在规则库中进行推理运算,并得到搜索结果。 02B according to one of the intelligent search method claims, characterized in that the information retrieving step includes: dimensional analysis step: the user entered search the contents of dimensional analysis, extraction multidimensional information; rule operation step: Use rules engine combined with the search word in the multidimensional information inference operation in the rule base, and get search results.
6.一种02B智能搜索系统,其特征在于,所述02B智能搜索系统包括: a)词库建构模块:用于将词汇进行逻辑关联,建立映射词库; b)语义分析模块:用于对用户输入的搜索内容进行语义分析; c) 二次检索模块:用于将语义分析结果,二次检索得出关联词,作为搜索词; d)信息检索模块:用于将所述搜索词在规则库中进行推理运算,并得到搜索结果; e)数据交互模块:用于接收所述用户输入的搜索内容,并输出所述搜索结果。 A 02B intelligent search system, characterized in that the 02B intelligent search system comprising: a) thesaurus building blocks: logical association for the vocabulary, the mapping thesaurus; b) semantic analysis module: for search the contents of user input semantic analysis; c) secondary retrieval module: for the semantic analysis results obtained secondary retrieve related words, as the search term; d) information retrieval module: For the search term in the rule base reasoning in operation and get the search results; e) data exchange module: means for receiving the user input search content, and outputs the search results.
7.根据权利要求7所述的02B智能搜索系统,其特征在于,所述词库建构模块包括: 词汇逻辑关联单元:用于运用图模型处理特征词权重,将词汇进行逻辑关联,并以此建A映射关系; 词汇映射关系库:用于储存所述词汇映射关系。 According to claim 7, wherein the intelligent search system 02B, wherein said thesaurus building blocks include: Vocabulary logically associated units: for processing using graphical models feature weighting, will logically associated vocabulary, and thus A mapping relationship built; vocabulary mapping between the Library: for storing the vocabulary mappings.
8.根据权利要求7所述的02B智能搜索系统,其特征在于,所述语义分析模块包括: 分词处理单元:用于利用分词技术对所述用户输入的搜索内容进行分词处理; 关键词提取单元:用于根据分词主谓宾等中文语法特点查找关键词。 Keyword extraction unit; segmentation techniques for utilizing the contents of the user entered search word processing: according to claim 7, wherein 02B intelligent search system, wherein the semantic analysis module include: word processing unit : Find keywords based on segmentation for SVO and other Chinese grammar characteristics.
9.根据权利要求7所述的02B智能搜索系统,其特征在于,所述二次检索模块用于对所述关键词进行相关度分析,根据特征词权重判断词汇关联度,获取与其在所述词库中具有映射关系的信息,作为搜索词。 According to claim 7, wherein 02B intelligent search system, characterized in that the secondary retrieval module for the keyword correlation analysis, based on vocabulary term weight determine the degree of association, access thereto in the Thesaurus has information mappings, as the search term.
10.根据权利要求7所述的02B智能搜索系统,其特征在于,所述信息检索模块包括: 维度分析单元:用于将所述用户输入的搜索内容进行维度分析,提取多维信息;规则引擎:用于将所述搜索词结合所述多维信息在规则库中进行推理运算,并得到搜索结果。 According to claim 7, wherein 02B intelligent search system, wherein the information retrieval module includes: dimensional analysis unit: the user input is used to search the contents of dimensional analysis, extraction multidimensional information; rules engine: for the search word in combination with the multi-dimensional information in the rule base inference operation, and get search results.
Beschreibung  übersetzt aus folgender Sprache: Chinesisch
一种02B智能搜索方法及系统 One kind of intelligent search method and system 02B

技术领域 TECHNICAL FIELD

[0001] 本发明涉及计算机技术领域,尤其涉及一种02B智能搜索方法及系统。 [0001] The present invention relates to the field of computer technology, more particularly to a method and system 02B intelligent search.

[0002] [0002]

背景技术 Background technique

[0003] 随着信息时代的不断发展,互联网也走进了人们生活的各个角落。 [0003] With the continuous development of the information age, the Internet has entered every corner of people's lives. 轻轻一点搜索引擎,人们往往足不出户就可以获取各种信息,信息的共享大大便利了人们的生活。 Gently point the search engines, people tend to stay at home you can get all kinds of information, sharing of information has greatly facilitated people's lives. 电子商务的蓬勃发展也为人们通过互联网获取各种服务提供了便利。 The booming e-commerce but also for people to obtain a variety of services via the Internet provides a convenient. 但是,随着互联网用户群的不断增大,如何更准确的把握用户的真实意图,提供更快捷便利的服务,逐渐引起了服务提供商及中间商们的关注。 However, with the growing population of Internet users, how to more accurately grasp the true intentions of users, providing a more efficient and convenient services, gradually attracted service providers and brokers' attention.

[0004] 02B (Order To Business),即采购报价模式,是一种新型的电子商务模式。 [0004] 02B (Order To Business), namely purchasing quotation mode is a new e-commerce model. 02B模式以订单信息精准有效匹配供应商为特色开展新一代贸易拓展服务。 02B models to order information accurate and effective matching suppliers to carry out a new generation of business to expand service characteristics. 相较于传统的B2B、B2C模式,02B模式能够更为精准的为用户提供匹配其需求的各项服务。 Compared to the traditional B2B, B2C model, 02B model can provide users with a more precise matching of the service to their needs.

[0005] 传统的搜索引擎通过关键字检索,从数据库中找到符合该关键词的相关网页索弓丨,并将结果按一定的排名规则最终呈现给搜索用户。 [0005] The traditional search engines through keyword search, found for the keyword related pages 丨 bow rope from the database, and the results are ranked according to certain rules of the final presentation to the search user. 在实现本发明的过程中,发明人发现现有技术存在以下问题: In the process of implementing the present invention, the inventors have found the following problems of the prior art:

首先,现有的搜索引擎主要采取基于关键词的布尔表达式作为输入,这就造成了搜索引擎很难准确掌握隐含在关键词背后的用户的真实需求。 First, the existing keyword-based search engines mainly taken Boolean expressions as input, which resulted in a search engine is difficult to accurately grasp the real needs implicit in the keywords behind the user.

[0006] 其次,基于关键词的索引方式,造成现有的搜索方式较为机械,获取的搜索结果数量巨大且不加筛选。 [0006] Secondly, keyword-based indexing methods, resulting in a huge number of existing search more mechanical way to get search results without added filtering. “关键词全文检索”的结果就像扔给我们所有能获取的答案,毫无导向性。 "Keyword text retrieval," the result is like throwing all the answers we can get, there is no orientation. 对于关键词指向不够明确的搜索,其效果不免差强人意。 Pointing to a lack of clarity for the keyword search, the effect can not help but passable. 例如一个用户搜索“我的冰箱坏了”,他得到的结果是包含有“我的冰箱坏了”这一句话的各种网页,而如果用户需要找到冰箱维修等解决方式,则需要输入更多的关键词,这无形中增加了对搜索引擎应用不熟悉的一类用户的负担。 For example, a user searches for "My refrigerator is broken," the result is he got that includes "My refrigerator is broken." This sentence various Web pages, but if you need to find the refrigerator maintenance solutions, you need to enter more Keywords, which virtually increased the burden on the search engine application are not familiar with a class of users.

[0007] [0007]

发明内容 SUMMARY OF THE INVENTION

[0008] 本发明的目的在于,克服现有搜索引擎不能满足用户真实需求的缺陷,提供一种改进的02B智能搜索方法,其通过语义分析与逻辑映射,将复杂多变的语言化繁为简,在所有零散的检索结果中“穿针引线”勾勒出我们所需要的真正有用的信息。 [0008] The object of the present invention is to overcome the deficiencies of existing search engines can not meet the real needs of users, to provide an improved 02B intelligent search method through semantic analysis and logic mapping, complex language to simplify , scattered in all search results in "go-between" outlines the truly useful information that we need. 准确翻译出用户的真实需求,进而提供精准的搜索服务。 Accurate translation of the real needs of users, thus providing accurate search services.

[0009] 本发明的目的还在于提供一种实现上述02B智能搜索方法的02B智能搜索系统。 [0009] The object of the present invention is to provide a method to achieve the above the 02B 02B intelligent search intelligent search system.

[0010] 为实现上述发明目的之一,本发明提供一种02B智能搜索方法。 [0010] To achieve one object of the invention described above, the present invention provides an intelligent search method 02B. 所述技术方案如下: The technical solutions are as follows:

一种02B智能搜索方法,包括以下步骤: One kind 02B intelligent search method comprising the steps of:

词库建构步骤:将词汇进行逻辑关联,建立映射词库; 信息录入步骤:获取用户输入的搜索内容; Thesaurus Construction steps: vocabulary logical association, the mapping thesaurus; information input step: get the user-entered search content;

语义分析步骤:对用户输入的搜索内容进行语义分析; Semantic analysis steps: search for user input semantic content analysis;

二次检索步骤:将语义分析结果二次检索得出关联词,作为搜索词; Secondary search step: the semantic analysis result obtained secondary retrieve related words, as the search term;

信息检索步骤:将所述搜索词在规则库中进行推理运算,并得到搜索结果; Information retrieval step: the search word reasoning operation in the rule base, and get the search results;

信息反馈步骤:输出所述搜索结果。 Feedback step: the output of the search results.

[0011] 作为上述技术方案的优选,所述词库建构步骤包括: [0011] As a preferred aspect of the above, the thesaurus construct comprises:

词汇逻辑关联步骤:运用图模型处理特征词权重,将词汇进行逻辑关联,并以此建立映身寸关系; Vocabulary logic associating step: using graphical models feature weighting process, the vocabulary logical association, and to establish relations reflect inch body;

保存映射信息步骤:将所述词汇映射关系存入所述词库。 Save mapping information steps: the mapping relationship is stored in the vocabulary thesaurus.

[0012] 作为上述技术方案的优选,所述语义分析步骤包括: [0012] As a preferred aspect of the above, the semantic analysis step includes:

分词处理步骤:利用分词技术对所述用户输入的搜索内容进行分词处理; Word processing steps of: word segmentation of the user entered search word processing content;

提取关键词步骤:根据分词主谓宾等中文语法特点查找关键词。 Keywords extraction step: Find keywords based on segmentation SVO grammar and other Chinese characteristics.

[0013] 作为上述技术方案的优选,所述二次检索步骤为对所述关键词进行相关度分析,根据特征词权重判断词汇关联度,获取与其在所述词库中关联度最高的信息,作为搜索词。 [0013] As a preferred aspect of the above, the secondary retrieval step for the relevance of the keyword analysis, based on vocabulary term weight determine the degree of association, access to information relating to their highest degree in the lexicon, as the search term.

[0014] 作为上述技术方案的优选,所述信息检索步骤包括: [0014] As a preferred aspect of the above, the information retrieving step comprises:

维度分析步骤:将所述搜索词进行维度分析,提取多维信息; Dimensional analysis step: the search terms were dimensional analysis, multi-dimensional information extraction;

规则运算步骤:使用规则引擎将所述搜索词结合所述多维信息在规则库中进行推理运算,并得到搜索结果。 Rules of operation step: Use the search term rules engine in combination with the multi-dimensional information in the rule base inference operation, and get search results.

[0015] 相应地,为实现上述另一目的,本发明还提供了一种02B智能搜索系统。 [0015] Accordingly, to achieve the above another object, the present invention also provides a smart search system 02B.

[0016] 一种02B智能搜索系统,包括: [0016] A 02B intelligent search system, comprising:

词库建构模块:用于将词汇进行逻辑关联,建立映射词库; Thesaurus building blocks: logical association for the vocabulary, the mapping thesaurus;

语义分析模块:用于对用户输入的搜索内容进行语义分析; Semantic Analysis Module: search for user input semantic content analysis;

二次检索模块:用于将语义分析结果二次检索得出关联词,作为搜索词; Secondary retrieval module: for the semantic analysis results obtained secondary retrieve related words, as the search term;

信息检索模块:用于将所述搜索词在规则库中进行推理运算,并得到搜索结果; Information retrieval module: the search word for reasoning operation in the rule base, and get the search results;

数据交互模块:用于接收所述搜索内容,并输出所述搜索结果。 Data exchange module: means for receiving the search for content, and outputs the search result.

[0017] 作为上述技术方案的优选,所述词库建构模块包括: [0017] As a preferred aspect of the above, the thesaurus building blocks include:

词汇逻辑关联单元:用于运用图模型处理特征词权重,将词汇进行逻辑关联,并以此建A映射关系; Vocabulary logical association unit: graph model for use word-processing feature weights will be logically associated vocabulary, and thus build A mapping relationship;

词汇映射关系库:用于储存所述词汇映射关系。 Vocabulary mapping between the library: for storing the vocabulary mappings.

[0018] 作为上述技术方案的优选,所述语义分析模块包括: [0018] As a preferred aspect of the above, the semantic analysis module comprises:

分词处理单元:用于利用分词技术对所述用户输入的搜索内容进行分词处理; Word Processing Unit: for use segmentation technique to search the contents of the user input word processing;

关键词提取单元:用于根据分词主谓宾等中文语法特点查找关键词。 Keyword extraction unit: Find keywords based on segmentation for SVO grammar and other Chinese characteristics.

[0019] 作为上述技术方案的优选,所述二次检索模块用于将所述关键词进行相关度分析,根据特征词权重判断词汇关联度,获取与其在所述词库中具有映射关系的信息,作为搜索词。 Information [0019] As a preferred aspect of the above, the secondary keyword retrieval module for the correlation analysis, based on vocabulary term weight determine the degree of association, access to which they have mapping in the lexicon as a search term.

[0020] 作为上述技术方案的优选,所述信息检索模块包括: [0020] As a preferred aspect of the above, the information retrieval module comprising:

维度分析单元:用于将所述搜索词进行维度分析,提取多维信息; Dimensional analysis unit: carry out the search word for dimensional analysis, multi-dimensional information extraction;

规则引擎:用于将所述搜索词结合所述多维信息在规则库中进行推理运算,并得到搜索结果。 Rules engine: the search for multi-dimensional information in combination with the word inference operation in the rule base, and get the search results.

[0021] 与现有技术相比,本发明的有益效果是:02B智能搜索方法或02B智能搜索系统,通过对用户语词的分析解读,更准确的挖掘隐含在关键字背后的用户的真实需求,从而化繁为简,筛选出更符合用户要求的搜索结果。 [0021] Compared with the prior art, the beneficial effects of the present invention are: 02B intelligent search methods or 02B intelligent search system, by analyzing the interpretation of the user's words, more accurate implicit dig real needs behind the user's keyword thereby simplify screened more in line with user requirements of search results.

[0022] [0022]

附图说明 Brief Description

[0023] 图1是本发明一个实施例中02B智能搜索方法的流程图; [0023] FIG. 1 is a flow chart showing 02B intelligent search method according to an embodiment of the present invention;

图2是本发明一个实施例中词库建构步骤的流程图; Figure 2 is a flow chart showing the steps of the construction of a thesaurus embodiment of the present invention;

图3是本发明一个实施例中语义分析步骤的流程图; Figure 3 is a flow diagram of one embodiment of semantic analysis step embodiment of the invention;

图4是本发明一个实施例中信息检索步骤的流程图; Figure 4 is a flowchart illustrating the steps of an information retrieval embodiment of the present invention;

图5是本发明一个实施例中02B智能搜索系统的结构示意图; Figure 5 is a block diagram representation of 02B intelligent search system according to one embodiment of the present invention;

图6是本发明一个实施例中词库建构模块包括的单元图; Figure 6 is a section view showing the construction of the thesaurus module includes an embodiment of the present invention;

图7是本发明一个实施例中语义分析模块包括的单元图; Figure 7 is a section view showing a semantic analysis module includes an embodiment of the present invention;

图8是本发明一个实施例中信息检索模块包括的单元图。 Figure 8 is a section view of the information retrieval module comprises an embodiment of the present invention.

[0024] [0024]

具体实施方式 DETAILED DESCRIPTION

[0025] 为了更清楚地展示本发明的目的、技术方案和优点,下面将结合具体实施例及附图对本发明的实施方式作较为详细地描述。 [0025] In order to more clearly show the object of the present invention, technical solutions and advantages will now be described with reference to specific embodiments and the drawings in greater detail embodiments of the present invention will be. 但这些实施方式并不限制本发明,本领域的普通技术人员根据这些实施方式所轻易做出的结构、方法、或功能上的变换均包含在本发明的保护范围内。 However, these embodiments do not limit the present invention, one of ordinary skill in the art based on these structures made easily embodiment, a method, or the conversion function are included within the scope of the present invention.

[0026] 如图1所示,本发明第一实施例提出了一种02B智能搜索方法,包括以下步骤: 词库建构步骤101、将词汇进行逻辑关联,建立映射词库; [0026] As shown in Figure 1, the first embodiment of the invention proposes a 02B intelligent search method comprising the steps of: thesaurus construction of step 101, the vocabulary logical association, the mapping thesaurus;

信息录入步骤102、获取用户输入的搜索内容; Information input step 102, get user input of search content;

语义分析步骤103、对用户输入的搜索内容进行语义分析; Semantic analysis step 103, the user entered search semantic content analysis;

二次检索步骤104、将语义分析结果二次检索得出关联词,作为搜索词; The second search step 104, the semantic analysis results obtained secondary retrieve related words, as the search term;

信息检索步骤105、将所述搜索词在规则库中进行推理运算,并得到搜索果; Information retrieval step 105, the reasoning of the search word operation in the rule base, and get search results;

信息反馈步骤106、输出所述搜索结果。 Feedback step 106, the output of the search results.

[0027] 如图2所示,在本发明中,所述词库建构步骤101包括: [0027] As shown, in the present invention, the step 101 includes a thesaurus 2 Construction:

词汇逻辑关联步骤201、将词汇进行逻辑关联,建立映射关系。 Vocabulary logical association step 201, the logic associated vocabulary, establish mappings.

[0028] 具体的,此步骤利用图模型处理特征词权重,将词汇进行逻辑关联,并以此建立映射关系。 [0028] Specifically, this step is the use of graphical models feature weighting process, the vocabulary logical association, and to establish mappings. 例如:“煤气”对应“灶头”、“冰箱”对应“家电”。 For example: "gas" corresponds to the "stove", "refrigerator" corresponds to the "home appliances." 需要说明的是:这里的特征词权重以自动采集和人为干预的方式调整,因而针对一些网络新词汇也能较为迅速地获取。 It should be noted that: the right word here feature weight to automatically collect and how human intervention to adjust, and thus can more quickly access the network for some of the new vocabulary. 例如:“高帅富”对应“男性”。 For example: "Gao Shuai rich" corresponds to the "male."

[0029] 保存映射信息步骤202、将所述词汇映射关系存入所述词库。 [0029] mapping information saving step 202, the stored vocabulary mapping between the thesaurus.

[0030] 如图3所示,在本发明中,所述语义分析步骤103包括: [0030] shown in Figure 3, in the present invention, the semantic analysis step 103 comprises:

分词处理步骤301、利用分词技术对所述用户输入的搜索内容进行分词处理。 Word processing step 301, the use of word segmentation of the user entered search word processing content. 例如:用户输入“我的煤气坏了”,通过分词处理,被分割为:“我的,煤气,坏了”。 For example: the user enters "My bad gas" through word processing, is divided into: "My, gas, bad."

[0031] 提取关键词步骤302、根据分词主谓宾等中文语法特点查找关键词。 [0031] Step 302 extracts keywords, based on segmentation SVO grammar and other Chinese characteristics to find keywords. 例如:上述“我的煤气坏了”关键词为“煤气”。 For example: the aforementioned "My bad gas" keyword "gas."

[0032] 具体的,在本发明中,所述二次检索步骤104为对所述关键词进行相关度分析,获取与其在所述词库中关联度最高的信息,作为搜索词。 [0032] Specifically, in the present invention, the secondary search step 104 keywords related to the analysis, to obtain information relating to their highest degree in the lexicon as a search term. 例如:上述关键词“煤气”经过相关度分析,转换为与其具有映射关系的信息“灶头”,并将“灶头”作为搜索词。 For example: The above keyword "gas" After correlation analysis, is converted to its relationship with a mapping information "stove" and "stove" as a search term.

[0033] 如图4所示,在本发明中,所述信息检索步骤105包括: [0033] 4, in the present invention, the information retrieving step 105 comprises:

维度分析步骤401、将所述用户输入的搜索内容进行维度分析,提取多维信息。 Dimensional analysis step 401, the user enters a search contents dimensional analysis, extraction multidimensional information. 例如:上述“我的煤气坏了”经过维度分析,“我的”可提取出隶属关系维度,“坏了”可提取出判定维度。 For example: the aforementioned "My bad gas" After dimensional analysis, "my" can extract the affiliation dimension, "bad" to extract the judgment dimension.

[0034] 规则运算步骤402、使用规则引擎将所述搜索词结合所述多维信息在规则库中进行推理运算,并得到搜索结果。 [0034] Rule computing step 402, using the rules engine to the search word in combination with the multi-dimensional information in the rule base inference operation, and get search results.

[0035] 如图5所不,本发明另一实施例提出了一种02B智能搜索系统,包括: [0035] FIG. 5 is not, another embodiment of the invention proposes a 02B intelligent search system, comprising:

词库建构模块501、用于将词汇进行逻辑关联,建立映射词库; Thesaurus construction module 501 for vocabulary logical association, the mapping thesaurus;

语义分析模块502、用于对用户输入的搜索内容进行语义分析; Semantic analysis module 502, is used to search the contents of user input semantic analysis;

二次检索模块503、用于将语义分析结果二次检索得出关键词,作为搜索词; Secondary retrieval module 503 for semantic analysis results obtained secondary retrieval keyword as a search term;

信息检索模块504、用于将所述搜索词在规则库中进行推理运算,并得到搜索结果; 数据交互模块505、用于接收所述用户输入的搜索内容,并输出搜索搜索结果。 Information retrieval module 504, for the search term were in the rule base inference computing, and get search results; data exchange module 505, for receiving the user input of search content, and outputs a search result.

[0036] 如图6所示,在本发明中,所述词库建构模块501包括: [0036] FIG. 6, in the present invention, the thesaurus construction module 501 includes:

词汇逻辑关联单元601、用于将词汇进行逻辑关联,建立映射关系。 Vocabulary logical association unit 601 for logically associated vocabulary, establish mappings.

[0037] 具体的,此单元利用图模型处理特征词权重,将词汇进行逻辑关联,并以此建立映射关系。 [0037] Specifically, this means using graphical models feature weighting process, the vocabulary logical association, and to establish mappings. 例如:“煤气”对应“灶头”、“冰箱”对应“家电”。 For example: "gas" corresponds to the "stove", "refrigerator" corresponds to the "home appliances." 需要说明的是:这里的特征词权重以自动采集和人为干预的方式调整,因而针对一些网络新词汇也能较为迅速地获取。 It should be noted that: the right word here feature weight to automatically collect and how human intervention to adjust, and thus can more quickly access the network for some of the new vocabulary. 例如:“高帅富”对应“男性”。 For example: "Gao Shuai rich" corresponds to the "male."

[0038] 词汇映射关系库602、用于储存所述词汇映射关系。 [0038] vocabulary mapping between the library 602 for storing the vocabulary mappings.

[0039] 如图7所示,在本发明中,所述语义分析模块502包括: [0039] As shown in FIG. 7, in the present invention, the semantic analysis module 502 comprises:

分词处理单元701、用于利用分词技术对所述用户输入的搜索内容进行分词处理。 Word processing unit 701, for use segmentation technology to the user-entered search word processing content. 例如:用户输入“我的煤气坏了”,通过分词处理,被分割为:“我的,煤气,坏了”。 For example: the user enters "My bad gas" through word processing, is divided into: "My, gas, bad."

[0040] 关键词提取单元702、用于根据分词主谓宾等中文语法特点查找关键词。 [0040] The keyword extracting unit 702, based on keywords used to find word SVO grammar and other Chinese characteristics. 例如:上述“我的煤气坏了”关键词为“煤气”。 For example: the aforementioned "My bad gas" keyword "gas."

[0041] 具体的,在本发明中,所述二次检索模块503用于对所述关键词进行相关度分析,获取与其在所述词库中关联度最高的信息,作为搜索词。 [0041] Specifically, in the present invention, the secondary retrieval module 503 for the keyword correlation analysis to obtain information relating to their highest degree in the lexicon as a search term. 例如:上述关键词“煤气”经过相关度分析,转换为与其具有映射关系的信息“灶头”,并将“灶头”作为搜索词。 For example: The above keyword "gas" After correlation analysis, is converted to its relationship with a mapping information "stove" and "stove" as a search term.

[0042] 如图8所示,在本发明中,所述信息检索步骤504包括: [0042] Figure 8, in the present invention, the information retrieving step 504 includes:

维度分析单元801、用于将所述户输入的搜索内容进行维度分析,提取多维信息。 Dimensional analysis unit 801 for the user entered search contents dimensional analysis, extraction multidimensional information. 例如:上述“我的煤气坏了”经过维度分析,“我的”可提取出隶属关系维度,“坏了”可提取出判定维度。 For example: the aforementioned "My bad gas" After dimensional analysis, "my" can extract the affiliation dimension, "bad" to extract the judgment dimension.

[0043] 规则引擎802、用于将所述搜索词结合所述多维信息在规则库中进行推理运算,并得到搜索结果。 [0043] Rule 802 engine for the search word in combination with the multi-dimensional information in the rule base inference operation, and get search results.

[0044] 本发明实施例提供的上述技术方案的全部或部分可以通过程序指令相关的硬件来完成,所述程序可以储存在可读取的存储介质中,所述存储介质包括:R0M、RAM、磁盘、光盘等各种可以存储程序代码的介质。 [0044] Examples provided in the technical proposal in whole or in part by a program instructing relevant hardware embodiment of the present invention is accomplished, the program may be stored in a storage medium that can be read, the storage medium includes: R0M, RAM, disk, CD-ROM, and other media may store program code.

[0045] 上文所述仅为本发明的可行性实施例,并不用以限制本发明,凡未脱离本发明的精神和原则所作的任何等效实施方式或变更,均应包含在本发明的保护范围之内。 [0045] The feasibility of the above embodiments of the invention only and not intended to limit the present invention and which do not depart from the spirit of the invention and any equivalent embodiments or changes made in principle, should be included in the present invention. within the scope of protection.

Patentzitate
Zitiertes PatentEingetragen Veröffentlichungsdatum Antragsteller Titel
CN1759392A *27. Febr. 200412. Apr. 2006Nhn株式会社Method for generating a search result list on a web search engine
CN101916294A *27. Aug. 201015. Dez. 2010黄斌Method for realizing exact search by utilizing semantic analysis
CN103064838A *19. Okt. 201124. Apr. 2013阿里巴巴集团控股有限公司Data searching method and device
US6269361 *28. Mai 199931. Juli 2001Goto.ComSystem and method for influencing a position on a search result list generated by a computer network search engine
Klassifizierungen
Internationale KlassifikationG06F17/30
Juristische Ereignisse
DatumCodeEreignisBeschreibung
3. Juni 2015C06Publication
22. Juli 2015C10Entry into substantive examination