WO2015154679A1 - Procédé et dispositif pour classer des résultats de recherche de multiples moteurs de recherche - Google Patents

Procédé et dispositif pour classer des résultats de recherche de multiples moteurs de recherche Download PDF

Info

Publication number
WO2015154679A1
WO2015154679A1 PCT/CN2015/076111 CN2015076111W WO2015154679A1 WO 2015154679 A1 WO2015154679 A1 WO 2015154679A1 CN 2015076111 W CN2015076111 W CN 2015076111W WO 2015154679 A1 WO2015154679 A1 WO 2015154679A1
Authority
WO
WIPO (PCT)
Prior art keywords
search
search engine
result
sorting
search results
Prior art date
Application number
PCT/CN2015/076111
Other languages
English (en)
Chinese (zh)
Inventor
杨浩
吴凯
Original Assignee
北京奇虎科技有限公司
奇智软件（北京）有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 北京奇虎科技有限公司, 奇智软件（北京）有限公司 filed Critical 北京奇虎科技有限公司
Publication of WO2015154679A1 publication Critical patent/WO2015154679A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/951—Indexing; Web crawling techniques

Definitions

  • the present invention relates to the field of the Internet, and in particular to a method and apparatus for sorting search results of multiple search engines.
  • search engine ranking systems There are two main types of existing search engine ranking systems, one is an independent search engine, such as Google Music, and the other is a multi-search engine, such as meta search.
  • the first type of search engine lacks certain resources due to copyright resource issues, etc., and cannot satisfy user requirements. Because of the sorting analysis based on its own user behavior, other engine user behavior data is missing, and non-optimal sorting in big data.
  • the second type of search engine while satisfying the resource requirements, displays multiple search engine ranking results according to a fixed sorting method. It does not calculate multi-search engine result weights based on big data, and lacks reordering of results within the engine. Sorting analysis between engines, while optimizing the search results of multiple search engines, the results of each search engine are not well ordered and cannot meet the needs of users.
  • the present invention has been made in order to provide a sorting method for a plurality of search engine search results and a corresponding sorting device for a plurality of search engine search results that overcome the above problems or at least partially solve the above problems.
  • a method for sorting search results of multiple search engines including:
  • each search engine According to the user behavior weight of each search result, the search results of each search engine are separately rearranged, and each search engine is sorted;
  • the ranking results of each search engine and the search results of each search engine after rearrangement are displayed in the form of an online application of the tab.
  • a sorting apparatus for a plurality of search engine search results including:
  • a user requirement obtaining module adapted to receive a search request of the user
  • a multiple search engine result obtaining module configured to respectively obtain search results obtained by each search engine in response to the search request
  • the sorting module is adapted to re-arrange the search results of the respective search engines according to the user behavior weights of the respective search results, and sort the search engines;
  • the multi-search engine result presentation module is adapted to display the sorting result of each search engine and the search result of each search engine after the rearrangement in the form of an online application of the tab page.
  • a computer program comprising computer readable code, when the computer readable code is run on a computing device, causing the computing device to perform the multiple search engine of the present invention The sorting method of the search results.
  • a computer readable medium storing the computer program of the present invention is provided.
  • the search results obtained by each search engine in response to the search request are respectively acquired, and the search results of the respective search engines are respectively rearranged according to the user behavior weights of the respective search results.
  • sorting each search engine displaying the sorting result of each search engine and the search results of each search engine after rearranging in the form of an online application of the tab page.
  • the method uses user behavior weights to reorder the search results of each search engine separately, and sorts each search engine, which is better than the lack of resources and sorting of independent search engines, so that the ranking of each search engine in multiple search engines is optimized. Engine-to-engine sequencing is also optimized.
  • FIG. 1 shows a flow chart of a method for ranking multiple search engine search results in accordance with one embodiment of the present invention
  • FIG. 2 is a flow chart showing a method of sorting search results of multiple search engines according to another embodiment of the present invention.
  • FIG. 3 is a block diagram showing the structure of a sorting apparatus for multiple search engine search results according to an embodiment of the present invention
  • FIG. 4 is a block diagram schematically showing a computing device for performing a ranking method of multiple search engine search results in accordance with the present invention
  • Fig. 5 schematically shows a storage unit for holding or carrying program code of a method of implementing a file upload cloud disk according to the present invention.
  • FIG. 1 shows a flow chart of a method of ranking multiple search engine search results in accordance with one embodiment of the present invention. As shown in FIG. 1, the method includes the following steps:
  • Step S100 Receive a search request of the user, and respectively obtain search results obtained by each search engine in response to the search request.
  • Step S110 Re-arrange the search results of the respective search engines according to the user behavior weights of the respective search results, and sort the search engines.
  • the user behavior weight mainly reflects the amount of clicks on a certain search result by the users of the whole network, or the amount of click statistics of the group users on a certain search result within a certain period of time.
  • Step S120 displaying the sorting result of each search engine and the search result of each search engine after the rearrangement in the form of an online application of the tab page.
  • the search results obtained by each search engine in response to the search request are respectively obtained, and the search results of the search engines are respectively performed according to the user behavior weights of the respective search results.
  • Rearrange and sort each search engine to display the sorting results of each search engine and the search results of each search engine after rearranging in the form of an online application of the tab.
  • the method uses user behavior weights to reorder the search results of each search engine separately, and sorts each search engine, which is better than the lack of resources and sorting of independent search engines, so that the ranking of each search engine in multiple search engines is optimized. Engine The ordering is also optimized.
  • FIG. 2 shows a flow chart of a method for ranking multiple search engine search results in accordance with another embodiment of the present invention. As shown in FIG. 2, the method includes the following steps:
  • Step S200 Receive a search request of the user, and respectively obtain search results obtained by each search engine in response to the search request.
  • the user inputs a certain artist name XX through the client, and after receiving the user's search request, the server sends the user's search request to various search engines such as A, B, C, D, and E to obtain respective search engine responses.
  • Searching for the search result obtained by the search engine A wherein the search results obtained by the search engine A are: a, c, d, e, b; the search results obtained by the search engine B are: c, e, f, h, a; the search engine C
  • the search results obtained are: f, k, m, d, b; the search results obtained by the search engine D are: m, n, k, a, h;
  • the search results obtained by the search engine E are: a, e, o, d, n, analyze each search result and the sort position of each search result in each search engine.
  • Step S210 performing weight removal processing on the search results of the respective search engines.
  • the search results of the search engines obtained in step S200 are subjected to weighting processing, and the results obtained by the weighting are: a, b, c, d, e, f, h, k, m, n, o.
  • Step S220 Acquire user behavior weights of each search result and search engine relevance weights.
  • User behavior weight mainly reflects the number of clicks on a certain search result by users on the whole network, rather than just based on the click volume of a certain user or several users on a certain search result.
  • User behavior weights are based on big data, big data is also called Big data, or massive data, massive data, which means that the amount of data involved is so large that it cannot be reached in a reasonable time through current mainstream software tools. Capture, manage, process, and organize information that is more positive for helping business decisions. Big data is characterized by scale, high speed, diversity, and value. It integrates different forms of data from different sources, enabling real-time analysis rather than batch analysis, and discovering internal correlations between massive and frequent data.
  • the invention utilizes the advantage of big data, counts the clicks of the search results of the whole network users, and calculates the user behavior weights of the search results according to the clicks. Specifically, by counting the click amount of the search result in the step S200 by the user of the whole network, the user behavior weights of the search result after the weighting in step S210 are calculated according to the click amount: 20, 18, 25, 10, 14, 15 respectively. , 12, 19, 21, 13, 16.
  • Search engine relevance weights For example, analyzing the positions of each search result in the search engines A, B, C, D, and E, and using the position weights to indicate the order in which the search results appear in the search engine, the search results are in the search engine.
  • the position weights are: 50, 40, 30, 20, 10, respectively. Since the same search result may appear in different search engines, the weighted average of the position weights of the search results in different engines is taken as the search result.
  • Analyze the number of times the search result appears on the partner, and the number of times the search result after the statistical weighting appears on the partner is: 4, 2, 2, 3, 3, 2, 2, 2, 2, 1, according to The number of times the search result appears on the partner is determined by the number of times 40, 20, 20, 30, 30, 20, 20, 20, 20, 20, 10.
  • the relevance weight of the search engine is the weighted weight of the position weight and the number of times of the search result, and the multi-engine correlation weights of each search result are: 72.5, 30, 65, 53.3, 80, 60, 35 , 55, 60, 45, 40.
  • Step S230 according to the user behavior weights of each search result and the search engine relevance weights, the search results of all the search engines are comprehensively sorted to obtain an ideal comprehensive sorting result.
  • the weighted weights of the user behavior weights and the search engine correlation weights of the respective search results are separately calculated as the weight values of the respective search results, and the weight values of the respective search results are ranked to all the search engines in descending order.
  • the search results are sorted, and the ideal integrated sorting result is obtained by sorting the weight values of the respective search results from high to low.
  • Calculated the weight values of each search result are: 92.5, 48, 90, 63.3, 94, 75, 47, 74, 81, 58, 56, according to the weight value of each search result from high to low for all search engines
  • the search results are sorted to obtain the ideal comprehensive sorting results: e, a, c, m, f, k, d, n, o, b, h.
  • step S240 the search results of the respective search engines are separately rearranged by using the ideal integrated sorting result, and each search engine is sorted.
  • the search results of the search engines in step S200 are rearranged according to the ideal integrated sorting result obtained in step S220.
  • the order of the search results rearranged by each search engine should be compared with the ideal integrated sort result.
  • the order of precedence is the same.
  • the search results of the rearranged search engine A are: e, a, c, d, b;
  • the search results of the search engine B are: e, a, c, f, h;
  • the search result of the search engine C is: m, f, k, d, b;
  • search engine D search results are: a, m, k, n, h;
  • search engine E search results are: e, a, d, n, o.
  • the sum of the weight values of all the search results of each search engine is calculated as the weight value of the search engine, and each search engine is sorted according to the weight values of the respective search engines.
  • each search engine is sorted, and the rankings of each search engine are: B, A, E, D, C.
  • an ideal comprehensive ranking result is obtained according to the user behavior weight and the search engine correlation weight.
  • the present invention is not limited to this, and the ideal integrated ranking result may be obtained only according to the user behavior weight or only the search engine correlation weight.
  • the specific method is similar to the above method, and details are not described herein again.
  • the specific calculation methods of the user behavior weights, the search engine correlation weights, the weight values of the respective search results, and the weight values of the search engines described in the foregoing embodiments are all specific examples of the present invention, and according to actual conditions, The specific calculation method can be adjusted.
  • Step S250 displaying the sorting result of each search engine and the search result of each search engine after the rearrangement in the form of an online application of the tab page.
  • the client After obtaining the sorting result of each search engine and the search results of each search engine after the rearrangement, the client displays in the form of an online application of the tab page.
  • a presentation manner may be various.
  • a plurality of buttons having a sequential order respectively represent respective search engines, wherein the order of the plurality of buttons is consistent with the order of the respective search engines.
  • the search results of the rearranged search engines are displayed by switching between a plurality of buttons.
  • a search request of a user is received, and search results obtained by each search engine in response to the search request are respectively obtained, and search results of each search engine are subjected to weight processing to obtain user behavior weights of each search result.
  • the search engine correlation weights according to the user behavior weights of each search result and the search engine relevance weights, the search results of all the search engines are comprehensively sorted to obtain an ideal comprehensive sorting result, and the ideal comprehensive sorting result is utilized.
  • the search results of each search engine are separately rearranged, and each search engine is sorted, and the sorting result of each search engine and the search results of each rearranged search engine are displayed in the form of an online application of the tab page.
  • the method first performs weight processing on the search results, and then uses User behavior weights and search engine correlation weights reorder search results and sort search engines, using big data about user behavior across the network to more accurately sort search results and search engines Sorting, which is better than the lack of resources, big data missing and sorting of independent search engines, makes the optimization of each search engine in multiple search engines and the inter-engine sorting is also optimized.
  • FIG. 3 is a block diagram showing the structure of a sorting apparatus for multiple search engine search results according to an embodiment of the present invention.
  • the device includes: a user requirement acquisition module 300, a multi-search engine result acquisition module 310, a ranking module 320, and a multi-search engine result presentation module 330.
  • the user requirement obtaining module 300 is adapted to receive a search request of the user.
  • the multiple search engine result obtaining module 310 is adapted to respectively obtain search results obtained by each search engine in response to the search request.
  • the sorting module 320 is adapted to re-arrange the search results of the respective search engines according to the user behavior weights of the respective search results, and sort the search engines.
  • the sorting module 320 further includes: an integrated sorting sub-module 340, a multi-search engine result rearranging sub-module 350, and a multi-search engine sorting sub-module 360.
  • the comprehensive sorting sub-module 340 is adapted to comprehensively sort the search results of all the search engines according to the user behavior weights of the respective search results, to obtain an ideal comprehensive sorting result.
  • the integrated sorting sub-module 340 further includes a weight calculating unit 370 adapted to acquire user behavior weights of the respective search results and search engine relevance weights.
  • User behavior weights mainly reflect the number of clicks on a search result by users across the network.
  • the weight calculation unit 370 is further adapted to: count the click amount of the search result of the whole network user, and calculate the user behavior weight of the search result according to the click amount.
  • the weight calculation unit 370 is further adapted to: analyze the position where the search result appears in each search engine and the number of occurrences in the partner, and calculate the search engine relevance according to the position of the search result appearing in each search engine and the number of occurrences in the partner. Weight.
  • the comprehensive sorting unit 380 is adapted to comprehensively sort the search results of all the search engines according to the user behavior weights of the respective search results and the search engine relevance weights, to obtain an ideal comprehensive sorting result.
  • the comprehensive sorting unit 380 is further adapted to: respectively calculate the weighted weights of the user behavior weights and the search engine correlation weights of the respective search results as the weight values of the respective search results, according to the weight values of the respective search results in descending order Sort the search results of all search engines.
  • the multiple search engine result rearrangement sub-module 350 is adapted to re-arrange the search results of the respective search engines by using the ideal integrated sorting result;
  • the multiple search engine sorting sub-module 360 is adapted to sort the individual search engines by using the ideal integrated sorting result.
  • the ideal comprehensive sorting result is obtained by sorting the weight values of the respective search results from high to low;
  • the multiple search engine ranking sub-module 360 is further adapted to: calculate a sum of weight values for all search results for each search engine as a weight value for the search engine; sort the individual search engines according to the weight values of the respective search engines.
  • the multiple search engine result presentation module 330 is adapted to present the ranking results of the respective search engines and the search results of the respective search engines after the rearrangement in the form of an online application of the tabs.
  • the multiple search engine result presentation module 330 is further adapted to: in the online application, each of the search engines is represented by a plurality of buttons having a sequence, wherein the order of the plurality of buttons is consistent with the order of the respective search engines, and the multiple Switch between buttons to show the search results of each search engine after rearrangement.
  • the sorting device further includes: a weighting module 390, configured to perform a weighting process on the search results of the respective search engines.
  • the weighting module 390 is adapted to perform a weighting process on the search results obtained by the multiple search engine result obtaining module 310.
  • the search results obtained by each search engine in response to the search request are respectively obtained, and the search results of the search engines are respectively performed according to the user behavior weights of the respective search results.
  • Rearrange and sort each search engine to display the sorting results of each search engine and the search results of each search engine after rearranging in the form of an online application of the tab.
  • the device first performs weight processing on the search results, and then rearranges the search results according to the user behavior weights and the search engine relevance weights, and sorts the search engines, and utilizes the big data about the user behaviors in the entire network. More accurate sorting of search results and sorting of search engines is superior to independent search engine resource shortage, big data missing and sorting, which optimizes the ranking of each search engine in multiple search engines and optimizes engine-to-engine sorting.
  • modules in the devices of the embodiments can be adaptively changed and placed in one or more devices different from the embodiment.
  • the modules or units or components of the embodiments may be combined into one module or unit or component, and further they may be divided into a plurality of sub-modules or sub-units or sub-components.
  • any combination of the features disclosed in the specification, including the accompanying claims, the abstract and the drawings, and any methods so disclosed, or All processes or units of the device are combined.
  • Each feature disclosed in this specification (including the accompanying claims, the abstract and the drawings) may be replaced by alternative features that provide the same, equivalent or similar purpose.
  • the various component embodiments of the present invention may be implemented in hardware, or in a software module running on one or more processors, or in a combination thereof.
  • Those skilled in the art will appreciate that some or all of some or all of the components of the ranking device for multiple search engine search results in accordance with embodiments of the present invention may be implemented in practice using a microprocessor or digital signal processor (DSP).
  • DSP digital signal processor
  • the invention can also be implemented as a device or device program (e.g., a computer program and a computer program product) for performing some or all of the methods described herein.
  • Such a program implementing the invention may be stored on a computer readable medium or may be in the form of one or more signals. Such signals can be downloaded from the Internet website, or Provided on the body signal, or in any other form.
  • Figure 4 illustrates a computing device that can implement ranking of multiple search engine search results in accordance with the present invention.
  • the computing device conventionally includes a processor 410 and a computer program product or computer readable medium in the form of a memory 420.
  • the memory 420 may be an electronic memory such as a flash memory, an EEPROM (Electrically Erasable Programmable Read Only Memory), an EPROM, a hard disk, or a ROM.
  • Memory 420 has a memory space 430 for program code 431 for performing any of the method steps described above.
  • storage space 430 for program code may include various program code 431 for implementing various steps in the above methods, respectively.
  • the program code can be read from or written to one or more computer program products.
  • These computer program products include program code carriers such as hard disks, compact disks (CDs), memory cards or floppy disks. Such computer program products are typically portable or fixed storage units as described with reference to FIG.
  • the storage unit may have storage segments, storage spaces, and the like that are similarly arranged to memory 420 in the computing device of FIG.
  • the program code can be compressed, for example, in an appropriate form.
  • the storage unit includes computer readable code 431', ie, code readable by a processor, such as 410, that when executed by a computing device causes the computing device to perform each of the methods described above step.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

La présente invention concerne un procédé et un dispositif pour classer des résultats de recherche de multiples moteurs de recherche. Le procédé consiste à : recevoir une demande de recherche d'un utilisateur et acquérir respectivement des résultats de recherche obtenus par divers moteurs de recherche en réponse à la demande de recherche ; selon des pondérations des divers résultats de recherche en fonction du comportement de l'utilisateur, reclasser respectivement les résultats de recherche des divers moteurs de recherche et classer les divers moteurs de recherche ; et afficher les résultats de classement des divers moteurs de recherche et les résultats de recherche reclassés des divers moteurs de recherche à la manière d'une application en ligne d'une page à onglets. Le procédé utilise des pondérations en fonction du comportement de l'utilisateur pour reclasser respectivement des résultats de recherche de divers moteurs de recherche, et classe et optimise les divers moteurs de recherche.
PCT/CN2015/076111 2014-04-08 2015-04-08 Procédé et dispositif pour classer des résultats de recherche de multiples moteurs de recherche WO2015154679A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410139320.5A CN103870607A (zh) 2014-04-08 2014-04-08 多搜索引擎搜索结果的排序方法及装置
CN201410139320.5 2014-04-08

Publications (1)

Publication Number Publication Date
WO2015154679A1 true WO2015154679A1 (fr) 2015-10-15

Family

ID=50909137

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/076111 WO2015154679A1 (fr) 2014-04-08 2015-04-08 Procédé et dispositif pour classer des résultats de recherche de multiples moteurs de recherche

Country Status (2)

Country Link
CN (1) CN103870607A (fr)
WO (1) WO2015154679A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022231901A1 (fr) * 2021-04-27 2022-11-03 Elasticsearch B.V. Systèmes et procédés de conservation automatique de réponses d'interrogation

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103870607A (zh) * 2014-04-08 2014-06-18 北京奇虎科技有限公司 多搜索引擎搜索结果的排序方法及装置
CN105589895B (zh) * 2014-11-13 2020-04-24 深圳市腾讯计算机系统有限公司 资源排行数据生成方法和装置
EP3026584A1 (fr) 2014-11-25 2016-06-01 Samsung Electronics Co., Ltd. Dispositif et procédé de fourniture de ressources multimédia
CN104462576B (zh) * 2014-12-29 2019-02-26 北京奇虎科技有限公司 基于标签页提供音乐综合搜索的方法和装置
CN107704467B (zh) * 2016-08-09 2021-08-24 百度在线网络技术（北京）有限公司 搜索质量评估方法及装置
CN106649554A (zh) * 2016-11-08 2017-05-10 北京奇虎科技有限公司 一种应用程序的搜索方法和装置、搜索服务器及系统
CN110544475B (zh) * 2019-07-08 2022-03-11 三星电子（中国）研发中心 一种多语音助手的实现方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040215607A1 (en) * 2003-04-25 2004-10-28 Travis Robert L. Method and system fo blending search engine results from disparate sources into one search result
CN102004782A (zh) * 2010-11-25 2011-04-06 北京搜狗科技发展有限公司 一种搜索结果排序方法和搜索结果排序器
CN102081668A (zh) * 2011-01-24 2011-06-01 熊晶 基于领域本体的信息检索优化方法
CN103870607A (zh) * 2014-04-08 2014-06-18 北京奇虎科技有限公司 多搜索引擎搜索结果的排序方法及装置

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100504866C (zh) * 2006-06-30 2009-06-24 腾讯科技（深圳）有限公司 一种综合搜索结果的排序系统及方法
CN100565516C (zh) * 2006-12-08 2009-12-02 腾讯科技（深圳）有限公司 一种移动终端进行网络搜索的方法及系统
CN100507920C (zh) * 2007-05-25 2009-07-01 清华大学 一种基于用户行为信息的搜索引擎检索结果重排序方法
US9495460B2 (en) * 2009-05-27 2016-11-15 Microsoft Technology Licensing, Llc Merging search results
CN102651022B (zh) * 2012-03-31 2017-05-10 北京奇虎科技有限公司 一种搜索方法和装置
CN103257985A (zh) * 2012-05-30 2013-08-21 韩俊 同时查询插入显示多个跨域名数据库的装置和方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040215607A1 (en) * 2003-04-25 2004-10-28 Travis Robert L. Method and system fo blending search engine results from disparate sources into one search result
CN102004782A (zh) * 2010-11-25 2011-04-06 北京搜狗科技发展有限公司 一种搜索结果排序方法和搜索结果排序器
CN102081668A (zh) * 2011-01-24 2011-06-01 熊晶 基于领域本体的信息检索优化方法
CN103870607A (zh) * 2014-04-08 2014-06-18 北京奇虎科技有限公司 多搜索引擎搜索结果的排序方法及装置

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022231901A1 (fr) * 2021-04-27 2022-11-03 Elasticsearch B.V. Systèmes et procédés de conservation automatique de réponses d'interrogation
US11899677B2 (en) 2021-04-27 2024-02-13 Elasticsearch B.V. Systems and methods for automatically curating query responses

Also Published As

Publication number Publication date
CN103870607A (zh) 2014-06-18

Similar Documents

Publication Publication Date Title
WO2015154679A1 (fr) Procédé et dispositif pour classer des résultats de recherche de multiples moteurs de recherche
JP7343568B2 (ja) 機械学習のためのハイパーパラメータの識別および適用
US11580168B2 (en) Method and system for providing context based query suggestions
JP5575902B2 (ja) クエリのセマンティックパターンに基づく情報検索
US10902052B2 (en) Search results through image attractiveness
JP5491678B2 (ja) ビデオコンテンツを管理するための方法および装置
US20140365452A1 (en) Discovering Trending Content of a Domain
CN111061954A (zh) 搜索结果排序方法、装置及存储介质
CN107003829B (zh) 各个结果类别的视野内和视野外的与请求相关的结果区域
CN106777201B (zh) 搜索结果页上的推荐数据的排序方法及装置
CN109241360B (zh) 组合字符串的匹配方法及装置和电子设备
CN110363570A (zh) 应用中的类目展示方法、装置、电子设备和存储介质
US20090063464A1 (en) System and method for visualizing and relevance tuning search engine ranking functions
WO2018205391A1 (fr) Procédé, système et appareil d'évaluation de la précision de recherche d'informations, et support de stockage lisible par ordinateur
WO2016033130A1 (fr) Amélioration d'un classificateur de dispositif informatique par l'intermédiaire d'un échantillonnage d'une entrée stratifiée à n dimensions
CN114780368B (zh) 表数据同步方法与装置
JP7213890B2 (ja) 高速化された大規模な類似度計算
CN108170664B (zh) 基于重点关键词的关键词拓展方法和装置
WO2015143911A1 (fr) Procédé et dispositif de distribution sélective de pages web contenant des informations d'opportunité
CN111241821A (zh) 确定用户的行为特征的方法和装置
JP5361090B2 (ja) 話題語獲得装置、方法、及びプログラム
WO2017219317A1 (fr) Procédé et dispositif de poussée d'informations basés sur un contenu de recherche
WO2017107695A1 (fr) Procédé et dispositif de classement d'actualités
CN111078972B (zh) 提问行为数据的获取方法、装置和服务器
CN109934689B (zh) 目标对象排名解释方法、装置、电子设备及可读存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15776009

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15776009

Country of ref document: EP

Kind code of ref document: A1