WO2018032789A1 - 一种计算互联网关键词的检索精度的方法及系统 - Google Patents

一种计算互联网关键词的检索精度的方法及系统 Download PDF

Info

Publication number
WO2018032789A1
WO2018032789A1 PCT/CN2017/080785 CN2017080785W WO2018032789A1 WO 2018032789 A1 WO2018032789 A1 WO 2018032789A1 CN 2017080785 W CN2017080785 W CN 2017080785W WO 2018032789 A1 WO2018032789 A1 WO 2018032789A1
Authority
WO
WIPO (PCT)
Prior art keywords
search result
retrieval
hit
group
level
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2017/080785
Other languages
English (en)
French (fr)
Inventor
白凡
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuhan Douyu Network Technology Co Ltd
Original Assignee
Wuhan Douyu Network Technology Co Ltd
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 Wuhan Douyu Network Technology Co Ltd filed Critical Wuhan Douyu Network Technology Co Ltd
Publication of WO2018032789A1 publication Critical patent/WO2018032789A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation

Definitions

  • the invention relates to the field of searching for Internet keywords, and particularly relates to a method and system for calculating the retrieval precision of Internet keywords.
  • the traditional Internet search platform generally calculates the accuracy of keyword retrieval:
  • the full-text search is performed on the Internet, that is, after the keyword is segmented according to a certain rule, the word segment is matched with the source data, and the score is divided according to the number of occurrences of the word segmentation, the proximity distance of the word segment, the weight, and the like.
  • Obtain a number of search results ie, several links); for example, if the keyword input by the user is "China", a large number of search results similar to "Chinese cuisine” and "Made in China" are returned.
  • the search result after the sorting is clicked and clicked; and if more The user clicks on the search result after sorting, which proves that the accuracy of the search result for the keyword is low; and when such a situation occurs, the above method for calculating the keyword retrieval accuracy may still be the search result.
  • the accuracy is higher, that is, there is a big difference between the accuracy of the actual search results.
  • the technical problem solved by the present invention is: accurately calculating the retrieval precision of the Internet keyword, and the retrieval accuracy score calculated by the present invention is the same as the actual retrieval precision, and provides for the secondary statistical development of the search platform. Very accurate data.
  • the method for calculating the retrieval precision of an Internet keyword includes the following steps:
  • S2 All the search results are sequentially selected in descending order of the number of clicks, and three sets of search result groups are sequentially selected: a level 1 hit search result group, a level 2 hit search result group, and a level 3 hit search result group, and each group of search result groups
  • the number of search results is at least 10; all unselected search results are used as the miss search result group;
  • a 1 is the sum of the number of clicks of each search result in the level 1 hit search result group
  • a 2 is the sum of the click times of each search result in the level 2 hit search result group
  • a 3 is a level 3 hit.
  • a no is the sum of the number of clicks for each search result in the miss search result group
  • P no is a custom variable
  • the calculation formulas of P 1 , P 2 and P 3 are:
  • the system for calculating the retrieval precision of an Internet keyword by the above method provided by the present invention comprises a retrieval result click number acquisition module, a retrieval result group division module and a retrieval precision score calculation module;
  • the search result click number obtaining module is configured to: count the number of clicks of each search result corresponding to the keyword that needs to calculate the retrieval precision;
  • the search result group dividing module is configured to: sequentially select all the search results in the descending order of the number of clicks: a level 1 hit search result group, a level 2 hit search result group, and a level 3 hit search result group, each The number of search results in the group search result group is at least 10; all unselected search results are used as the miss search result group;
  • the retrieval accuracy score calculation module is configured to: calculate a retrieval precision score S of the current keyword, and the calculation formula is:
  • a 1 is the sum of the number of clicks of each search result in the level 1 hit search result group
  • a 2 is the sum of the click times of each search result in the level 2 hit search result group
  • a 3 is a level 3 hit.
  • a no is the sum of the number of clicks for each search result in the miss search result group
  • P no is a custom variable
  • the calculation formulas of P 1 , P 2 and P 3 are:
  • the present invention calculates a keyword based on the number of clicks and hit impact factors of the four sets of search result groups (the first, second, and third hit search result groups of the search result clicks, and the remaining miss search result groups) Search accuracy score:
  • the retrieval accuracy score calculated by the present invention is the same as the actual retrieval precision, so the result of the retrieval accuracy score is very Accurate, provides accurate data for the secondary statistical development of the search platform, and is highly applicable.
  • FIG. 1 is a flowchart of a method for calculating retrieval precision of an Internet keyword according to an embodiment of the present invention.
  • a method for calculating retrieval precision of an Internet keyword in the embodiment of the present invention includes the following steps:
  • S2 All the search results are sequentially selected in descending order of clicks, and three sets of search result groups are sequentially selected, and the number of search results in each set of search result groups is at least 10, and all unselected search results are used as miss search results. group.
  • the search result group with the highest number of clicks among the three groups of search result groups is the level 1 hit search result group, the search result group with medium clicks is the level 2 hit search result group, and the search result group with the lowest click count is the level 3 hit search result group.
  • the search result in the first-level hit search result group is 1 to 10 in descending order of the number of clicks; the search result in the second-level hit search result group is 10 to 20 arranged in descending order of the number of clicks.
  • the search results in the third-level hit search result group are 20 to 30 in descending order of clicks.
  • a 1 is the sum of the number of clicks of each search result in the level 1 hit search result group, and P 1 represents the hit influence factor of a 1 ;
  • a 2 is the sum of the number of clicks of each search result in the level 2 hit search result group, and P 2 represents the hit influence factor of a 2 ;
  • a 3 is the sum of the number of clicks of each search result in the level 3 hit search result group;
  • P 3 represents the hit influence factor of a 3 ;
  • a no is the sum of the number of clicks of each search result in the missed search result group, P no represents the hit influence factor of a no , and P no is a custom variable in the formula;
  • the process of selecting the specific value of P no is: determining whether N*m is greater than a no , and the value range of m is 90% to 99.5% (99.5% in this embodiment), if it is proved that N>>a no , P no is set to 0, otherwise
  • the system for calculating the retrieval precision of an Internet keyword by the above method provided by the present invention comprises a retrieval result click number acquisition module, a retrieval result group division module, a retrieval precision score calculation module and a retrieval precision determination module.
  • the search result click number obtaining module is configured to: count the number of clicks of each search result corresponding to the keyword that needs to calculate the retrieval precision;
  • the search result group dividing module is configured to: sequentially select all the search results in the descending order of the number of clicks: a level 1 hit search result group, a level 2 hit search result group, and a level 3 hit search result group, each The number of search results in the group search result group is at least 10; all unselected search results are used as the miss search result group.
  • the search result in the first-order hit search result group divided by the search result group division module is 1 to 10 in descending order of the number of clicks; the search result in the second-level hit search result group is arranged in descending order of clicks.
  • the search results in the 10 to 20 and 3rd-level hit search result groups are 20 to 30 in descending order of clicks.
  • the retrieval accuracy score calculation module is configured to: calculate a retrieval precision score S of the current keyword, and the calculation formula is:
  • a 1 is the sum of the number of clicks of each search result in the level 1 hit search result group
  • a 2 is the sum of the click times of each search result in the level 2 hit search result group
  • a 3 is a level 3 hit.
  • a no is the sum of the number of clicks for each search result in the miss search result group
  • P no is a custom variable
  • the calculation formulas of P 1 , P 2 and P 3 are:
  • the workflow for selecting the specific value of P no is: determining whether N*m is greater than a no , and the value range of m is 90% to 99.5%. If yes, P no is set to 0, otherwise
  • the retrieval accuracy determining module is configured to: determine whether the retrieval accuracy score S of the keyword is greater than or equal to an average value of the retrieval accuracy scores S of all the keywords, and if yes, determine the accuracy of the retrieval result corresponding to the current keyword to be high, otherwise The accuracy of the search result corresponding to the current keyword is determined to be low.

Landscapes

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

Abstract

一种计算互联网关键词的检索精度的方法及系统，涉及互联网关键词的检索领域。该方法的步骤为：统计需要计算检索精度的关键词对应的每个检索结果的点击次数(S1)；按照点击次数降序的顺序，顺次选取3组检索结果组，将剩余的所有检索结果作为未命中检索结果组(S2)；计算关键词的检索精度评分(S3)。该方法计算得出的检索精度评分与实际检索精度相同，检索精度评分的结果非常准确，为搜索平台的二次统计开发提供了准确的数据，适用性极强。

Description

一种计算互联网关键词的检索精度的方法及系统 技术领域
本发明涉及互联网关键词的检索领域，具体涉及一种计算互联网关键词的检索精度的方法及系统。
背景技术
随着互联网行业的飞速发展，互联网平台之间的竞争日益加剧，如何快速有效地分析预研用户行为已成为所有互联网平台面临的重要议题；在分析预研用户行为的过程中，用户对于互联网搜索平台的搜索行为是一个重要的分析环节。
目前，传统的互联网搜索平台计算关键词检索精度的方法一般为：
1、用户输入关键词后，在互联网中进行全文检索，即对关键词按一定的规则分词后，再将分词与源数据进行匹配，根据出现分词次数、分词邻近距离、权重等数据进行打分，获得若干检索结果(即若干条链接)；例如用户输入的关键词为“中国”，则返回大量类似于“中国美食”、“中国制造”等检索结果。
2、统计用户点击检索结果(即点击根据关键词检索的链接)次数，若指定数量的用户对同一关键词的检索结果的点击次数超过了阈值，则证明对于当前输入的关键词而言，该检索结果的精度较高。
但是，上述计算关键词检索精度的方法存在以下不足：
当用户发现排序靠前的检索结果(例如第一页网页中的检索结果)不是想要的时候，会观看和点击排序靠后的检索结果；而若较多 的用户点击排序靠后的检索结果，则证明对于关键词而言的检索结果的精度较低；而当出现此种情形时，上述计算关键词检索精度的方法得出的结论可能依然是检索结果精度较高，即与实际的检索结果精度有较大的出入。
下面通过1个例子进行说明：
例如用户想得到“中国人”数据，却输入了“中国”，此时若返回大量类似于“中国美食”、“中国制造”等数据的结果，必然不符合“中国人”数据。用户为了得到自己想要结果，则会选择一些排序较后的检索结果；因此对于“中国人”数据而言，关键词“中国”的检索结果的精度较低，进而使得推荐效果较差。但是即使如此，若用户点击排序较后的检索结果的次数较多，上述计算关键词检索精度的方法依然会认为对于“中国人”数据而言，关键词“中国”的检索结果的精度较高。
发明内容
针对现有技术中存在的缺陷，本发明解决的技术问题为：准确计算互联网关键词的检索精度，本发明计算得出的检索精度评分与实际检索精度相同，为搜索平台的二次统计开发提供了非常准确的数据。
为达到以上目的，本发明提供的计算互联网关键词的检索精度的方法，包括以下步骤：
S1：统计需要计算检索精度的关键词对应的每个检索结果的点击次数；
S2：将所有检索结果按照点击次数降序的顺序，顺次选取3组检索结果组：1级命中检索结果组、2级命中检索结果组和3级命中检索结果组，每组检索结果组中的检索结果的数量为至少10条；将未选取的所有检索结果作为未命中检索结果组；
S3：计算当前关键词的检索精度评分S，计算公式为：
S＝P1*a1+P2 2*a2+P3 3*a3+In(Pno)*ano；
上述公式中：a1为1级命中检索结果组中每条检索结果的点击次数之和，a2为2级命中检索结果组中每条检索结果的点击次数之和，a3为3级命中检索结果组中每条检索结果的点击次数之和；ano为未命中检索结果组中每条检索结果的点击次数之和；
Pno为自定义变量，P1、P2和P3的计算公式分别为：
Figure PCTCN2017080785-appb-000001
本发明提供的实现上述方法的计算互联网关键词的检索精度的系统，包括检索结果点击次数获取模块、检索结果组划分模块和检索精度评分计算模块；
检索结果点击次数获取模块用于：统计需要计算检索精度的关键词对应的每个检索结果的点击次数；
检索结果组划分模块用于：将所有检索结果按照点击次数降序的顺序，顺次选取3组检索结果组：1级命中检索结果组、2级命中检索结果组和3级命中检索结果组，每组检索结果组中的检索结果的数量为至少10条；将未选取的所有检索结果作为未命中检索结果组；
检索精度评分计算模块用于：计算当前关键词的检索精度评分S，计算公式为：
S＝P1*a1+P2 2*a2+P3 3*a3+In(Pno)*ano；
上述公式中：a1为1级命中检索结果组中每条检索结果的点击次数之和，a2为2级命中检索结果组中每条检索结果的点击次数之和，a3为3级命中检索结果组中每条检索结果的点击次数之和；ano为未命中检索结果组中每条检索结果的点击次数之和；
Pno为自定义变量，P1、P2和P3的计算公式分别为：
Figure PCTCN2017080785-appb-000002
与现有技术相比，本发明的优点在于：
本发明根据4组检索结果组(检索结果点击次数排序靠前的1级、2级和3级命中检索结果组、以及剩余的未命中检索结果组)的点击次数和命中影响因子计算关键词的检索精度评分：
当未命中检索结果组中检索结果的点击次数较大时，证明有大量用户未在关键词的1级、2级和3级命中检索结果组中找到自己想要的检索结果，即该关键词的检索结果精度较低，此时本发明计算得出的检索精度评分较小。
当未命中检索结果组中检索结果的点击次数较小时，证明有大量用户在关键词的1级、2级和3级命中检索结果组中能够找到自己想要的关键词检索结果，即该关键词的检索结果精度较高，此时本发明计算得出的检索精度评分较大。
有鉴于此，与现有技术中根据关键词的所有检索结果点击次数，确定关键词的检索精度相比，本发明计算得出的检索精度评分与实际检索精度相同，因此检索精度评分的结果非常准确，为搜索平台的二次统计开发提供了准确的数据，适用性极强。
附图说明
图1为本发明实施例中计算互联网关键词的检索精度的方法的流程图。
具体实施方式
以下结合附图及实施例对本发明作进一步详细说明。
参见图1所示，本发明实施例中的计算互联网关键词的检索精度的方法，包括以下步骤：
S1：统计需要计算检索精度的关键词对应的每个检索结果的点击次数。
S2：将所有检索结果按照点击次数降序的顺序，顺次选取3组检索结果组，每组检索结果组中的检索结果的数量为至少10条，将未选取的所有检索结果作为未命中检索结果组。3组检索结果组中点击次数最高的检索结果组为1级命中检索结果组，点击次数中等的检索结果组为2级命中检索结果组，点击次数最低的检索结果组为3级命中检索结果组。其原因在于：检索结果中排序越靠前的数据被操作，说明该检索效果良好，则给予更高的权重；相反，越靠后的数据被操作或者未被操作，说明检索效果极差，则给予相对偏低的权重值。
本实施例中1级命中检索结果组中的检索结果，为按点击次数降序进行排列的1～10名；2级命中检索结果组中的检索结果，为按点击次数降序进行排列的10～20名，3级命中检索结果组中的检索结果，为按点击次数降序进行排列的20～30名。
S3：计算当前关键词的检索精度评分S，计算公式为：
S＝P1*a1+P2 2*a2+P3 3*a3+In(Pno)*ano；
上述公式中：
a1为1级命中检索结果组中每条检索结果的点击次数之和，P1代表a1的命中影响因子；
a2为2级命中检索结果组中每条检索结果的点击次数之和，P2代表a2的命中影响因子；
a3为3级命中检索结果组中每条检索结果的点击次数之和；P3代表a3的命中影响因子；
ano为未命中检索结果组中每条检索结果的点击次数之和，Pno代表ano的命中影响因子，本公式中Pno为自定义变量；
P1、P2和P3的计算公式分别为：
Figure PCTCN2017080785-appb-000003
S4：判断当前关键词的检索精度评分S，是否大于等于所有关键词的检索精度评分S的平均值，若是，则将当前关键词对应的检索结果的精度确定为高，否则将当前关键词对应的检索结果的精度确定为低。
S3中选取Pno的具体值时，可先将
Figure PCTCN2017080785-appb-000004
此时由于N＝a1+a2+a3+ano，因此0≤P1≤1，0≤P2≤1，0≤P3≤1，0≤Pno≤1。对于关键词的检索结果数据而言，其a1的命中影响因子应当远大于其余命中影响因子，且命中影响因子随等级递减。因为0≤Pno≤1，故In(Pno)≤0，所以ano会对检索精度评分S造成扣分影响。而当N>>ano时，即未命中的检索结果占检索结果总量比极少。说明该关键词的检索精度非常高，此时可直接将Pno设置为0，忽略其影响。
因此选取Pno的具体值的流程为：判断N*m是否大于ano，m的取值范围为90％～99.5％(本实施例中为99.5％)，若是证明N>>ano，将Pno设置为0，否则
Figure PCTCN2017080785-appb-000005
本发明提供的实现上述方法的计算互联网关键词的检索精度的系统，包括检索结果点击次数获取模块、检索结果组划分模块、检索精度评分计算模块和检索精度确定模块。
检索结果点击次数获取模块用于：统计需要计算检索精度的关键词对应的每个检索结果的点击次数；
检索结果组划分模块用于：将所有检索结果按照点击次数降序的顺序，顺次选取3组检索结果组：1级命中检索结果组、2级命中检索结果组和3级命中检索结果组，每组检索结果组中的检索结果的数量为至少10条；将未选取的所有检索结果作为未命中检索结果组。
检索结果组划分模块划分的1级命中检索结果组中的检索结果，为按点击次数降序进行排列的1～10名；2级命中检索结果组中的检索结果，为按点击次数降序进行排列的10～20名，3级命中检索结果组中的检索结果，为按点击次数降序进行排列的20～30名。
检索精度评分计算模块用于：计算当前关键词的检索精度评分S，计算公式为：
S＝P1*a1+P2 2*a2+P3 3*a3+In(Pno)*ano；
上述公式中：a1为1级命中检索结果组中每条检索结果的点击次数之和，a2为2级命中检索结果组中每条检索结果的点击次数之和，a3为3级命中检索结果组中每条检索结果的点击次数之和；ano为未命中检索结果组中每条检索结果的点击次数之和；
Pno为自定义变量，P1、P2和P3的计算公式分别为：
Figure PCTCN2017080785-appb-000006
选取Pno的具体值的工作流程为：判断N*m是否大于ano，m的取值范围为90％～99.5％，若是，将Pno设置为0，否则
Figure PCTCN2017080785-appb-000007
检索精度确定模块用于：判断关键词的检索精度评分S，是否大于等于所有关键词的检索精度评分S的平均值，若是，则将当前关键词对应的检索结果的精度确定为高，否则将当前关键词对应的检索结果的精度确定为低。
本发明不局限于上述实施方式，对于本技术领域的普通技术人员 来说，在不脱离本发明原理的前提下，还可以做出若干改进和润饰，这些改进和润饰也视为本发明的保护范围之内。本说明书中未作详细描述的内容属于本领域专业技术人员公知的现有技术。

Claims (8)

  1. 一种计算互联网关键词的检索精度的方法，其特征在于，该方法包括以下步骤：
    S1：统计需要计算检索精度的关键词对应的每个检索结果的点击次数；
    S2：将所有检索结果按照点击次数降序的顺序，顺次选取3组检索结果组：1级命中检索结果组、2级命中检索结果组和3级命中检索结果组，每组检索结果组中的检索结果的数量为至少10条；将未选取的所有检索结果作为未命中检索结果组；
    S3：计算当前关键词的检索精度评分S，计算公式为：
    S＝P1*a1+P2 2*a2+P3 3*a3+In(Pno)*ano；
    上述公式中：a1为1级命中检索结果组中每条检索结果的点击次数之和，a2为2级命中检索结果组中每条检索结果的点击次数之和，a3为3级命中检索结果组中每条检索结果的点击次数之和；ano为未命中检索结果组中每条检索结果的点击次数之和；
    Pno为自定义变量，P1、P2和P3的计算公式分别为：
    Figure PCTCN2017080785-appb-100001
  2. 如权利要求1所述的计算互联网关键词的检索精度的方法，其特征在于：S2中所述1级命中检索结果组中的检索结果，为按点击次数降序进行排列的1～10名；所述2级命中检索结果组中的检索结果，为按点击次数降序进行排列的10～20名，所述3级命中检索结果组中的检索结果，为按点击次数降序进行排列的20～30名。
  3. 如权利要求1所述的计算互联网关键词的检索精度的方法，其特征在于：S3中选取所述Pno的具体值的流程为：判断N*m是否 大于ano，m的取值范围为90％～99.5％，若是，将Pno设置为0，否则
    Figure PCTCN2017080785-appb-100002
  4. 如权利要求1至3任一项所述的计算互联网关键词的检索精度的方法，其特征在于：S3之后还包括以下步骤：S4：判断当前关键词的检索精度评分S，是否大于等于所有关键词的检索精度评分S的平均值，若是，则将当前关键词对应的检索结果的精度确定为高，否则将当前关键词对应的检索结果的精度确定为低。
  5. 一种实现权利要求1至4任一项所述方法的计算互联网关键词的检索精度的系统，其特征在于：该系统包括检索结果点击次数获取模块、检索结果组划分模块和检索精度评分计算模块；
    检索结果点击次数获取模块用于：统计需要计算检索精度的关键词对应的每个检索结果的点击次数；
    检索结果组划分模块用于：将所有检索结果按照点击次数降序的顺序，顺次选取3组检索结果组：1级命中检索结果组、2级命中检索结果组和3级命中检索结果组，每组检索结果组中的检索结果的数量为至少10条；将未选取的所有检索结果作为未命中检索结果组；
    检索精度评分计算模块用于：计算当前关键词的检索精度评分S，计算公式为：
    S＝P1*a1+P2 2*a2+P3 3*a3+In(Pno)*ano；
    上述公式中：a1为1级命中检索结果组中每条检索结果的点击次数之和，a2为2级命中检索结果组中每条检索结果的点击次数之和，a3为3级命中检索结果组中每条检索结果的点击次数之和；ano为未命中检索结果组中每条检索结果的点击次数之和；
    Pno为自定义变量，P1、P2和P3的计算公式分别为：
    Figure PCTCN2017080785-appb-100003
  6. 如权利要求5所述的计算互联网关键词的检索精度的系统，其特征在于：所述检索结果组划分模块划分的1级命中检索结果组中的检索结果，为按点击次数降序进行排列的1～10名；2级命中检索结果组中的检索结果，为按点击次数降序进行排列的10～20名，3级命中检索结果组中的检索结果，为按点击次数降序进行排列的20～30名。
  7. 如权利要求5所述的计算互联网关键词的检索精度的系统，其特征在于：所述检索精度评分计算模块选取Pno的具体值的工作流程为：判断N*m是否大于ano，m的取值范围为90％～99.5％，若是，将Pno设置为0，否则
    Figure PCTCN2017080785-appb-100004
  8. 如权利要求5至7任一项所述的计算互联网关键词的检索精度的系统，其特征在于：该系统还包括检索精度确定模块，其用于：判断关键词的检索精度评分S，是否大于等于所有关键词的检索精度评分S的平均值，若是，则将当前关键词对应的检索结果的精度确定为高，否则将当前关键词对应的检索结果的精度确定为低。
PCT/CN2017/080785 2016-08-16 2017-04-17 一种计算互联网关键词的检索精度的方法及系统 Ceased WO2018032789A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610671286.5A CN106560808A (zh) 2016-08-16 2016-08-16 一种计算互联网关键词的检索精度的方法及系统
CN201610671286.5 2016-08-16

Publications (1)

Publication Number Publication Date
WO2018032789A1 true WO2018032789A1 (zh) 2018-02-22

Family

ID=58485624

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/080785 Ceased WO2018032789A1 (zh) 2016-08-16 2017-04-17 一种计算互联网关键词的检索精度的方法及系统

Country Status (2)

Country Link
CN (1) CN106560808A (zh)
WO (1) WO2018032789A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106560808A (zh) * 2016-08-16 2017-04-12 武汉斗鱼网络科技有限公司 一种计算互联网关键词的检索精度的方法及系统
CN113935316B (zh) * 2020-07-13 2024-12-03 中国电信股份有限公司 数据处理方法及装置、计算机可存储介质
CN114116736A (zh) * 2020-09-01 2022-03-01 上海智臻智能网络科技股份有限公司 知识库更新、更新验证和基于知识库的搜索方法及装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050050045A1 (en) * 2002-08-23 2005-03-03 Hiroshi Taira Program, system and method for analyzing retrieval keyword
CN101609472A (zh) * 2009-08-13 2009-12-23 腾讯科技（深圳）有限公司 一种基于问答平台的关键词评价方法和装置
CN103246740A (zh) * 2013-05-17 2013-08-14 重庆大学 迭代的基于用户点击的优化搜索和满意度提升方法和系统
CN103383685A (zh) * 2012-05-02 2013-11-06 腾讯科技（深圳）有限公司 一种基于用户点击数据的关键词属性量化方法及装置
CN106560808A (zh) * 2016-08-16 2017-04-12 武汉斗鱼网络科技有限公司 一种计算互联网关键词的检索精度的方法及系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050050045A1 (en) * 2002-08-23 2005-03-03 Hiroshi Taira Program, system and method for analyzing retrieval keyword
CN101609472A (zh) * 2009-08-13 2009-12-23 腾讯科技（深圳）有限公司 一种基于问答平台的关键词评价方法和装置
CN103383685A (zh) * 2012-05-02 2013-11-06 腾讯科技（深圳）有限公司 一种基于用户点击数据的关键词属性量化方法及装置
CN103246740A (zh) * 2013-05-17 2013-08-14 重庆大学 迭代的基于用户点击的优化搜索和满意度提升方法和系统
CN106560808A (zh) * 2016-08-16 2017-04-12 武汉斗鱼网络科技有限公司 一种计算互联网关键词的检索精度的方法及系统

Also Published As

Publication number Publication date
CN106560808A (zh) 2017-04-12

Similar Documents

Publication Publication Date Title
CN105653562B (zh) 一种文本内容与查询请求之间相关性的计算方法及装置
JP5450842B2 (ja) 単語情報エントロピの決定
US8463806B2 (en) Methods and systems for creating and using an adaptive thesaurus
CN103136329B (zh) 多个查询修订模型的集成
TWI554895B (zh) Search results sorting methods and systems, search results sorting optimization methods and systems
CN103365910B (zh) 一种信息检索的方法和系统
CN106339404B (zh) 一种搜索词识别方法及装置
CN110347701B (zh) 一种面向实体检索查询的目标类型标识方法
CN113722478B (zh) 多维度特征融合相似事件计算方法、系统及电子设备
CN104699817B (zh) 一种基于改进谱聚类的搜索引擎排序方法与系统
CN108027814A (zh) 停用词识别方法与装置
CN105653553B (zh) 词权重生成方法和装置
CN106095780B (zh) 一种基于位置特征的检索方法
US11650999B2 (en) Database search enhancement and interactive user interface therefor
CN105786810A (zh) 类目映射关系的建立方法与装置
CN111079035A (zh) 基于动态图谱链接分析的领域搜索排序方法
CN119226603B (zh) 一种基于多维度特征的垂直搜索引擎精排方法及系统
CN106560808A (zh) 一种计算互联网关键词的检索精度的方法及系统
CN105488209A (zh) 一种词权重的分析方法及装置
Tschuggnall et al. Reduce & attribute: Two-step authorship attribution for large-scale problems
Siddiqi et al. Keyword and keyphrase extraction from single Hindi document using statistical approach
US20230022476A1 (en) Systems and methods to facilitate prioritization of documents in electronic discovery
CN112507181B (zh) 搜索请求分类方法、装置、电子设备及存储介质
CN112765311A (zh) 一种裁判文书的搜索方法
CN107423298B (zh) 一种搜索方法和装置

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: 17840765

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: 17840765

Country of ref document: EP

Kind code of ref document: A1