WO2018161710A1 - 一种基于输入搜索词来推荐搜索词的方法、装置和存储介质 - Google Patents

一种基于输入搜索词来推荐搜索词的方法、装置和存储介质 Download PDF

Info

Publication number
WO2018161710A1
WO2018161710A1 PCT/CN2017/120266 CN2017120266W WO2018161710A1 WO 2018161710 A1 WO2018161710 A1 WO 2018161710A1 CN 2017120266 W CN2017120266 W CN 2017120266W WO 2018161710 A1 WO2018161710 A1 WO 2018161710A1
Authority
WO
WIPO (PCT)
Prior art keywords
application
search
search term
similarity
preset
Prior art date
Application number
PCT/CN2017/120266
Other languages
English (en)
French (fr)
Inventor
潘岸腾
Original Assignee
广州优视网络科技有限公司
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 广州优视网络科技有限公司 filed Critical 广州优视网络科技有限公司
Publication of WO2018161710A1 publication Critical patent/WO2018161710A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/951Indexing; Web crawling techniques
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/955Retrieval from the web using information identifiers, e.g. uniform resource locators [URL]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/22Matching criteria, e.g. proximity measures

Definitions

  • the present application relates to the field of information processing technologies, and in particular, to a method, an apparatus, and a storage medium for recommending search terms based on inputting search terms.
  • Searching for pages in the app store or app market is the most important entry for users to download apps.
  • search-recommended referrals on the search page such as "everyone is still searching”
  • the function or type recommendation function, as shown in FIG. 1, is to provide other related search words as display recommendations according to the search words currently input by the user.
  • search words that are identical or similar to the input search term portion are used as search words guided by the secondary search.
  • the input search term is “Mimi Music”, and the recommended search words are used. It is "miy” music, "music", “miy” reading, etc., but the recommended search words obtained by simple association of words are difficult to match the needs of users.
  • the search term provided by the existing search term recommendation method as a secondary search guide has some shortcomings in conformity with the user's interest, and the intention of the user to perform the secondary search using the recommended search term is not high. Therefore, it is necessary to continuously develop a new search term recommendation method to help users find more interesting applications through the recommended search terms, thereby improving the user experience.
  • the purpose of the present application is a method, apparatus, and storage medium for recommending search terms based on inputting search terms to improve the above problems.
  • the embodiment of the present application provides a method for recommending a search term based on inputting a search term, including:
  • the embodiment of the present application further provides an apparatus for recommending a search term based on inputting a search term, including:
  • a matching degree determining unit configured to determine a matching degree between the search word input by the user and the application
  • a tag set determining unit for searching for a tag set for the most matching application to cooperate as a tag set of the input search term
  • a first similarity determining unit configured to determine a first similarity between the two applications in the preset application library
  • a second similarity determining unit configured to determine the input search term and other search terms in the preset search term set based on the matching degree, the input tag set of the search term, and the first similarity Second similarity between;
  • a recommendation unit configured to select, according to the second similarity, a certain number of search terms from the preset search word set in a preset manner to recommend to the user.
  • the embodiment of the present application further provides a terminal device, including: one or more processors; a memory; one or more applications, wherein the one or more applications are stored in the memory and configured to Executing by the one or more processors, the one or more programs configured to determine a degree of matching between a search term input by a user and an application in a preset application library;
  • the embodiment of the present application further provides a computer readable storage medium carrying one or more computer instruction programs, where the computer instruction program is executed by one or more processors, the one or more processors executing the above method.
  • a method and apparatus for recommending a search term based on an input search term based on a degree of matching between a search term input by the user and an application i in the application library, and an application between the application i and the application j in the application library a similarity degree, a matching degree between the search term k in the search word set and the application j in the application library, a search result input by the user, a label set, and a search mark k in the search word set respectively.
  • Searching for the search term k having the same tag as the search word input by the user, and according to the established associations, the search term input by the user and other search words in the preset search word set can be determined.
  • the similarity between the two, so as to select a plurality of search words with high similarity as the recommended search words, so that the recommended search words are more in line with the user's interest, and the intention of the user to use the recommended search words for secondary search is improved, and the intention is improved.
  • the user experience is improved.
  • FIG. 1 is a screenshot of an example of providing a recommended search term using a "everyone still searching" recommendation function based on a user-entered search term on a search page of an application store according to the prior art;
  • FIG. 2 is a flowchart of a method for recommending a search term based on an input search term according to an embodiment of the present application
  • FIG. 3 is a schematic block diagram of an apparatus for recommending a search term based on an input search term according to an embodiment of the present application
  • FIG. 4 is a schematic structural diagram of a terminal device according to an embodiment of the present application.
  • FIG. 2 is a flowchart of a method for recommending search terms based on input search words according to an embodiment of the present application. As shown in FIG. 2, the method for recommending a search term based on inputting a search term in the embodiment of the present application includes the following steps:
  • S1 Determine the matching degree between the search words input by the user and the application in the preset application library.
  • search terms that may be of interest to the user based on the search terms entered by the user
  • the search terms are downloaded from which apps, the number of users who downloaded each app, and more.
  • an application library is preset in an application store or an application market, and a large number of various developers developed by different third-party application developers are placed in the application library.
  • the method for determining the degree of matching between the search term input by the user and the application in the preset application library is to obtain the number of users who download an application through the input search term and download a plurality of different applications through the input search term.
  • the ratio of the sum of the number of users is the ratio of the number of users who download an application through the input search term to the sum of the number of users who download a plurality of different applications through the input search term, and the calculation method is as follows:
  • P l,i represents the degree of matching between the search term 1 input by the user and the application i in the preset application library
  • D l,i represents the number of users who downloaded the application i through the search term l;
  • n the number of applications in the application library.
  • the number of users is 100, and the number of users who downloaded the application B provided by the application store or the application market is 200 by using the search term "landlord", and the application provided by the application store or the application market is downloaded by the search term "Dodi”.
  • the number of users is 300, according to the method for determining the matching degree between the search words input by the user and the application in the preset application library provided by the embodiment of the present application, it can be known that the search term “landlord” input by the user matches the application A.
  • the search keyword's label collection method is implemented. Since the application provided in the application store or the application market has one or more tags, according to the matching value of the user-entered search word determined by the above calculation and the application in the preset application library, the application with the largest matching value is obtained. A set of tags having as a set of tags for the entered search term.
  • S3 Determine the first similarity between the two applications in the preset application library.
  • the method of calculating the Jaccard similarity coefficient is used to determine the first similarity between the two applications in the preset application library:
  • Sim i,j represents the first similarity between the application i and the application j in the application library
  • n the number of applications in the application library
  • U i represents a collection of users who have installed application i;
  • U j represents the set of users who installed the application j.
  • the ratio of the number of users who simultaneously installed the application i and the application j to the sum of the number of users who installed the application i and the number of users who installed the application j is used as the first similarity.
  • S4 determining a second similarity between the input search term and other search terms in the preset search word set based on the matching degree, the set of the input search word tags, and the first similarity .
  • the input search word and other search words in the preset search word set may be determined.
  • the second similarity between the two is
  • the popular search words described herein may use search terms ranked first in a predetermined period of time, for example, the search volume may be ranked in the top 500, or the first 1,000, or the first 10,000 within one week.
  • the search terms (which can be set according to the actual needs) are grouped together as a set of preset search words.
  • W l,k represents a second similarity between the search term 1 input by the user and the search term k in the preset search term set;
  • n the number of applications in the application library
  • n the number of search words in the preset search word set
  • K l represents a label set of the search word 1 input by the user
  • K k represents a set of tags of the search term k in the preset search word set
  • P l,i represents the degree of matching between the search term 1 input by the user and the application i in the application library
  • P k,j represents the matching degree between the search term k in the preset search word set and the application j in the application library
  • Sim i,j represents the first similarity between the application i and the application j in the application library.
  • the calculation method of the matching degree P k,j of the search word k in the preset search word set and the application j in the application library and the matching degree P l of the search word 1 input by the user and the application i in the application library is the same, that is, the ratio of the number of users who download an application j by the search word k and the total number of users who download a plurality of different applications through the search word k is obtained as the matching degree P k,j ; in the preset search word set
  • the tag set K k of the search word k is the same as the tag set K l of the search word 1 input by the user, and is determined by using the tag set possessed by the application with the highest matching as the tag set of the search term. .
  • the matching degree between the search term 1 based on the user input and the application i in the application library, the first similarity between the application i and the application j in the application library, and the search term in the search word set k After matching the application j in the application library with the search term k in the search word set, respectively, the search term k having the same tag as the search word input by the user is found. Based on the established associations, a second similarity between the input search terms and other search terms in the preset search term set can be determined. In this way, a certain number of search words can be selected as the recommended search words according to the size of the second similarity, as described in the following step.
  • S5 Select a certain number of search words from the preset search word set to recommend to the user according to the second similarity in a preset manner.
  • a second similarity between the search word input by the user and other search words in the preset search word set may be calculated, and the second similarity may be selected from the preset search word set according to the second similarity.
  • a number of search terms are recommended to the user.
  • a certain number of search terms may be selected from the preset search word set in the order of the second similarity value to be recommended to the user, for example, the second similarity value will be corresponding to the order of the second similarity value.
  • the search words are arranged in descending order, and the top ranked search words such as 10 or 20 are selected as recommended search words to recommend to the user.
  • the number of search terms can be set by itself, not limited to 10 or 20 as exemplified here.
  • a threshold may be preset, and a certain number of search terms are randomly selected from a plurality of search words in the preset search word set corresponding to the second similarity greater than or equal to the preset threshold to be recommended to the user.
  • the corresponding plurality of search words may be selected in descending order of the second similarity value among the plurality of second similarities greater than or equal to the preset threshold.
  • the number of selected search words is the same as above, and can be set by itself, and the top ranked or randomly selected search words such as 10 or 20 are used as recommended search words, and of course, more can be selected. Or other number of search terms as recommended search terms and are not limited to 10 or 20 exemplified herein.
  • the method for recommending a search term based on an input search term in the embodiment of the present application is disclosed in the above embodiments, those skilled in the art know that in determining the input search term and the preset search term set.
  • the first application between the application i and the application j in the application library are determined in advance
  • the degree of similarity, the matching degree of the search term k in the search word set and the application j in the application library, the search word input to the user are marked with a label set, and the search word k in the search word set is marked with a label set.
  • the step S3 in the specific embodiment disclosed may also be implemented in the first step, and the original steps S1-S2 are changed to the second step and the third step; although the specific implementation is disclosed.
  • the matching degree between the search term k in the search word set and the application j in the application library and the search word k in the search word set are marked in the step S4, but actually in step S1.
  • Execution Calculate the matching degree between the search term k in the search term set and the application j in the application library, and then mark the search term k in the search word set with a label set. The other sequences are not listed here.
  • a plurality of search words having high similarity with the search words input by the user can be found from among a plurality of search words in the preset search word set as recommendations Searching for words, so that the recommended search terms are more in line with the user's interest, improving the user's intention to use the recommended search terms for secondary search, and improving the user experience.
  • FIG. 3 is a schematic block diagram of an apparatus for recommending a search term based on an input search term according to an embodiment of the present application.
  • the apparatus for recommending a search term based on inputting a search term in the embodiment of the present application includes:
  • a matching degree determining unit configured to determine a matching degree between the search word input by the user and the application
  • a first similarity determining unit configured to determine a first similarity between the two applications in the preset application library
  • a second similarity determining unit configured to determine the input search term and other search terms in the preset search term set based on the matching degree, the input tag set of the search term, and the first similarity Second similarity between;
  • a recommendation unit configured to select, according to the second similarity, a certain number of search terms from the preset search word set in a preset manner to recommend to the user.
  • the matching degree determining unit is configured to calculate, as the matching degree, a ratio of a number of users who download an application by using the input search term and a total number of users who download a plurality of different applications by using the input search term.
  • the calculation method is as follows:
  • P l, i l represents a degree of matching with the search word input by the user application i;
  • D l,i represents the number of users who downloaded the application i through the search term l;
  • n the number of applications in the application library.
  • the first similarity determining unit is configured to determine the first similarity by using a method of calculating a Jaccard similarity coefficient:
  • Sim i,j represents the first similarity between the application i and the application j in the application library
  • n the number of applications in the application library
  • U i represents a collection of users who have installed application i;
  • U j represents the set of users who installed the application j.
  • the second similarity determining unit is configured to determine the second similarity using the following method:
  • W l,k represents a second similarity between the search term 1 input by the user and the search term k in the preset search term set;
  • n the number of applications in the application library
  • n the number of search words in the preset search word set
  • K l represents a label set of the search word 1 input by the user
  • K k represents a set of tags of the search term k in the preset search word set
  • P l,i represents the degree of matching between the search term 1 input by the user and the application i in the application library
  • P k,j represents the matching degree between the search term k in the preset search word set and the application j in the application library
  • Sim i,j represents the first similarity between the application i and the application j in the application library.
  • the second similarity determining unit calculates a matching degree P k,j of the search term k and the application j in the preset search word set and user input
  • the search term l is the same as the matching degree P l,i of the application i, and is a ratio of the number of users who download an application by the search term k to the sum of the number of users who download a plurality of different applications through the search term k.
  • the matching degree; the label set K k of the search word k in the preset search word set is the same as the label set K l of the search word 1 input by the user, and the label set of the application with the highest matching degree is used as the label set.
  • the method of searching for a tag set of words is determined.
  • the recommending unit is configured to select a certain number of search words from the preset search word set to recommend to the user according to the second similarity value from the largest to the smallest; or, from the second similarity greater than or equal to the preset threshold A plurality of search words in the corresponding preset search word set randomly select a certain number of search words to recommend to the user.
  • a method of randomly selecting a certain number of search words from a plurality of search words in a preset search word set corresponding to a second similarity greater than or equal to a preset threshold to recommend to the user In addition to the random selection, in the plurality of second similarities greater than or equal to the preset threshold, the corresponding plurality of search words may be selected in descending order of the second similarity value.
  • the number of search terms is the same as that of the above method embodiment, and may be set by itself, for example, 10 or 20 search words, or more or other number of search words may be selected as recommended search words.
  • a terminal device provided by an embodiment of the present application includes: one or more processors; a memory; one or more applications, wherein the one or more applications are stored in the memory and configured to Executing by the one or more processors, the one or more programs configured to: determine a degree of matching between a search term input by the user and an application in the preset application library;
  • the terminal device includes a processor 310, a memory 320, an internal memory 330, a network interface 340, and a display screen 350 connected by a system bus.
  • the processor 310 is configured to implement a function of recommending a search term based on inputting a search term, and the processor 310 is configured to perform the method of recommending a search term based on the input search term provided by the above embodiment.
  • the processor 310 is configured to determine a matching degree between the search term input by the user and the application in the preset application library; use the tag set of the application with the highest matching degree as the tag set of the input search term; determine the preset application library a first similarity between the two-two applications; determining the input search term and the preset search term set based on the matching degree, the input tag set of the search term, and the first similarity a second similarity between the other search terms; and selecting, by the second similarity, a predetermined number of search terms from the preset search word set in a preset manner to recommend to the user.
  • the memory 320 is a non-volatile storage medium storing an operating system 321, a database 322, and a computer program for implementing the search word method based on the input search term provided by the above embodiments, and a candidate for executing the computer program generation. Intermediate data and result data.
  • Network interface 340 is used to communicate with the server, and network interface 340 includes a radio frequency transceiver.
  • the processor is configured to: in determining a degree of matching between a search term input by the user and an application in the preset application library, the matching degree is a user who downloads an application by using the input search term
  • the ratio of the quantity to the sum of the number of users who download a plurality of different applications through the input search term is calculated as follows:
  • P l,i represents the degree of matching between the search term 1 input by the user and the application i in the preset application library
  • D l,i represents the number of users who downloaded the application i through the search term l;
  • n the number of applications in the application library.
  • the processor is configured to determine the first similarity by using a method for calculating a Jaccard similarity coefficient in the step of determining a first similarity between the two applications in the preset application library. :
  • Sim i,j represents the first similarity between the application i and the application j in the application library
  • n the number of applications in the application library
  • U i represents a collection of users who have installed application i;
  • U j represents the set of users who installed the application j.
  • the processor is configured to determine the input search term and the preset search term set based on the matching degree, the input tag set of the search term, and the first similarity
  • the following method is used to determine the second similarity:
  • W l,k represents a second similarity between the search term 1 input by the user and the search term k in the preset search term set;
  • n the number of applications in the application library
  • n the number of search words in the preset search word set
  • K l represents a label set of the search word 1 input by the user
  • K k represents a set of tags of the search term k in the preset search word set
  • P l,i represents the degree of matching between the search term 1 input by the user and the application i in the application library
  • P k,j represents the matching degree between the search term k in the preset search word set and the application j in the application library
  • Sim i,j represents the first similarity between the application i and the application j in the application library.
  • the calculating method of the matching degree P k,j of the search term k and the application j in the preset search word set includes:
  • the method for determining the tag set K k of the search term k in the preset search word set includes: using the tag set possessed by the application having the highest matching degree with the search term k as the tag set K k of the search term k .
  • Embodiments of the present application provide a computer readable storage medium carrying one or more computer instruction programs thereon, when the computer instruction program is executed by one or more processors, the one or more processors perform the above method.
  • the embodiment of the present application first establishes a quantitative relationship between a search term and an application, and then converts the search term association search problem into an application association application problem according to the relationship, and the application similarity can be calculated by the Jacques formula. To achieve personalized recommendation of search term related search terms.
  • a plurality of search words having high similarity with the search words input by the user can be found from the plurality of search words in the preset search word set as recommendations Searching for words, so that the recommended search terms are more in line with the user's interest, improving the user's intention to use the recommended search terms for secondary search, and improving the user experience.
  • the foregoing program may be stored in a computer readable storage medium, and the program is executed when executed.
  • the foregoing storage medium includes: a mobile storage device, a random access memory (RAM), a read-only memory (ROM), a magnetic disk, or an optical disk.
  • RAM random access memory
  • ROM read-only memory
  • magnetic disk or an optical disk.
  • optical disk A medium that can store program code.
  • the above-described integrated unit of the present application may be stored in a computer readable storage medium if it is implemented in the form of a software function module and sold or used as a stand-alone product.
  • the technical solution of the embodiments of the present application may be embodied in the form of a software product in essence or in the form of a software product, which is stored in a storage medium and includes a plurality of instructions for making
  • a computer device which may be a personal computer, server, or network device, etc.
  • the foregoing storage medium includes various media that can store program codes, such as a mobile storage device, a RAM, a ROM, a magnetic disk, or an optical disk.

Abstract

一种基于输入搜索词来推荐搜索词的方法、装置和存储介质。所述方法包括:确定用户输入的搜索词与预置应用库里的应用的匹配度(S1);将匹配度最高的应用具有的标签集合作为所述输入的搜索词的标签集合(S2);确定预置应用库里的两两应用之间的第一相似度(S3);基于所述匹配度、所述输入的搜索词的标签集合和所述第一相似度来确定所述输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度(S4);基于所述第二相似度按预设方式从预置搜索词集合中选取一定数量的搜索词向用户推荐(S5)。

Description

一种基于输入搜索词来推荐搜索词的方法、装置和存储介质
交互参考
本申请要求以下优先权:2017年03月07日提出的申请号:201710130848.X,名称:“一种基于输入搜索词来推荐搜索词的方法和装置”的中国专利,本申请参考引用了如上所述申请的全部内容。
技术领域
本申请涉及信息处理技术领域,具体而言涉及一种基于输入搜索词来推荐搜索词的方法、装置和存储介质。
背景技术
随着互联网技术和智能移动终端技术的快速发展,很多在计算机终端上实现的功能(例如购物、阅读)也都可以在智能移动终端上实现,例如使用智能手机或平板电脑等。另外,这些功能的实现需要在智能移动终端上安装相应的应用程序(APP)。例如,网上购物,需要安装例如淘宝客户端,听音乐需要安装音乐播放器客户端等。由此,很多软件公司提供了应用商店或应用市场,例如豌豆荚或者PP助手等。用户可以打开应用商店或者应用市场,从而能够快速搜索和下载所需要的各种应用程序,包括影音播放类、系统工具类、通讯社交类、网上购物类、阅读类等,当然还可以下载游戏等休闲娱乐类应用程序。
在应用商店或应用市场中搜索页面是用户下载应用的最重要入口,为了帮助用户发现更多有趣的应用,在搜索页面还增加有二次搜索引导的推荐功能,例如“大家还在搜”推荐功能或类型推荐功能,如图1所示,该推荐功能是根据用户当前输入的搜索词提供其他相关联的搜索词作为展示推荐。
在实践中,目前仅将与输入的搜索词部分相同或相近的搜索词作为二次搜索引导的搜索词,例如参见图1所示,输入的搜索词为“咪咕音乐”,推荐的搜索词为“咪咕”音乐、“音乐”达人、“咪咕”阅读等,但是,仅是通过词语的简单关联得到的推荐搜索词很难与用户的需求匹配。正是如此,现有的搜索词推荐方法提供的作为二次搜索引导的搜索词在符合用户的兴趣方面存在一些不足,导致用户使用推荐的搜索词进行二次搜索的意向不高。因此需要不断开发新的搜索词推荐方法以帮助用户通过推荐的搜索词发现更多有趣的应用,从而提高用户的使用体验感。
发明内容
本申请的目的在于一种基于输入搜索词来推荐搜索词的方法、装置和存储介质,以改善上述的问题。
本申请实施例提供了一种基于输入搜索词来推荐搜索词的方法,包括:
确定用户输入的搜索词与预置应用库里的应用的匹配度;
将匹配度最高的应用具有的标签集合作为所述输入的搜索词的标签集合;
确定预置应用库里的两两应用之间的第一相似度;
基于所述匹配度、所述输入的搜索词的标签集合和所述第一相似度来确定所述输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度;
基于所述第二相似度按预设方式从预置搜索词集合中选取一定数量的搜索词向用户推荐。
本申请实施例还提供了一种基于输入搜索词来推荐搜索词的装置,包括:
匹配度确定单元,用于确定用户输入的搜索词与应用的匹配度;
搜索词的标签集合确定单元,用于将匹配度最高的应用具有的标签集 合作为所述输入的搜索词的标签集合;
第一相似度确定单元,用于确定预置应用库里的两两应用之间的第一相似度;
第二相似度确定单元,用于基于所述匹配度、所述输入的搜索词的标签集合和所述第一相似度来确定所述输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度;
推荐单元,用于基于所述第二相似度按预设方式从预置搜索词集合中选取一定数量的搜索词向用户推荐。
本申请实施例还提供一种终端设备,其包括:一个或多个处理器;存储器;一个或多个应用程序,其中所述一个或多个应用程序被存储在所述存储器中并被配置为由所述一个或多个处理器执行,所述一个或多个程序配置用于确定用户输入的搜索词与预置应用库里的应用的匹配度;
将匹配度最高的应用具有的标签集合作为所述输入的搜索词的标签集合;
确定预置应用库里的两两应用之间的第一相似度;
基于所述匹配度、所述输入的搜索词的标签集合和所述第一相似度来确定所述输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度;以及
基于所述第二相似度按预设方式从预置搜索词集合中选取预设数量的搜索词向用户推荐。
本申请实施例还提供一种计算机可读存储介质,其上承载一个或多个计算机指令程序,所述计算机指令程序被一个或多个处理器执行时,所述一个或多个处理器执行上述方法。
根据本申请实施例的基于输入搜索词来推荐搜索词的方法和装置,基于用户输入的搜索词l与应用库里的应用i的匹配度、应用库里的应用i和应用j之间的第一相似度、搜索词集合里的搜索词k与应用库里的应用j的 匹配度、再给用户输入的搜索词l标上标签集合、分别给搜索词集合里的搜索词k标上标签集合,从中找出具有与用户输入的搜索词l所具有的标签相同的标签的搜索词k,根据建立的这些关联性,可以确定出用户输入的搜索词与预置搜索词集合中的其它搜索词之间的相似度,由此选择相似度高的多个搜索词作为推荐搜索词,从而所推荐的搜索词更符合用户的兴趣,提高了用户使用推荐的搜索词进行二次搜索的意向,提高了用户体验。
附图说明
图1是根据现有技术在应用商店的搜索页面上基于用户输入的搜索词使用“大家还在搜”推荐功能提供推荐搜索词的一个例子的截图;
图2是本申请实施例提供的基于输入搜索词来推荐搜索词的方法的流程图;
图3是本申请实施例提供的基于输入搜索词来推荐搜索词的装置的示意性框图;
图4是本申请实施例提供的一种终端设备的结构示意图。
具体实施方式
下面将结合本申请实施例和附图,对本申请实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅是本申请一部分实施例,而不是全部的实施例。通常在此处附图中描述和示出的本申请实施例的组件可以以各种不同的配置来布置和设计。因此,以下对在附图中提供的本申请的实施例的详细描述并非旨在限制要求保护的本申请的范围,而是仅仅表示本申请的选定实施例。基于本申请的实施例,本领域技术人员在没有做出创造性劳动的前提下所获得的所有其他实施例,都属于本申请保护的范围。
图2是本申请实施例提供的基于输入搜索词来推荐搜索词的方法的流程图。如图2所示,本申请实施例的基于输入搜索词来推荐搜索词的方法包括以下步骤:
S1:确定用户输入的搜索词与预置应用库里的应用的匹配度。
要想实现根据用户输入的搜索词来推荐可能让用户感兴趣的相关联搜索词,首先需要知道通过用户输入的搜索词都可以下载哪些应用,继而确定所输入的搜索词与那些下载的应用之间匹配度的排序。通过用户输入的搜索词都下载了哪些应用是可以通过后台服务器统计出来的。本领域技术人员都知道,可以对这种统计数据限定一个期限,例如统计1周内、2周内、1个月内、3个月内、6个月内、12个月内等通过用户输入的搜索词都下载了哪些应用、下载每种应用的用户数量等信息。另外,本领域技术人员都知道开发应用商店或应用市场的目的和作用,应用商店或应用市场里预置了应用库,该应用库里放置了由不同第三方应用程序开发商开发的大量的各种应用。这样,确定用户输入的搜索词与预置应用库里的应用的匹配度的方法就是获得通过所述输入的搜索词下载一个应用的用户数量与通过所述输入的搜索词下载多个不同应用的用户数量总和的比值,即所述匹配度为通过所述输入的搜索词下载一个应用的用户数量与通过所述输入的搜索词下载多个不同应用的用户数量总和之比,其计算方法如下:
Figure PCTCN2017120266-appb-000001
其中:P l,i表示用户输入的搜索词l与预置应用库里的应用i的匹配度;
D l,i表示通过搜索词l下载了应用i的用户数量;
Figure PCTCN2017120266-appb-000002
表示通过搜索词l下载了多个不同应用j的用户数量总和;
n表示应用库里的应用数量。
本领域技术人员都知道,统计通过搜索词l下载了应用i的用户数量时,需要限定一个时间段才能有统计结果,如上所述,可以统计1周、2周、1个月、3个月、6个月内、12个月等时间段内的用户下载行为。
例如,假设一个用户在应用商店或应用市场中搜索页面上输入了一个 搜索词“斗地主”,在1周的时间内通过搜索词“斗地主”下载了应用商店或应用市场所提供的应用A的用户数量为100,通过搜索词“斗地主”下载了应用商店或应用市场所提供的应用B的用户数量为200,通过搜索词“斗地主”下载了应用商店或应用市场所提供的应用C的用户数量为300,则根据本申请实施例提供的确定用户输入的搜索词与预置应用库里的应用的匹配度的方法可知,该用户输入的搜索词“斗地主”与应用A的匹配度为100/100+200+300=0.167,该用户输入的搜索词“斗地主”与应用B的匹配度为200/100+200+300=0.333,该用户输入的搜索词“斗地主”与应用C的匹配度为300/100+200+300=0.5。
S2:将匹配度最高的应用具有的标签集合作为所述输入的搜索词的标签集合。
在确定了用户输入的搜索词与预置应用库里的应用的匹配度之后,需要确定与用户输入的搜索词匹配的标签集合,这里采用将匹配度最高的应用具有的标签集合作为所述输入的搜索词的标签集合的方法来实现。由于应用商店或应用市场中提供的应用都具有1个或多个标签,根据上面计算确定的用户输入的搜索词与预置应用库里的应用的匹配度值,将匹配度值最大的应用所具有的标签集合作为所述输入的搜索词的标签集合。
S3:确定预置应用库里的两两应用之间的第一相似度。
在本步骤中,使用计算杰卡德相似系数的方法来确定预置应用库里的两两应用之间的第一相似度:
Figure PCTCN2017120266-appb-000003
其中:Sim i,j表示应用库里的应用i和应用j之间的第一相似度;
n表示应用库里的应用数量;
U i表示安装了应用i的用户集合;
U j表示安装了应用j的用户集合。
在这里,使用同时安装有应用i和应用j的用户数量与安装了应用i的用户数量和安装了应用j的用户数量之和的比值作为第一相似度。
S4:基于所述匹配度、所述输入的搜索词的标签集合和所述第一相似度来确定所述输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度。
在经过以上步骤得到了所述匹配度、所述输入的搜索词的标签集合和所述第一相似度后,就可以确定所述输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度。
本领域技术人员都知道在应用商店或应用市场增加有二次搜索引导的推荐功能后,应用商店或应用市场都会预置一些热门的搜索词作为推荐使用。这里所述的热门搜索词可以采用在预定时间段内搜索量排在前面的搜索词,例如可以将在1周内搜索量排名在前500名、或者前1千名、或者前1万名以内的搜索词(可以根据实践需要自行设定取数量)集合在一起作为预置搜索词集合。
由此基于所述匹配度、所述输入的搜索词的标签集合和所述第一相似度,使用下述方法来确定所述第二相似度:
Figure PCTCN2017120266-appb-000004
i=1,2,…,n;j=1,2,…,n;k=1,2,…,m
其中:W l,k表示用户输入的搜索词l与预置搜索词集合里搜索词k之间的第二相似度;
n表示应用库里的应用数量;
m表示表示预置搜索词集合里的搜索词数量;
K l表示用户输入的搜索词l的标签集合;
K k表示预置搜索词集合里搜索词k的标签集合;
P l,i表示用户输入的搜索词l与应用库里的应用i的匹配度;
P k,j表示预置搜索词集合里搜索词k与应用库里的应用j的匹配度;
Sim i,j表示应用库里的应用i和应用j之间的第一相似度。
另外,所述预置搜索词集合里搜索词k与应用库里的应用j的匹配度P k,j的计算方法与用户输入的搜索词l与应用库里的应用i的匹配度P l,i的计算方法相同,即获得通过搜索词k下载一个应用j的用户数量与通过该搜索词k下载多个不同应用的用户数量总和的比值作为匹配度P k,j;预置搜索词集合里搜索词k的标签集合K k与用户输入的搜索词l的标签集合K l的确定方法相同,都是使用将匹配度最高的应用具有的标签集合作为所述搜索词的标签集合的方法来确定。
由该公式可以看出,基于用户输入的搜索词l与应用库里的应用i的匹配度、应用库里的应用i和应用j之间的第一相似度、搜索词集合里的搜索词k与应用库里的应用j的匹配度、再分别给搜索词集合里的搜索词k标上标签集合之后,从中找出具有与用户输入的搜索词l所具有的标签相同的标签的搜索词k,根据建立的这些关联性,可以确定出所述输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度。这样可以根据第二相似度的大小来选取一定数量的搜索词作为推荐的搜索词,如下一步骤所述。
S5:基于所述第二相似度按预设方式从预置搜索词集合中选取一定数量的搜索词向用户推荐。
根据上一步可以计算出用户输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度,基于所述第二相似度按预设方式从预置搜索词集合中选取一定数量的搜索词向用户推荐。优选的,可以按第二相似度值从大到小的顺序从预置搜索词集合中选取一定数量的搜索词向用户推荐,例如按第二相似度值从大到小的顺序将相对应的搜索词做降序排列,选取排名靠前的例如10个或者20个等搜索词作为推荐搜索词向用户推荐。当然,选取搜索词的数量可以自行设定,不限于这里举例的10或20。优选的,还可以预先设置一个阈值,从大于或等于预设阈值的第二相似度所对应的 预置搜索词集合中的多个搜索词中随机选取一定数量的搜索词向用户推荐。当然,除了随机选取之外,也可以在大于或等于预设阈值的多个第二相似度中,按第二相似度值从大到小的顺序来选取相应的多个搜索词。在这个优选实施例中,选取搜索词的数量与上述一样,可以自行设定,选取排名靠前的或者随机选取例如10个或者20个等搜索词作为推荐搜索词,当然还可以选取更多个或其他数量的搜索词作为推荐搜索词而不局限于这里举例的10或20。
另外,尽管上面以具体实施例的方式公开了本申请实施例的基于输入搜索词来推荐搜索词的方法,本领域技术人员都知道,在确定所述输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度的过程中,只要预先确定出用户输入的搜索词l与应用库里的应用i的匹配度、应用库里的应用i和应用j之间的第一相似度、搜索词集合里的搜索词k与应用库里的应用j的匹配度、给用户输入的搜索词l标上标签集合、给搜索词集合里的搜索词k标上标签集合即可,而与得到结果的先后顺序无关,即公开的具体实施例中的步骤S3也可以放在第一步实施,原步骤S1-S2改为第二步和第三步实施;虽然在公开的具体实施例中搜索词集合里的搜索词k与应用库里的应用j的匹配度和给搜索词集合里的搜索词k标上标签集合都是在步骤S4中完成,但实际上也可以在步骤S1中执行计算搜索词集合里的搜索词k与应用库里的应用j的匹配度,然后给搜索词集合里的搜索词k标上标签集合,这里不对其他顺序做列举说明。
根据本申请实施例的基于输入搜索词来推荐搜索词的方法,能够从预置搜索词集合中的多个搜索词中找出与用户输入的搜索词的相似度高的多个搜索词作为推荐搜索词,从而所推荐的搜索词更符合用户的兴趣,提高了用户使用推荐的搜索词进行二次搜索的意向,提高了用户体验。
图3是本申请实施例提供的基于输入搜索词来推荐搜索词的装置的示意性框图。如图3所示,本申请实施例的基于输入搜索词来推荐搜索词的装置包括:
匹配度确定单元,用于确定用户输入的搜索词与应用的匹配度;
搜索词的标签集合确定单元,用于将匹配度最高的应用具有的标签集合作为所述输入的搜索词的标签集合;
第一相似度确定单元,用于确定预置应用库里的两两应用之间的第一相似度;
第二相似度确定单元,用于基于所述匹配度、所述输入的搜索词的标签集合和所述第一相似度来确定所述输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度;
推荐单元,用于基于所述第二相似度按预设方式从预置搜索词集合中选取一定数量的搜索词向用户推荐。
优选的,所述匹配度确定单元用于计算通过所述输入的搜索词下载一个应用的用户数量与通过所述输入的搜索词下载多个不同应用的用户数量总和的比值来作为所述匹配度,其计算方法如下:
Figure PCTCN2017120266-appb-000005
其中:P l,i表示用户输入的搜索词l与应用i的匹配度;
D l,i表示通过搜索词l下载了应用i的用户数量;
Figure PCTCN2017120266-appb-000006
表示通过搜索词l下载了多个不同应用j的用户数量总和;
n表示应用库里的应用数量。
优选的,所述第一相似度确定单元用于使用计算杰卡德相似系数的方法来确定所述第一相似度:
Figure PCTCN2017120266-appb-000007
其中:Sim i,j表示应用库里的应用i和应用j之间的第一相似度;
n表示应用库里的应用数量;
U i表示安装了应用i的用户集合;
U j表示安装了应用j的用户集合。
优选的,所述第二相似度确定单元用于使用下述方法来确定所述第二相似度:
Figure PCTCN2017120266-appb-000008
i=1,2,…,n;j=1,2,…,n;k=1,2,…,m
其中:W l,k表示用户输入的搜索词l与预置搜索词集合里搜索词k之间的第二相似度;
n表示应用库里的应用数量;
m表示表示预置搜索词集合里的搜索词数量;
K l表示用户输入的搜索词l的标签集合;
K k表示预置搜索词集合里搜索词k的标签集合;
P l,i表示用户输入的搜索词l与应用库里的应用i的匹配度;
P k,j表示预置搜索词集合里搜索词k与应用库里的应用j的匹配度;
Sim i,j表示应用库里的应用i和应用j之间的第一相似度。
优选的,所述第二相似度确定单元在确定所述第二相似度的过程中,所述预置搜索词集合里搜索词k与应用j的匹配度P k,j的计算方法与用户输入的搜索词l与应用i的匹配度P l,i的计算方法相同,是将通过搜索词k下载一个应用的用户数量与通过该搜索词k下载多个不同应用的用户数量总和的比值来作为所述匹配度;预置搜索词集合里搜索词k的标签集合K k与用户输入的搜索词l的标签集合K l的确定方法相同,都是使用将匹配度最高的应用具有的标签集合作为所述搜索词的标签集合的方法来确定。
优选的,推荐单元用于按第二相似度值从大到小的顺序从预置搜索词集合中选取一定数量的搜索词向用户推荐;或者,从大于或等于预设阈值的第二相似度所对应的预置搜索词集合中的多个搜索词中随机选取一定数量的搜索词向用户推荐。如对应的方法步骤描述的那样,在使用从大于或 等于预设阈值的第二相似度所对应的预置搜索词集合中的多个搜索词中随机选取一定数量的搜索词向用户推荐的方法时,除了随机选取之外,也可以在大于或等于预设阈值的多个第二相似度中,按第二相似度值从大到小的顺序来选取相应的多个搜索词。选取搜索词的数量与上述方法实施例一样,可以自行设定,例如10个或者20个等搜索词,或者选取更多个或其他数量的搜索词作为推荐搜索词。
本申请实施例提供的一种终端设备,其包括:一个或多个处理器;存储器;一个或多个应用程序,其中所述一个或多个应用程序被存储在所述存储器中并被配置为由所述一个或多个处理器执行,所述一个或多个程序配置用于:确定用户输入的搜索词与预置应用库里的应用的匹配度;
将匹配度最高的应用具有的标签集合作为所述输入的搜索词的标签集合;
确定预置应用库里的两两应用之间的第一相似度;
基于所述匹配度、所述输入的搜索词的标签集合和所述第一相似度来确定所述输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度;以及
基于所述第二相似度按预设方式从预置搜索词集合中选取预设数量的搜索词向用户推荐。
在一实施例中,参见图4所示,终端设备包括通过系统总线连接的处理器310、存储器320、内存储器330、网络接口340和显示屏350。处理器310用于实现基于输入搜索词来推荐搜索词的功能,处理器310被配置为执行上述实施例提供的基于输入搜索词来推荐搜索词方法。处理器310用于确定用户输入的搜索词与预置应用库里的应用的匹配度;将匹配度最高的应用具有的标签集合作为所述输入的搜索词的标签集合;确定预置应用库里的两两应用之间的第一相似度;基于所述匹配度、所述输入的搜索词的标签集合和所述第一相似度来确定所述输入的搜索词与预置搜索词集 合中的其它搜索词之间的第二相似度;以及基于所述第二相似度按预设方式从预置搜索词集合中选取预设数量的搜索词向用户推荐。存储器320是一种非易失性存储介质,存储有操作系统321、数据库322和用于实现上述实施例提供的的基于输入搜索词来推荐搜索词方法的计算机程序,以及执行计算机程序产生的候选中间数据以及结果数据。网络接口340用于与服务器通信,网络接口340包括射频收发器。
在一实施例中,所述处理器用于在确定用户输入的搜索词与预置应用库里的应用的匹配度的步骤中,所述匹配度为通过所述输入的搜索词下载一个应用的用户数量与通过所述输入的搜索词下载多个不同应用的用户数量总和之比,其计算方法如下:
Figure PCTCN2017120266-appb-000009
其中:P l,i表示用户输入的搜索词l与预置应用库里的应用i的匹配度;
D l,i表示通过搜索词l下载了应用i的用户数量;
Figure PCTCN2017120266-appb-000010
表示通过搜索词l下载了多个不同应用j的用户数量总和;
n表示应用库里的应用数量。
在一实施例中,所述处理器用于在确定预置应用库里的两两应用之间的第一相似度的步骤中,使用计算杰卡德相似系数的方法来确定所述第一相似度:
Figure PCTCN2017120266-appb-000011
其中:Sim i,j表示应用库里的应用i和应用j之间的第一相似度;
n表示应用库里的应用数量;
U i表示安装了应用i的用户集合;
U j表示安装了应用j的用户集合。
在一实施例中,所述处理器用于在基于所述匹配度、所述输入的搜索词的标签集合和所述第一相似度来确定所述输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度的步骤中,使用下述方法来确定所述第二相似度:
Figure PCTCN2017120266-appb-000012
i=1,2,…,n;j=1,2,…,n;k=1,2,…,m
其中:W l,k表示用户输入的搜索词l与预置搜索词集合里搜索词k之间的第二相似度;
n表示应用库里的应用数量;
m表示表示预置搜索词集合里的搜索词数量;
K l表示用户输入的搜索词l的标签集合;
K k表示预置搜索词集合里搜索词k的标签集合;
P l,i表示用户输入的搜索词l与应用库里的应用i的匹配度;
P k,j表示预置搜索词集合里搜索词k与应用库里的应用j的匹配度;
Sim i,j表示应用库里的应用i和应用j之间的第一相似度。
在一实施例中,所述处理器用于所述预置搜索词集合里搜索词k与应用j的匹配度P k,j的计算方法包括:
获得通过搜索词k下载一个应用的用户数量与通过该搜索词k下载多个不同应用的用户数量总和的比值;
所述预置搜索词集合里搜索词k的标签集合K k的确定方法包括:将与所述搜索词k匹配度最高的应用具有的标签集合作为所述搜索词k的标签集合K k
本申请实施例提供一种计算机可读存储介质,其上承载一个或多个计算机指令程序,所述计算机指令程序被一个或多个处理器执行时,所述一 个或多个处理器执行上述的方法。
综上,本申请实施例首先建立搜索词与应用的量化关系,然后根据该关系,把搜索词关联搜索的问题,转化为应用关联应用的问题,而应用的相似性可以通过杰卡德公式计算,从而实现搜索词关联搜索词的个性化推荐。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的装置的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再重复描述。
根据本申请实施例的基于输入搜索词来推荐搜索词的装置,能够从预置搜索词集合中的多个搜索词中找出与用户输入的搜索词的相似度高的多个搜索词作为推荐搜索词,从而所推荐的搜索词更符合用户的兴趣,提高了用户使用推荐的搜索词进行二次搜索的意向,提高了用户体验。
本领域普通技术人员可以理解:实现上述方法实施例的全部或部分步骤可以通过程序指令相关的硬件来完成,前述的程序可以存储于一计算机可读取存储介质中,该程序在执行时,执行包括上述任意方法实施例的步骤;而前述的存储介质包括:移动存储设备、随机存取存储器(RAM,Random Access Memory)、只读存储器(ROM,Read-Only Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
或者,本申请上述集成的单元如果以软件功能模块的形式实现并作为独立的产品销售或使用时,也可以存储在一个计算机可读取存储介质中。基于这样的理解,本申请实施例的技术方案本质上或者说对相关技术做出贡献的部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机、服务器、或者网络设备等)执行申请各个实施例所述方法的全部或部分。而前述的存储介质包括:移动存储设备、RAM、ROM、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述,仅为本申请的具体实施方式,但本申请的保护范围并不局限于此,任何熟悉本技术领域的技术人员在本申请揭露的技术范围内,可轻易想到变化或替换,都应涵盖在本申请的保护范围之内。因此,本申请的保护范围应以所述权利要求的保护范围为准。

Claims (16)

  1. 一种基于输入搜索词来推荐搜索词的方法,包括:
    确定用户输入的搜索词与预置应用库里的应用的匹配度;
    将匹配度最高的应用具有的标签集合作为所述输入的搜索词的标签集合;
    确定预置应用库里的两两应用之间的第一相似度;
    基于所述匹配度、所述输入的搜索词的标签集合和所述第一相似度来确定所述输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度;以及
    基于所述第二相似度按预设方式从预置搜索词集合中选取预设数量的搜索词向用户推荐。
  2. 根据权利要求1所述的方法,在确定用户输入的搜索词与预置应用库里的应用的匹配度的步骤中,所述匹配度为通过所述输入的搜索词下载一个应用的用户数量与通过所述输入的搜索词下载多个不同应用的用户数量总和之比,其计算方法如下:
    Figure PCTCN2017120266-appb-100001
    其中:P l,i表示用户输入的搜索词l与预置应用库里的应用i的匹配度;
    D l,i表示通过搜索词l下载了应用i的用户数量;
    Figure PCTCN2017120266-appb-100002
    表示通过搜索词l下载了多个不同应用j的用户数量总和;
    n表示应用库里的应用数量。
  3. 根据权利要求1所述的方法,在确定预置应用库里的两两应用之间的第一相似度的步骤中,使用计算杰卡德相似系数的方法来确定所述第一相似度:
    Figure PCTCN2017120266-appb-100003
    其中:Sim i,j表示应用库里的应用i和应用j之间的第一相似度;
    n表示应用库里的应用数量;
    U i表示安装了应用i的用户集合;
    U j表示安装了应用j的用户集合。
  4. 根据权利要求1所述的方法,在基于所述匹配度、所述输入的搜索词的标签集合和所述第一相似度来确定所述输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度的步骤中,使用下述方法来确定所述第二相似度:
    Figure PCTCN2017120266-appb-100004
    其中:W l,k表示用户输入的搜索词l与预置搜索词集合里搜索词k之间的第二相似度;
    n表示应用库里的应用数量;
    m表示表示预置搜索词集合里的搜索词数量;
    K l表示用户输入的搜索词l的标签集合;
    K k表示预置搜索词集合里搜索词k的标签集合;
    P l,i表示用户输入的搜索词l与应用库里的应用i的匹配度;
    P k,j表示预置搜索词集合里搜索词k与应用库里的应用j的匹配度;
    Sim i,j表示应用库里的应用i和应用j之间的第一相似度。
  5. 根据权利要求4所述的方法,所述预置搜索词集合里搜索词k与应用j的匹配度P k,j的计算方法包括:
    获得通过搜索词k下载一个应用的用户数量与通过该搜索词k下载多个 不同应用的用户数量总和的比值;
    所述预置搜索词集合里搜索词k的标签集合K k的确定方法包括:将与所述搜索词k匹配度最高的应用具有的标签集合作为所述搜索词k的标签集合K k
  6. 一种基于输入搜索词来推荐搜索词的装置,包括:
    匹配度确定单元,用于确定用户输入的搜索词与应用的匹配度;
    搜索词的标签集合确定单元,用于将匹配度最高的应用具有的标签集合作为所述输入的搜索词的标签集合;
    第一相似度确定单元,用于确定预置应用库里的两两应用之间的第一相似度;
    第二相似度确定单元,用于基于所述匹配度、所述输入的搜索词的标签集合和所述第一相似度来确定所述输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度;和
    推荐单元,用于基于所述第二相似度按预设方式从预置搜索词集合中选取一定数量的搜索词向用户推荐。
  7. 根据权利要求6所述的装置,所述匹配度确定单元用于计算通过所述输入的搜索词下载一个应用的用户数量与通过所述输入的搜索词下载多个不同应用的用户数量总和的比值来作为所述匹配度,其计算方法如下:
    Figure PCTCN2017120266-appb-100005
    其中:P l,i表示用户输入的搜索词l与预置应用库里的应用i的匹配度;
    D l,i表示通过搜索词l下载了应用i的用户数量;
    Figure PCTCN2017120266-appb-100006
    表示通过搜索词l下载了多个不同应用j的用户数量总和;
    n表示应用库里的应用数量。
  8. 根据权利要求6所述的装置,所述第一相似度确定单元用于使用计 算杰卡德相似系数的方法来确定所述第一相似度:
    Figure PCTCN2017120266-appb-100007
    其中:Sim i,j表示应用库里的应用i和应用j之间的第一相似度;
    n表示应用库里的应用数量;
    U i表示安装了应用i的用户集合;
    U j表示安装了应用j的用户集合。
  9. 根据权利要求6所述的装置,所述第二相似度确定单元用于使用下述方法来确定所述第二相似度:
    Figure PCTCN2017120266-appb-100008
    其中:W l,k表示用户输入的搜索词l与预置搜索词集合里搜索词k之间的第二相似度;
    n表示应用库里的应用数量;
    m表示表示预置搜索词集合里的搜索词数量;
    K l表示用户输入的搜索词l的标签集合;
    K k表示预置搜索词集合里搜索词k的标签集合;
    P l,i表示用户输入的搜索词l与应用库里的应用i的匹配度;
    P k,j表示预置搜索词集合里搜索词k与应用库里的应用j的匹配度;
    Sim i,j表示应用库里的应用i和应用j之间的第一相似度。
  10. 根据权利要求9所述的装置,所述第二相似度确定单元用于用于将通过搜索词k下载一个应用的用户数量与通过该搜索词k下载多个不同应用的用户数量总和的比值来作为所述预置搜索词集合里搜索词k与应用j的匹配度P k,j;以及用于将与所述搜索词k匹配度最高的应用具有的标签集合 作为所述搜索词k的标签集合K k的方法来确定所述预置搜索词集合里搜索词k的标签集合K k
  11. 一种终端设备,所述终端设备包括:
    一个或多个处理器;
    存储器;
    一个或多个应用程序,其中所述一个或多个应用程序被存储在所述存储器中并被配置为由所述一个或多个处理器执行,所述一个或多个程序配置用于:确定用户输入的搜索词与预置应用库里的应用的匹配度;
    将匹配度最高的应用具有的标签集合作为所述输入的搜索词的标签集合;
    确定预置应用库里的两两应用之间的第一相似度;
    基于所述匹配度、所述输入的搜索词的标签集合和所述第一相似度来确定所述输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度;以及
    基于所述第二相似度按预设方式从预置搜索词集合中选取预设数量的搜索词向用户推荐。
  12. 根据权利要求11所述的终端设备,所述处理器用于在确定用户输入的搜索词与预置应用库里的应用的匹配度的步骤中,所述匹配度为通过所述输入的搜索词下载一个应用的用户数量与通过所述输入的搜索词下载多个不同应用的用户数量总和之比,其计算方法如下:
    Figure PCTCN2017120266-appb-100009
    其中:P l,i表示用户输入的搜索词l与预置应用库里的应用i的匹配度;
    D l,i表示通过搜索词l下载了应用i的用户数量;
    Figure PCTCN2017120266-appb-100010
    表示通过搜索词l下载了多个不同应用j的用户数量总和;
    n表示应用库里的应用数量。
  13. 根据权利要求11所述的终端设备,所述处理器用于在确定预置应用库里的两两应用之间的第一相似度的步骤中,使用计算杰卡德相似系数的方法来确定所述第一相似度:
    Figure PCTCN2017120266-appb-100011
    其中:Sim i,j表示应用库里的应用i和应用j之间的第一相似度;
    n表示应用库里的应用数量;
    U i表示安装了应用i的用户集合;
    U j表示安装了应用j的用户集合。
  14. 根据权利要求11所述的终端设备,所述处理器用于在基于所述匹配度、所述输入的搜索词的标签集合和所述第一相似度来确定所述输入的搜索词与预置搜索词集合中的其它搜索词之间的第二相似度的步骤中,使用下述方法来确定所述第二相似度:
    Figure PCTCN2017120266-appb-100012
    其中:W l,k表示用户输入的搜索词l与预置搜索词集合里搜索词k之间的第二相似度;
    n表示应用库里的应用数量;
    m表示表示预置搜索词集合里的搜索词数量;
    K l表示用户输入的搜索词l的标签集合;
    K k表示预置搜索词集合里搜索词k的标签集合;
    P l,i表示用户输入的搜索词l与应用库里的应用i的匹配度;
    P k,j表示预置搜索词集合里搜索词k与应用库里的应用j的匹配度;
    Sim i,j表示应用库里的应用i和应用j之间的第一相似度。
  15. 根据权利要求14所述的终端设备,所述处理器用于所述预置搜索词集合里搜索词k与应用j的匹配度P k,j的计算方法包括:
    获得通过搜索词k下载一个应用的用户数量与通过该搜索词k下载多个不同应用的用户数量总和的比值;
    所述预置搜索词集合里搜索词k的标签集合K k的确定方法包括:将与所述搜索词k匹配度最高的应用具有的标签集合作为所述搜索词k的标签集合K k
  16. 一种计算机可读存储介质,其上承载一个或多个计算机指令程序,所述计算机指令程序被一个或多个处理器执行时,所述一个或多个处理器执行权利要求1至5任一项所述的方法。
PCT/CN2017/120266 2017-03-07 2017-12-29 一种基于输入搜索词来推荐搜索词的方法、装置和存储介质 WO2018161710A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710130848.XA CN106909688B (zh) 2017-03-07 2017-03-07 一种基于输入搜索词来推荐搜索词的方法和装置
CN201710130848.X 2017-03-07

Publications (1)

Publication Number Publication Date
WO2018161710A1 true WO2018161710A1 (zh) 2018-09-13

Family

ID=59186183

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/120266 WO2018161710A1 (zh) 2017-03-07 2017-12-29 一种基于输入搜索词来推荐搜索词的方法、装置和存储介质

Country Status (2)

Country Link
CN (1) CN106909688B (zh)
WO (1) WO2018161710A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111967518A (zh) * 2020-08-18 2020-11-20 深圳市欢太科技有限公司 应用标注方法、应用标注装置及终端设备
CN113505230A (zh) * 2021-09-10 2021-10-15 明品云(北京)数据科技有限公司 一种承揽服务推荐方法及系统

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106909688B (zh) * 2017-03-07 2020-10-16 阿里巴巴(中国)有限公司 一种基于输入搜索词来推荐搜索词的方法和装置
WO2019041193A1 (zh) * 2017-08-30 2019-03-07 深圳市云中飞网络科技有限公司 应用资源处理方法及相关产品
CN108446378B (zh) * 2018-03-16 2022-04-12 蜜芽宝贝(北京)网络科技有限公司 基于用户搜索的方法、系统及计算机存储介质
CN110609956B (zh) * 2019-09-18 2022-10-11 苏州达家迎信息技术有限公司 一种信息搜索方法、装置、介质及设备

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103714088A (zh) * 2012-10-09 2014-04-09 深圳市世纪光速信息技术有限公司 搜索词获取方法、服务器、搜索词推荐方法及系统
CN105893440A (zh) * 2015-12-15 2016-08-24 乐视网信息技术(北京)股份有限公司 一种关联应用的推荐方法及装置
CN105938494A (zh) * 2016-04-14 2016-09-14 乐视控股(北京)有限公司 一种应用搜索方法和装置
CN106445963A (zh) * 2015-08-10 2017-02-22 北京奇虎科技有限公司 App平台的广告索引关键词自动生成方法和装置
CN106909688A (zh) * 2017-03-07 2017-06-30 广州优视网络科技有限公司 一种基于输入搜索词来推荐搜索词的方法和装置

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101840420B (zh) * 2010-04-02 2011-12-28 清华大学 搜索辅助系统与搜索辅助方法
CN102799594B (zh) * 2011-05-27 2016-04-13 腾讯科技(深圳)有限公司 一种信息推送方法及装置
KR101955463B1 (ko) * 2011-11-29 2019-03-11 에스케이플래닛 주식회사 컨텐츠 분석에 따른 어플리케이션 추천 시스템 및 방법
US20150088921A1 (en) * 2013-09-20 2015-03-26 Ebay Inc. Search guidance
CN103530339A (zh) * 2013-10-08 2014-01-22 北京百度网讯科技有限公司 移动应用信息推送方法和装置
CN104750760B (zh) * 2013-12-31 2018-11-23 中国移动通信集团上海有限公司 一种推荐应用软件的实现方法及装置
CN105095474B (zh) * 2015-08-11 2018-12-14 北京奇虎科技有限公司 建立搜索词与应用数据推荐关系的方法及装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103714088A (zh) * 2012-10-09 2014-04-09 深圳市世纪光速信息技术有限公司 搜索词获取方法、服务器、搜索词推荐方法及系统
CN106445963A (zh) * 2015-08-10 2017-02-22 北京奇虎科技有限公司 App平台的广告索引关键词自动生成方法和装置
CN105893440A (zh) * 2015-12-15 2016-08-24 乐视网信息技术(北京)股份有限公司 一种关联应用的推荐方法及装置
CN105938494A (zh) * 2016-04-14 2016-09-14 乐视控股(北京)有限公司 一种应用搜索方法和装置
CN106909688A (zh) * 2017-03-07 2017-06-30 广州优视网络科技有限公司 一种基于输入搜索词来推荐搜索词的方法和装置

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111967518A (zh) * 2020-08-18 2020-11-20 深圳市欢太科技有限公司 应用标注方法、应用标注装置及终端设备
CN111967518B (zh) * 2020-08-18 2023-10-13 深圳市欢太科技有限公司 应用标注方法、应用标注装置及终端设备
CN113505230A (zh) * 2021-09-10 2021-10-15 明品云(北京)数据科技有限公司 一种承揽服务推荐方法及系统
WO2023035941A1 (zh) * 2021-09-10 2023-03-16 上海明品医学数据科技有限公司 一种承揽服务推荐方法及系统

Also Published As

Publication number Publication date
CN106909688B (zh) 2020-10-16
CN106909688A (zh) 2017-06-30

Similar Documents

Publication Publication Date Title
WO2018161710A1 (zh) 一种基于输入搜索词来推荐搜索词的方法、装置和存储介质
US10261973B2 (en) System and method for causing downloads of applications based on user intents
US9208437B2 (en) Personalized information pushing method and device
WO2018121700A1 (zh) 基于已安装应用来推荐应用信息的方法、装置、终端设备及存储介质
US10210243B2 (en) Method and system for enhanced query term suggestion
US8166062B1 (en) Search-caching and threshold alerting for commerce sites
JP5736469B2 (ja) ユーザ意図の有無に基づく検索キーワードの推薦
US9934293B2 (en) Generating search results
WO2020155877A1 (zh) 信息推荐
WO2015175384A1 (en) Query categorizer
CN111046221A (zh) 歌曲推荐方法、装置、终端设备以及存储介质
WO2016202214A2 (zh) 一种关键词的展示方法和装置
US10963916B2 (en) Systems and methods for assessing advertisement
WO2018137420A1 (zh) 一种信息推荐列表的生成方法及装置
JP2020170538A (ja) 検索データを処理するための方法、装置及びプログラム
WO2020258481A1 (zh) 个性化文本智能推荐方法、装置及计算机可读存储介质
CN112685648A (zh) 一种资源推荐方法、电子设备及计算机可读存储介质
WO2018171288A1 (zh) 给信息流标注标签的方法、装置、终端设备及存储介质
WO2018149244A1 (zh) 一种基于目标应用推荐相关联应用的方法和装置
CN110209687B (zh) 多维度归因的查询方法和装置
CN112328881B (zh) 文章推荐方法、装置、终端设备及存储介质
EP3255566A1 (en) Text retrieval method and apparatus
CN111767457B (zh) 推荐方法及装置
WO2017049760A1 (zh) 一种搜索结果的处理方法、装置、设备及计算机存储介质
WO2018133680A1 (zh) 一种基于用户已安装应用来推荐热词的方法、装置、终端设备及计算机可读存储介质

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: 17899970

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: 17899970

Country of ref document: EP

Kind code of ref document: A1