WO2017167071A1 - 一种对应用程序进行项目评估的方法及系统 - Google Patents

一种对应用程序进行项目评估的方法及系统 Download PDF

Info

Publication number
WO2017167071A1
WO2017167071A1 PCT/CN2017/077503 CN2017077503W WO2017167071A1 WO 2017167071 A1 WO2017167071 A1 WO 2017167071A1 CN 2017077503 W CN2017077503 W CN 2017077503W WO 2017167071 A1 WO2017167071 A1 WO 2017167071A1
Authority
WO
WIPO (PCT)
Prior art keywords
target application
feature
data
obtaining
credit
Prior art date
Application number
PCT/CN2017/077503
Other languages
English (en)
French (fr)
Inventor
王瑜
杨洋
叶舟
顾海杰
车品觉
Original Assignee
阿里巴巴集团控股有限公司
王瑜
杨洋
叶舟
顾海杰
车品觉
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 阿里巴巴集团控股有限公司, 王瑜, 杨洋, 叶舟, 顾海杰, 车品觉 filed Critical 阿里巴巴集团控股有限公司
Publication of WO2017167071A1 publication Critical patent/WO2017167071A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063Operations research, analysis or management
    • G06Q10/0639Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/01Social networking

Definitions

  • the present application relates to the field of data processing technologies, and in particular, to a method for project evaluation of an application, and a system for project evaluation of an application.
  • the credit assessment system most relevant to the app credit rating system is a credit assessment of people and companies or projects.
  • the credit evaluation system for people is based on the human dimension to build a credit system, and there is no portability for the app.
  • the technical problem to be solved by the embodiments of the present application is to provide a method for project evaluation of an application, which is used for comprehensive evaluation of the app, and obtains a project credit coefficient with more reference value.
  • the embodiment of the present application further provides a system for evaluating an application program to ensure implementation and application of the foregoing method.
  • the embodiment of the present application discloses a method for performing project evaluation on an application, where the method includes:
  • the step of acquiring multiple heterogeneous data comprises:
  • the heterogeneous data is organized into heterogeneous data sets, respectively.
  • the heterogeneous data includes at least: log data, public relationship data, and social network service data;
  • the heterogeneous data set includes at least: a log data set, a public relationship data set, and a social network service data set.
  • the feature information includes at least: an access behavior feature, a public relationship feature, and a social attribute feature;
  • the step of separately obtaining feature information of the target application to be evaluated from the heterogeneous data includes:
  • the step of determining the target application to be evaluated comprises:
  • the N applications ranked first are determined as the target application to be evaluated, where N is a positive integer.
  • the access behavior feature includes at least: a daily average independent visitor quantity of the target application, and/or a number of daily active users, and/or a daily average average usage duration;
  • the public relationship feature includes at least: a number of documents associated with the target application, and/or a number of comments and/or forwardings of the document associated with the target application;
  • the social attribute feature includes at least: the number of times the target application is downloaded, and/or the number of fans in the official microblog of the target application and/or the number of large V fans and/or the number of attention and/or The total number of comments and/or forwardings of the official Weibo, the name of the relevant person of the target application, and/or the number of fans of the relevant person's Weibo and/or the number of fans and/or the number of followers and/or / or all comments and/or forwarding numbers of the relevant person's Weibo.
  • the relevant person includes an investor and/or a developer and/or an operation and promotion promoter.
  • the project credit coefficient includes a desired credit amount
  • the step of acquiring the project credit coefficient of the target application based on the feature information of the target application comprises:
  • the step of creating a knowledge map of the target application comprises:
  • the step of estimating a desired credit amount of the target application based on the feature data table includes:
  • a desired credit amount for the target application is calculated.
  • the item credit coefficient further includes a credit score
  • the method further includes:
  • the embodiment of the present application further provides a system for performing project evaluation on an application, where the system includes:
  • a heterogeneous data acquisition module for acquiring a plurality of heterogeneous data
  • a feature information acquiring module configured to respectively acquire feature information of the target application to be evaluated from the heterogeneous data
  • a project credit coefficient obtaining module configured to acquire a project credit coefficient of the target application based on feature information of the target application.
  • the heterogeneous data acquisition module comprises:
  • a heterogeneous data obtaining submodule configured to respectively obtain corresponding heterogeneous data from a preset plurality of data resource sites
  • the heterogeneous data includes at least: log data, public relationship data, and social network service data;
  • the heterogeneous data set includes at least: a log data set, a public relationship data set, and a social network service data set.
  • the feature information includes at least: an access behavior feature, a public relationship feature, and a social attribute feature;
  • the feature information obtaining module includes:
  • An access feature obtaining submodule configured to acquire, from the log data set, an access behavior feature of the target application
  • a public relationship feature obtaining submodule configured to acquire a public relationship feature of the target application from the public relationship data set
  • a social feature acquisition sub-module configured to acquire a social attribute feature of the target application from the social network service data set.
  • the target application determining submodule comprises:
  • An access obtaining unit configured to acquire, according to the log data set, the number of accesses of each application within a preset time period
  • a sorting unit configured to sort the application based on the number of accesses
  • a determining unit for determining the top N applications to be the target application to be evaluated, where N is a positive integer.
  • the access behavior feature includes at least: a daily average independent visitor quantity of the target application, and/or a number of daily active users, and/or a daily average average usage duration;
  • the public relationship feature includes at least: a number of documents associated with the target application, and/or a number of comments and/or forwardings of the document associated with the target application;
  • the social attribute feature includes at least: the number of times the target application is downloaded, and/or the number of fans in the official microblog of the target application and/or the number of large V fans and/or the number of attention and/or The total number of comments and forwarding numbers of the official Weibo, and/or the name of the relevant person of the target application, and/or the number of fans of the related person's Weibo and/or the number of fans and/or the number of fans. Number and/or number of comments and/or forwarding numbers of the relevant person's Weibo.
  • the relevant person includes an investor and/or a developer and/or an operation and promotion promoter.
  • the item credit coefficient includes a desired credit amount
  • the item credit coefficient obtaining module includes:
  • a knowledge map creation sub-module for creating a knowledge map of the target application
  • a feature merging sub-module configured to combine the knowledge map, the access behavior feature, the public relationship feature, and the social attribute feature into a feature data table of the target application with a single target application as a primary key ;
  • the amount estimation sub-module is configured to estimate an expected credit amount of the target application based on the feature data table.
  • the knowledge map creation submodule comprises:
  • An associated document obtaining unit configured to obtain a document associated with the target application from the public relationship data set
  • a document word segmentation unit configured to perform word segmentation processing on the associated document, obtain valid words of the manuscript, and count word frequency of the valid word;
  • a knowledge determining unit configured to sort the word frequency into a valid word of the top M, as the entity knowledge of the target application, where M is a positive integer;
  • a type identifying unit configured to identify an entity type of the entity knowledge
  • An entity mapping unit configured to generate a mapping relationship between the target application, the entity type, and the entity knowledge
  • a knowledge map construction unit for organizing all mapping relationships into a knowledge map of the target application.
  • the amount estimation sub-module comprises:
  • a case obtaining unit for obtaining credit data of a credit application that has previously obtained an investment as a case set
  • a sample obtaining unit configured to generate a training sample according to the case set and the feature data table
  • a model training unit configured to perform model training on the training samples, and generate a prediction model
  • the credit amount calculation unit is configured to calculate a desired credit amount of the target application for the prediction model.
  • the project credit coefficient further includes a credit score
  • the system further comprises:
  • the credit score obtaining module is configured to perform a logarithm operation and a normalization process on the expected credit amount to obtain a credit score of the target application.
  • the embodiments of the present application include the following advantages:
  • the embodiment of the present application proposes a general project evaluation method for an application, which is not limited to an application in a certain industry, and can realize cross-industry application comparison and improve application process. Comparability of the order.
  • the embodiment of the present application can automatically combine various heterogeneous data, obtain various feature information of the target application to be evaluated, and obtain a project credit coefficient of the target application according to the feature information, and multiple heterogeneous data can improve the data.
  • FIG. 1 is a flow chart showing the steps of a first embodiment of a method for project evaluation of an application according to the present application
  • FIG. 2 is a flow chart of steps of a second embodiment of a method for performing project evaluation on an application according to the present application
  • FIG. 3 is a structural block diagram of an embodiment of a system for project evaluation of an application according to the present application.
  • FIG. 1 a flow chart of a first embodiment of a method for evaluating an application of an application according to the present application is shown.
  • the method may include the following steps:
  • Step 101 Acquire multiple heterogeneous data
  • Step 102 Obtain feature information of the target application to be evaluated from the heterogeneous data, respectively.
  • Step 103 Acquire an item credit coefficient of the target application based on the feature information of the target application.
  • the embodiment of the present application proposes a general project evaluation method for an application, which is not limited to an application in a certain industry, and can realize cross-industry application comparison and improve application comparability.
  • the embodiment of the present application can automatically combine various heterogeneous data, obtain various feature information of the target application to be evaluated, and obtain a project credit coefficient of the target application according to the feature information, and multiple heterogeneous data can improve the data.
  • FIG. 2 a flow chart of the steps of the second embodiment of the method for evaluating an application of the application is shown in the present application.
  • the embodiment of the present application can be applied to a project credit evaluation scenario of an application, such as an investor or a bank. Provides a more reference-oriented project evaluation program for investors or banks.
  • Step 201 Obtain corresponding log data, public relationship data, and social network service data from preset multiple data resource sites, respectively.
  • corresponding log data, public relation (PR) data, and social network service (SNS) data may be obtained from a plurality of preset data resource sites. data.
  • the data resource site may be a relatively independent site, including an industry hotspot website, a development platform or an analysis platform, a social service website, and the like.
  • Corresponding log data, public relations data, and social network service data can be crawled from multiple data resource sites through a web crawler.
  • the log data may be crawled from the database of the developer platform or the data analysis platform by using a web crawler
  • the developer platform or the data analysis platform may include a platform such as a mobile developer service platform, the alliance, and a Chinese website statistical analysis platform, cnzz.
  • Another example is that you can crawl PR data from industry hotspots through web crawlers, such as obtaining PR data from databases of IT industry hotspots such as Tiger Sniff.
  • SNS data can be crawled from a social service website through a web crawler, such as obtaining SNS data from a meager database.
  • Step 202 Organize the log data, public relationship data, and social network service data into corresponding log data sets, public relationship data sets, and social network service data sets, respectively;
  • all the acquired log data can be further organized into a log data set, and all acquired public relation data is obtained.
  • Organized into a public relation data set also known as a PR data set
  • organizes all acquired social network service data into a social network service data set also referred to as an SNS data set.
  • the log data set records information such as user access behavior to the application and operations on the application.
  • the PR data set records industry information related to the application, dynamic information of the application, and the like.
  • the SNS data collection records the social attribute information of the application, including the social attribute information of the developer, founder, and the like of the application.
  • a log database can be created to save the log data set, and a PR database is created to save the PR data set, and an SNS database is created to save the SNS data set.
  • Step 203 determining a target application to be evaluated
  • step 203 may include the following sub-steps:
  • Sub-step S11 acquiring the records in the log data set, each application is in advance Set the number of visits within the time period;
  • Sub-step S12 sorting the application based on the number of accesses
  • Sub-step S13 determining the top N applications to be the target application to be evaluated, where N is a positive integer.
  • the number of accesses of each application in a preset time period (for example, one month) can be counted, and the application is sorted by the number of accesses.
  • Get a more active application with the first N (N is a positive integer), and compose the list of evaluation objects to be evaluated, that is, the target application list.
  • the foregoing method for determining the target application is only one embodiment of the embodiment of the present application, but the embodiment of the present application is not limited thereto, and those skilled in the art may determine that the target application is applicable by other means.
  • the application that the evaluator needs to evaluate the project is used as the target application, and so on.
  • embodiments of the present application are not limited to the evaluation of an application in a certain industry, and can be applied to evaluation of various applications in various industries, realizing vertical application evaluation in the same industry and application evaluation across industries. Comparability of different application evaluations.
  • Step 204 Obtain an access behavior feature of the target application from the log data set.
  • the access behavior feature reflects the feature related to the access, is the performance of the current data of the target application, and can obtain the access behavior feature of the target application from the log data set with the single target application as the primary key.
  • the access behavior feature may include at least information: a daily average visitor (UV) of the target application, and/or a number of daily active users, And/or, average daily usage time, etc.
  • a daily average visitor (UV) of the target application and/or a number of daily active users, And/or, average daily usage time, etc.
  • the attribute information of the target application may be extracted from the log data set, and the attribute information of the target application may include at least: The industry to which the target application belongs, the age of the target application, the average age of the applications within the industry to which it belongs (the ratio of the ages of all apps in the industry to the number of all apps in the industry), and so on.
  • Step 205 Obtain a public relationship feature of the target application from the public relationship data set.
  • the public relations feature of the target application reflects public relations related features such as operational promotion involving the target application, wherein the public relationship of the target application refers to the communication and communication relationship between the target application and the public environment.
  • a document in an industry hotspot is one of the manifestations of an application's public relations.
  • the document associated with the target application may first be extracted from the public relation data set.
  • the ETL Extract-Transform-Load, which is used to describe the process of extracting, transforming, and loading the data from the source to the destination
  • the PR data set can be obtained in the PR data set. Structured information for all documents.
  • the structured information may include, but is not limited to, a title of each document in the PR data set, a title segmentation, an article tag, a digest, and the like.
  • the title of a manuscript is "A genuine road of music”
  • the article label is: A music, music copyright and entrepreneurship
  • the content is the text
  • its structured information is: "A music's genuine road” +Domi/Music/Genuine+A Music/Music Copyright/Entrepreneurship+NULL (because the data obtained only has the body content, and the abstract is empty, so replace it with NULL).
  • the matching factor of the target application may be matched from the structured information, and if the matching is successful, the document is The document associated with the target application, if it matches If it is unsuccessful, the document is a document that is not relevant to the target application.
  • the public relationship characteristics of the target application can be obtained based on the associated document.
  • the public relationship feature may include at least information such as the number of documents associated with the target application, and/or the number of comments and/or forwarding numbers of the document associated with the target application, and the like. .
  • the number of the associated documents can be calculated as one of the public relations characteristics of the target application.
  • the number of comments and the number of forwards of the document associated with the target application may be the number of comments and/or the number of forwards for each associated document obtained from the PR data set.
  • the number of comments and the number of forwardings of the document associated with the target application may also be a summary value obtained by summing the number of comments of all associated documents after obtaining the number of comments and the number of forwards of each associated document, and all the forwarding numbers.
  • the number of comments and the number of forwards of the document associated with the target application may also be the average number of comments obtained by summarizing and averaging the number of comments of all associated documents after obtaining the number of comments and the number of forwards of each associated document.
  • the public relationship feature of the generated music app can be: music app name + 12 (number of manuscripts) + 3327 (number of comments) + 58 ( Forwarding number).
  • Step 206 Obtain a social attribute feature of the target application from the social network service data set.
  • the social properties of the target application are characterized by the target application and related personnel.
  • the social ability of the target application is at least related to the promotion attribute attribute of the target application itself and the social attribute feature of the relevant person of the target application.
  • the promotion attribute feature of the target application itself may be embodied in the official microblog of the target application
  • the social attribute characteristics of the relevant person of the target application may be embodied in the microblog of the related person.
  • the relevant person may include at least one of the following: an investor (when the target application already has an investor), a developer, an operation and maintenance promoter.
  • the promotion attribute feature of the target application itself may include at least: the number of times the target application is downloaded, and/or the number of fans and/or the number of large V fans in the official microblog of the target application. / or attention to the number and / or official Weibo all comments and / or forwarding number.
  • the social attribute feature of the relevant person of the target application may include at least: a related person name of the target application, and/or a number of fans of the related person's Weibo and/or a large number of V fans and/or attention Number and/or number of comments and/or forwardings for the relevant person's Weibo.
  • Step 207 Create a knowledge map of the target application.
  • the Knowledge Knowledge Domain is also known as the scientific knowledge map. It is called the knowledge domain visualization or the knowledge domain mapping map in the library and information community. It is a series of different graphs showing the relationship between knowledge development process and structure. Using visualization technology. Describe knowledge resources and their carriers, and mine, analyze, construct, map, and display knowledge and their interrelationships.
  • the knowledge map is essentially a semantic network. It is a graph-based data structure consisting of a node and an edge. In the knowledge map, each node represents the "entity” that exists in the real world, and each edge is the “relationship” between the entity and the entity.
  • Knowledge maps are the most effective representation of relationships. In a nutshell, a knowledge map is a network of relationships that combines all kinds of different information (Heterogeneous Information). Knowledge The map provides the ability to analyze problems from a "relationship" perspective.
  • step 207 may include the following sub-steps:
  • Sub-step S21 obtaining a document associated with the target application from the public relationship data set
  • the matching factor of the target application may be matched in the structured information of all the documents of the PR data set, and if the matching is successful, the document is associated with the target application. If the match is unsuccessful, the document is a document that is not relevant to the target application.
  • Sub-step S22 performing word segmentation processing on the associated document, obtaining valid words of the document, and counting word frequency of the valid word;
  • a general word segmentation method may be adopted, for example, a word segmentation based on a string matching, an understanding based method, a statistical-based word segmentation method, and a word segmentation result are obtained, and the word segmentation result is obtained.
  • the embodiment of the present application does not limit the specific word segmentation method. .
  • the frequency of occurrence of each valid word of all associated documents can be counted in units of valid words, and the word frequency of each valid word is obtained.
  • Sub-step S23 sorting the word frequency into the valid words of the top M as the entity knowledge of the target application
  • the valid words can be sorted according to the word frequency, and the valid words sorted in the front M (M is a positive integer) are obtained as the entity knowledge of the target application, that is, as the node in the knowledge map.
  • M is a positive integer
  • the physical knowledge of the A music app can be B music, C music, Zhang San, and the like.
  • Sub-step S24 identifying an entity type of the entity knowledge
  • the entity type corresponding to the entity knowledge can be found in the preset dictionary.
  • the preset dictionary may include a list of investors, a list of developers, a list of well-known microblogs, a list of applications in the same industry, and the like.
  • the target application is an A music app and one of the entity knowledge is Zhang San
  • the target application is an A music app and one of the entity knowledge is a B music app
  • the B music can be found in the above-mentioned investor list, developer list, famous microblog V list, and industry application list. App, finally find the B music app in the same industry application list, then get the B music app entity type as a competitor.
  • the preset dictionary is stored in the SNS database.
  • Sub-step S25 generating a mapping relationship between the target application, the entity type, and the entity knowledge
  • a mapping relationship of the three can be generated. For example, referring to the above example, a mapping relationship of "A music app-investor-Zhangsan" is generated. Another example is to generate a mapping relationship of "A music app - competitor - B music app”.
  • Sub-step S26 all mapping relationships are organized into a knowledge map of the target application.
  • mapping relationships After obtaining the mapping relationship of all entity knowledge, all the mapping relationships can be organized. Go to the knowledge map of the target application.
  • mapping relationship obtained subsequently exists in the knowledge map the mapping relationship is abandoned, and if not, the mapping relationship is added to the knowledge map.
  • Step 208 Combine the knowledge map, the access behavior feature, the public relationship feature, and the social attribute feature into a feature data table of the target application by using a single target application as a primary key;
  • the public application characteristics, access behavior characteristics, social attribute characteristics and knowledge maps can be combined and obtained by the target application as the primary key.
  • factors such as attribute information of the target application, structured information of each associated document, and the like, and attribute information of the target application and structure of each associated document may be considered.
  • Information such as information is added to the feature data table.
  • the feature data table of the A music app is: A music + 12 (the number of related documents) + 3327 (the number of comments of related documents) + 58 (the number of related documents to be forwarded) + ...
  • Step 209 Estimate a desired credit amount of the target application based on the feature data table.
  • the expected credit amount (or the expected investment amount) of the target application can be estimated based on the feature data table.
  • step 209 may include the following sub-steps:
  • Sub-step S31 obtaining the credit data of the trusted application that obtained the investment first, as a case set;
  • the SNS set may also include a credited application that has previously obtained an investment.
  • the credit data of the program, the credit data may include the investment amount.
  • the credit data of the previously trusted investment application can be obtained from the SNS collection as a collection of cases.
  • Sub-step S32 generating a training sample according to the case set and the feature data table
  • the wide table of the corresponding format may be extracted from the case set as a feature variable according to the feature data table, and the investment amount of the trusted application is used as the target variable to obtain the training sample.
  • Sub-step S33 performing model training on the training samples to generate a prediction model
  • the regression model can be used to supervise the training samples to generate a prediction model, which is used to predict the expected credit amount of the target application.
  • Sub-step S34 calculating a desired credit amount of the target application for the prediction model.
  • the feature data table of the target application may be input into the prediction model for calculation, and finally, the expected credit amount of the target application is obtained.
  • Step 210 Perform logarithmic operation and normalization processing on the expected credit amount to obtain a credit score of the target application.
  • the logarithm operation of the expected credit amount may be first performed, and the obtained operation result may be normalized to obtain a credit score.
  • the credit score can range from 0-100.
  • the credit rating of the target application may also be obtained according to the expected credit amount or the credit score, and the credit rating corresponding to the expected investment amount may be obtained according to the association relationship between the preset credit rating and the investment amount. For example, if the expected investment amount is 1 million, and the 1 million is within the five-star rating, the target app's credit rating is determined to be five stars.
  • the embodiments of the present application generate and apply the integration and self-learning scores of social attribute characteristics (app and related person's social ability), public relationship characteristics (operation promotion ability), and access behavior characteristics (the current status of the app).
  • a credit evaluation system that is horizontally compared in the vertical direction of the industry and the entire app market, thereby obtaining a more reference project credit coefficient.
  • FIG. 3 there is shown a structural block diagram of a system embodiment for performing project evaluation on an application, which may include the following modules:
  • the heterogeneous data obtaining module 301 is configured to obtain a plurality of heterogeneous data
  • the feature information obtaining module 302 is configured to obtain feature information of the target application to be evaluated from the heterogeneous data, respectively.
  • the project credit coefficient obtaining module 303 is configured to acquire an item credit coefficient of the target application based on the feature information of the target application.
  • the heterogeneous data obtaining module 301 may include the following submodules:
  • a heterogeneous data obtaining submodule configured to respectively obtain corresponding heterogeneous data from a preset plurality of data resource sites
  • the heterogeneous data includes at least: log data, public relationship data, and social network service data; the heterogeneous data set is at least Includes: log data collections, public relations data collections, and social network service data collections.
  • the feature information includes at least: an access behavior feature, a public relationship feature, and a social attribute feature;
  • the feature information obtaining module 302 can include the following sub-modules:
  • An access feature obtaining submodule configured to acquire, from the log data set, an access behavior feature of the target application
  • a public relationship feature obtaining submodule configured to acquire a public relationship feature of the target application from the public relationship data set
  • a social feature acquisition sub-module configured to acquire a social attribute feature of the target application from the social network service data set.
  • the target application determining submodule includes:
  • An access obtaining unit configured to acquire, according to the log data set, the number of accesses of each application within a preset time period
  • a sorting unit configured to sort the application based on the number of accesses
  • a determining unit for determining the top N applications to be the target application to be evaluated, where N is a positive integer.
  • the access behavior feature includes at least: a daily average independent visitor quantity of the target application, and/or a number of daily active users, and/or an average daily average usage. duration;
  • the public relationship feature includes at least: a number of documents associated with the target application, and/or a number of comments and/or forwardings of the document associated with the target application;
  • the social attribute feature includes at least: the number of times the target application is downloaded, and/or the number of fans in the official microblog of the target application and/or the number of large V fans and/or the number of attention and/or The total number of comments and/or forwardings of the official Weibo, and/or the name of the relevant person of the target application, and/or the number of fans of the relevant person's Weibo and/or the number of fans of the large V and/or Or attention to the number and/or the number of comments and/or forwarding numbers of the relevant person's Weibo.
  • the related person includes an investor and/or a developer and/or an operation and promotion promoter.
  • the item credit coefficient includes a desired credit amount
  • the item credit coefficient obtaining module 303 may include the following sub-modules:
  • a knowledge map creation sub-module for creating a knowledge map of the target application
  • a feature merging sub-module configured to combine the knowledge map, the access behavior feature, the public relationship feature, and the social attribute feature into a feature data table of the target application with a single target application as a primary key ;
  • the amount estimation sub-module is configured to estimate an expected credit amount of the target application based on the feature data table.
  • the knowledge map creation submodule includes:
  • An associated document obtaining unit configured to obtain a document associated with the target application from the public relationship data set
  • a document word segmentation unit configured to perform word segmentation processing on the associated document, obtain valid words of the manuscript, and count word frequency of the valid word;
  • a knowledge determining unit configured to sort the word frequency into a valid word of the top M, as the entity knowledge of the target application, where M is a positive integer;
  • a type identifying unit configured to identify an entity type of the entity knowledge
  • An entity mapping unit configured to generate the target application, the entity type, and Describe the mapping relationship of entity knowledge
  • a knowledge map construction unit for organizing all mapping relationships into a knowledge map of the target application.
  • the amount estimation sub-module includes:
  • a case obtaining unit for obtaining credit data of a credit application that has previously obtained an investment as a case set
  • a sample obtaining unit configured to generate a training sample according to the case set and the feature data table
  • a model training unit configured to perform model training on the training samples, and generate a prediction model
  • the credit amount calculation unit is configured to calculate a desired credit amount of the target application for the prediction model.
  • the item credit coefficient further includes a credit score
  • the system further includes:
  • the credit score obtaining module is configured to perform a logarithm operation and a normalization process on the expected credit amount to obtain a credit score of the target application.
  • embodiments of the embodiments of the present application can be provided as a method, apparatus, or computer program product. Therefore, the embodiment of the present application can adopt complete hardware implementation. A form, an entirely software embodiment, or a combination of software and hardware aspects. Moreover, embodiments of the present application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.
  • computer-usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.
  • Embodiments of the present application are described with reference to flowcharts and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the present application. It will be understood that each flow and/or block of the flowchart illustrations and/or FIG.
  • These computer program operating instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing terminal device to produce a machine for execution by a processor of a computer or other programmable data processing terminal device
  • the operational instructions generate means for implementing the functions specified in one or more of the flow or in a block or blocks of the flowchart.
  • the computer program operating instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that operational instructions stored in the computer readable memory produce manufacturing including the operational command device
  • the operation instruction means implements the functions specified in one block or a plurality of blocks of a flow or a flow and/or a block diagram of the flowchart.
  • These computer program operating instructions can also be loaded onto a computer or other programmable data processing terminal device such that a series of operational steps are performed on the computer or other programmable terminal device to produce computer-implemented processing, such that the computer or other programmable terminal
  • the operational instructions executed on the device provide steps for implementing the functions specified in one or more blocks of the flowchart or in a flow or block of the flowchart.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Educational Administration (AREA)
  • Marketing (AREA)
  • Development Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Computing Systems (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)

Abstract

一种对应用程序进行项目评估的方法及系统,其中所述方法包括:获取多种异构数据(101);分别从所述异构数据中获取待评估的目标应用程序的特征信息(102);基于所述目标应用程序的特征信息,获取所述目标应用程序的项目信用系数(103)。所述方法及系统可以提高应用程序的项目评估的准确率,使得项目信用系数的参考意义更大。

Description

一种对应用程序进行项目评估的方法及系统 技术领域
本申请涉及数据处理技术领域,特别是涉及一种对应用程序进行项目评估的方法,以及一种对应用程序进行项目评估的系统。
背景技术
随着信息技术的发展,应用程序app的数量呈指数增长,仅中国就已有将近百万的量级。然而,这些app中,只有很少一部分可以获得投资,大部分有潜力的app因无法及时获得投资而得不到发展。因此,迫切需要建立一种app信用评估体系。
目前,与app信用评估体系最为相关的信用评估体系是对人和公司或项目的信用评估。
然而,针对人的信用评估体系,是以人为维度进行信用体系的构建,对于app并无移植性。
针对公司或项目的信用评估,往往都是以公司或项目等实体为维度进行评估,在授信时,对每个实体进行实地考察和精细估值。然而app市场由于其庞大的app数量,对每个app进行考察和精细估值并不可取。
因此,目前需要本领域技术人员迫切解决的一个技术问题就是:构建一种适用于app评估的信用评估体系,用以对app进行全面的评估,得到更具参考价值的项目信用系数。
发明内容
本申请实施例所要解决的技术问题是提供一种对应用程序进行项目评估的方法,用以对app进行全面的评估,得到更具参考价值的项目信用系数。
相应的,本申请实施例还提供了一种对应用程序进行项目评估的系统,用以保证上述方法的实现及应用。
为了解决上述问题,本申请实施例公开了一种对应用程序进行项目评估的方法,所述方法包括:
获取多种异构数据;
分别从所述异构数据中获取待评估的目标应用程序的特征信息;
基于所述目标应用程序的特征信息,获取所述目标应用程序的项目信用系数。
优选地,所述获取多种异构数据的步骤包括:
分别从预设的多个数据资源站点中获取对应的异构数据;
分别将所述异构数据组织成异构数据集合。
优选地,所述异构数据至少包括:日志数据、公共关系数据以及社交网络服务数据;所述异构数据集合至少包括:日志数据集合、公共关系数据集合以及社交网络服务数据集合。
优选地,所述特征信息至少包括:访问行为特征、公共关系特征、社交属性特征;
所述分别从所述异构数据中获取待评估的目标应用程序的特征信息的步骤包括:
确定待评估的目标应用程序;
从所述日志数据集合中获取所述目标应用程序的访问行为特征;
从所述公共关系数据集合中获取所述目标应用程序的公共关系特征;
从所述社交网络服务数据集合中获取所述目标应用程序的社交属 性特征。
优选地,所述确定待评估的目标应用程序的步骤包括:
获取所述日志数据集合中记录的,每个应用程序在预设时间段内的访问次数;
基于所述访问次数,对所述应用程序进行排序;
将排序在前的N个应用程序确定为待评估的目标应用程序,其中,N为正整数。
优选地,所述访问行为特征至少包括:所述目标应用程序的日均独立访客量,和/或,日活跃用户数,和/或,日均平均使用时长;
和/或,
所述公共关系特征至少包括:与所述目标应用程序关联的文稿的数量,和/或,所述与所述目标应用程序关联的文稿的评论数和/或转发数;
和/或,
所述社交属性特征至少包括:所述目标应用程序被下载的次数,和/或,所述目标应用程序的官方微博中的粉丝数和/或大V粉丝数和/或关注数和/或官方微博的所有评论数和/或转发数,所述目标应用程序的相关人员名称,和/或,所述相关人员的微博的粉丝数和/或大V粉丝数和/或关注数和/或所述相关人员的微博的所有评论数和/或转发数。
优选地,所述相关人员包括投资者和/或开发者和/或运维推广者。
优选地,所述项目信用系数包括期望授信金额,所述基于所述目标应用程序的特征信息,获取所述目标应用程序的项目信用系数的步骤包括:
创建所述目标应用程序的知识图谱;
以单个目标应用程序为主键,将所述知识图谱、所述访问行为特征、所述公共关系特征、所述社交属性特征,拼合成所述目标应用程序的特 征数据表;
基于所述特征数据表,预估所述目标应用程序的期望授信金额。
优选地,所述创建所述目标应用程序的知识图谱的步骤包括:
从所述公共关系数据集合中获得与所述目标应用程序相关联的文稿;
对所述相关联的文稿进行分词处理,获得所述文稿的有效词,并统计所述有效词的词频;
将词频排序在前M的有效词,作为所述目标应用程序的实体知识,其中,M为正整数;
识别所述实体知识的实体类型;
生成所述目标应用程序、所述实体类型以及所述实体知识的映射关系;
将所有映射关系组织成所述目标应用程序的知识图谱。
优选地,所述基于所述特征数据表,预估所述目标应用程序的期望授信金额的步骤包括:
获取在先获得投资的已授信应用程序的授信数据,作为案例集合;
根据所述案例集合以及所述特征数据表,生成训练样本;
对所述训练样本进行模型训练,生成预测模型;
针对所述预测模型,计算所述目标应用程序的期望授信金额。
优选地,所述项目信用系数还包括信用评分,所述方法还包括:
对所述期望授信金额进行对数运算以及归一化处理,得到所述目标应用程序的信用评分。
本申请实施例还提供了一种对应用程序进行项目评估的系统,所述系统包括:
异构数据获取模块,用于获取多种异构数据;
特征信息获取模块,用于分别从所述异构数据中获取待评估的目标应用程序的特征信息;
项目信用系数获取模块,用于基于所述目标应用程序的特征信息,获取所述目标应用程序的项目信用系数。
优选地,所述异构数据获取模块包括:
异构数据获取子模块,用于分别从预设的多个数据资源站点中获取对应的异构数据;
组织子模块,用于分别将所述异构数据组织成异构数据集合。
优选地,所述异构数据至少包括:日志数据、公共关系数据以及社交网络服务数据;所述异构数据集合至少包括:日志数据集合、公共关系数据集合以及社交网络服务数据集合。
优选地,所述特征信息至少包括:访问行为特征、公共关系特征、社交属性特征;
所述特征信息获取模块包括:
目标应用程序确定子模块,用于确定待评估的目标应用程序;
访问特征获取子模块,用于从所述日志数据集合中获取所述目标应用程序的访问行为特征;
公共关系特征获取子模块,用于从所述公共关系数据集合中获取所述目标应用程序的公共关系特征;
社交特征获取子模块,用于从所述社交网络服务数据集合中获取所述目标应用程序的社交属性特征。
优选地,所述目标应用程序确定子模块包括:
访问次数获取单元,用于获取所述日志数据集合中记录的,每个应用程序在预设时间段内的访问次数;
排序单元,用于基于所述访问次数,对所述应用程序进行排序;
确定单元,用于将排序在前的N个应用程序确定为待评估的目标应用程序,其中,N为正整数。
优选地,所述访问行为特征至少包括:所述目标应用程序的日均独立访客量,和/或,日活跃用户数,和/或,日均平均使用时长;
和/或,
所述公共关系特征至少包括:与所述目标应用程序关联的文稿的数量,和/或,所述与所述目标应用程序关联的文稿的评论数和/或转发数;
和/或,
所述社交属性特征至少包括:所述目标应用程序被下载的次数,和/或,所述目标应用程序的官方微博中的粉丝数和/或大V粉丝数和/或关注数和/或官方微博的所有评论数以及转发数,和/或,所述目标应用程序的相关人员名称,和/或,所述相关人员的微博的粉丝数和/或大V粉丝数和/或关注数和/或所述相关人员的微博的所有评论数和/或转发数。
优选地,所述相关人员包括投资者和/或开发者和/或运维推广者。
优选地,所述项目信用系数包括期望授信金额,所述项目信用系数获取模块包括:
知识图谱创建子模块,用于创建所述目标应用程序的知识图谱;
特征合并子模块,用于以单个目标应用程序为主键,将所述知识图谱、所述访问行为特征、所述公共关系特征、所述社交属性特征,拼合成所述目标应用程序的特征数据表;
金额预估子模块,用于基于所述特征数据表,预估所述目标应用程序的期望授信金额。
优选地,所述知识图谱创建子模块包括:
关联文稿获取单元,用于从所述公共关系数据集合中获得与所述目标应用程序相关联的文稿;
文稿分词单元,用于对所述相关联的文稿进行分词处理,获得所述文稿的有效词,并统计所述有效词的词频;
知识确定单元,用于将词频排序在前M的有效词,作为所述目标应用程序的实体知识,其中,M为正整数;
类型识别单元,用于识别所述实体知识的实体类型;
实体映射单元,用于生成所述目标应用程序、所述实体类型以及所述实体知识的映射关系;
知识图谱构建单元,用于将所有映射关系组织成所述目标应用程序的知识图谱。
优选地,所述金额预估子模块包括:
案例获取单元,用于获取在先获得投资的已授信应用程序的授信数据,作为案例集合;
样本获取单元,用于根据所述案例集合以及所述特征数据表,生成训练样本;
模型训练单元,用于对所述训练样本进行模型训练,生成预测模型;
授信金额计算单元,用于针对所述预测模型,计算所述目标应用程序的期望授信金额。
优选地,所述项目信用系数还包括信用评分,所述系统还包括:
信用评分获取模块,用于对所述期望授信金额进行对数运算以及归一化处理,得到所述目标应用程序的信用评分。
与背景技术相比,本申请实施例包括以下优点:
本申请实施例提出一种通用的对应用程序的项目评估方式,并不限于某一行业的应用程序,可以实现跨行业应用程序的对比,提高应用程 序的可比性。
另外,本申请实施例能够自动结合多种异构数据,获取待评估的目标应用程序的各项特征信息,并根据特征信息来获取目标应用程序的项目信用系数,多种异构数据可以提高数据来源的全面性,从而使得项目信用系数真实反映目标应用程序的特征,提高项目评估的准确率,使得项目信用系数的参考意义更大。
附图说明
图1是本申请的一种对应用程序进行项目评估的方法实施例一的步骤流程图;
图2是本申请的一种对应用程序进行项目评估的方法实施例二的步骤流程图;
图3是本申请的一种对应用程序进行项目评估的系统实施例的结构框图。
具体实施方式
为使本申请的上述目的、特征和优点能够更加明显易懂,下面结合附图和具体实施方式对本申请作进一步详细的说明。
参照图1,示出了本申请的一种对应用程序进行项目评估的方法实施例一的步骤流程图,所述方法可以包括如下步骤:
步骤101,获取多种异构数据;
步骤102,分别从所述异构数据中获取待评估的目标应用程序的特征信息;
步骤103,基于所述目标应用程序的特征信息,获取所述目标应用程序的项目信用系数。
本申请实施例提出一种通用的对应用程序的项目评估方式,并不限于某一行业的应用程序,可以实现跨行业应用程序的对比,提高应用程序的可比性。
另外,本申请实施例能够自动结合多种异构数据,获取待评估的目标应用程序的各项特征信息,并根据特征信息来获取目标应用程序的项目信用系数,多种异构数据可以提高数据来源的全面性,从而使得项目信用系数真实反映目标应用程序的特征,提高项目评估的准确率,使得项目信用系数的参考意义更大。
参照图2,示出了本申请的一种对应用程序进行项目评估的方法实施例二的步骤流程图,本申请实施例可以应用在投资方或银行等对应用程序的项目信用评估场景中,为投资方或银行等提供更具参考价值的针对应用程序的项目评估方案。
本申请实施例可以包括如下步骤:
步骤201,分别从预设的多个数据资源站点中获取对应的日志数据、公共关系数据以及社交网络服务数据;
在本申请实施例中,可以从预设的多个数据资源站点中获取对应的日志数据、公共关系(Public Relation,简称PR)数据以及社交网络服务(Social Networking Services,简称SNS)数据等异构数据。
在具体实现中,数据资源站点可以为相对独立的站点,包括行业热点网站、开发平台或分析平台、社交服务网站等。
可以通过网络爬虫从多个数据资源站点中爬取对应的日志数据、公共关系数据以及社交网络服务数据。
例如,可以通过网络爬虫从开发者平台或数据分析平台的数据库中爬取日志数据,该开发者平台或数据分析平台可以包括如移动开发者服务平台友盟、中文网站统计分析平台cnzz等平台。
又如,可以通过网络爬虫从行业热点网站中爬取PR数据,如从虎嗅网等IT行业热点网站的数据库中获取PR数据。
又如,可以通过网络爬虫从社交服务网站中爬取SNS数据,如从微薄的数据库中获取SNS数据。
步骤202,分别将所述日志数据、公共关系数据以及社交网络服务数据组织成对应的日志数据集合、公共关系数据集合以及社交网络服务数据集合;
从多个数据资源站点中获取对应的日志数据、公共关系数据以及社交网络服务数据等数据以后,进一步可以将所有获取到的日志数据组织成日志数据集合,以及,将所有获取到的公共关系数据组织成公共关系数据集合(又称PR数据集合),以及,将所有获取到的社交网络服务数据组织成社交网络服务数据集合(又称SNS数据集合)。
在具体实现中,日志数据集合记录了用户对应用程序的访问行为以及在应用程序上的操作等信息。
PR数据集合记录了与应用程序相关的行业信息以及该应用程序的动态信息等。
SNS数据集合记录了应用程序的社交属性信息,包括该应用程序的开发者、创始人等相关人员的社交属性信息。
在实际中,可以创建日志数据库保存该日志数据集合,以及,创建PR数据库保存该PR数据集合,以及,创建SNS数据库保存该SNS数据集合。
步骤203,确定待评估的目标应用程序;
在本申请实施例的一种优选实施例中,步骤203可以包括如下子步骤:
子步骤S11,获取所述日志数据集合中记录的,每个应用程序在预 设时间段内的访问次数;
子步骤S12,基于所述访问次数,对所述应用程序进行排序;
子步骤S13,将排序在前的N个应用程序确定为待评估的目标应用程序,其中,N为正整数。
具体而言,根据日志数据集合中记录的对每个应用程序的访问行为,可以统计预设时间段(例如,一个月)内每个应用程序的访问次数,并以访问次数对应用程序进行排序,获得排序在前N(N为正整数)的较为活跃的应用程序,组成待进行项目评估的评估对象列表,即目标应用程序列表。
需要说明的是,上述确定目标应用程序的方式仅仅是本申请实施例的一种实施方式,但本申请实施例并不限于此,本领域技术人员采用其他方式确定目标应用程序均是可以的,例如,根据银行或投资方等评估方的实际需求,将评估方需要进行项目评估的应用程序作为目标应用程序,等等。
另外,本申请实施例并不限于某一行业的应用程序的评估,可以通用于各种行业的各种应用程序的评估,实现同行业纵向的应用程序评估以及跨行业横向的应用程序评估,提高不同应用程序评估的可比性。
步骤204,从所述日志数据集合中获取所述目标应用程序的访问行为特征;
在实际中,该访问行为特征反映了与访问相关的特征,是目标应用程序当前数据的表现,可以以单个目标应用程序为主键,从日志数据集合中获取该目标应用程序的访问行为特征。
作为本申请实施例的一种优选示例,所述访问行为特征至少可以包括如下信息:所述目标应用程序的日均独立访客量(unique visitor,简称uv),和/或,日活跃用户数,和/或,日均平均使用时长等。
在具体实现中,除了可以从日志数据集合中提取目标应用程序的访问行为特征以外,还可以从日志数据集合中提取该目标应用程序的属性信息,该目标应用程序的属性信息至少可以包括:该目标应用程序所属的行业、该目标应用程序的年龄、该所属的行业内应用程序的平均年龄(该行业内所有app的年龄之和与该行业内所有app的数量的比值)等。
步骤205,从所述公共关系数据集合中获取所述目标应用程序的公共关系特征;
目标应用程序的公共关系特征为反映了涉及目标应用程序的运营推广等公共关系相关的特征,其中,目标应用程序的公共关系是指目标应用程序与公众环境之间的沟通与传播关系。例如,行业热点网站中的文稿是应用程序的公共关系的表现形式之一。
在本申请实施例中,首先可以从公共关系数据集合中提取与目标应用程序关联的文稿。在具体实现中,可以通过ETL(Extract-Transform-Load,用来描述将数据从来源端经过抽取(extract)、转换(transform)、加载(load)至目的端的过程)技术获取PR数据集合中的所有文稿的结构化信息。
作为一种示例,该结构化信息可以包括但不限于:PR数据集合中每个文稿的标题、标题分词、文章标签、摘要等。例如,某一文稿的标题为《A音乐的正版之路》,文章标签为:A音乐、音乐版权和创业三个,内容为正文,则其结构化信息为:《A音乐的正版之路》+多米/音乐/正版+A音乐/音乐版权/创业+NULL(因为获取的数据中只有正文内容,而摘要为空,所以用NULL代替)。
得到PR数据集合中的每个文稿的结构化信息以后,可以从该结构化信息中匹配目标应用程序的匹配因子(包括目标应用程序的名称、功能等),若匹配成功,则该文稿为与目标应用程序关联的文稿,若匹配 不成功,则该文稿为与目标应用程序不相关的文稿。
确定与目标应用程序的关联的文稿以后,可以根据该关联的文稿,获取目标应用程序的公共关系特征。
作为本申请实施例的一种优选示例,公共关系特征至少可以包括如下信息:与目标应用程序关联的文稿的数量,和/或,与目标应用程序关联的文稿的评论数和/或转发数等。
具体而言,得到目标应用程序的关联的文稿以后,可以计算该关联的文稿的数量,作为目标应用程序的公共关系特征之一。
目标应用程序关联的文稿的评论数和转发数可以为从PR数据集合中获取的每一个关联的文稿的评论数和/或转发数。
和/或,
目标应用程序关联的文稿的评论数和转发数也可以是在获得每一个关联的文稿的评论数以及转发数以后,将所有关联的文稿的评论数进行汇总得到的汇总值以及将所有的转发数进行汇总得到的汇总值。
和/或,
目标应用程序关联的文稿的评论数和转发数也可以是在获得每一个关联的文稿的评论数以及转发数以后,将所有关联的文稿的评论数进行汇总求平均后得到的平均评论数以及将所有的转发数进行汇总求平均后得到的平均转发数。
如虎嗅网中一个月内有多篇PR稿件是关于某音乐app的,因此生成的该音乐app的公共关系特征可以为:音乐app名称+12(稿件数量)+3327(评论数)+58(转发数)。
步骤206,从所述社交网络服务数据集合中获取所述目标应用程序的社交属性特征;
目标应用程序的社交属性特征为反映了目标应用程序及相关人员 的社交能力与关注程度相关的特征,该目标应用程序的社交属性特征至少可以包括目标应用程序本身的推广属性特征以及目标应用程序的相关人员的社交属性特征。例如,该目标应用程序本身的推广属性特征可以从该目标应用程序的官方微博中体现,该目标应用程序的相关人员的社交属性特征可以从该相关人员的微博中体现。
作为一种示例,该相关人员至少可以包括如下人员的一种:投资者(当目标应用程序已有投资者时)、开发者、运维推广者。
作为一种示例,目标应用程序本身的推广属性特征至少可以包括:该目标应用程序被下载的次数,和/或,该目标应用程序的官方微博中的粉丝数和/或大V粉丝数和/或关注数和/或官方微博的所有评论数和/或转发数等。
该目标应用程序的相关人员的社交属性特征至少可以包括:所述目标应用程序的相关人员名称,和/或,所述相关人员的微博的粉丝数和/或大V粉丝数和/或关注数和/或该相关人员的微博的所有评论数和/或转发数。
步骤207,创建所述目标应用程序的知识图谱;
知识图谱(Mapping Knowledge Domain)也被称为科学知识图谱,在图书情报界称为知识域可视化或知识领域映射地图,是显示知识发展进程与结构关系的一系列各种不同的图形,用可视化技术描述知识资源及其载体,挖掘、分析、构建、绘制和显示知识及它们之间的相互联系。
知识图谱本质上是语义网络,是一种基于图的数据结构,由节点(Point)和边(Edge)组成。在知识图谱里,每个节点表示现实世界中存在的“实体”,每条边为实体与实体之间的“关系”。知识图谱是关系的最有效的表示方式。通俗地讲,知识图谱就是把所有不同种类的信息(Heterogeneous Information)连接在一起而得到的一个关系网络。知识 图谱提供了从“关系”的角度去分析问题的能力。
在本申请实施例的一种优选实施例中,步骤207可以包括如下子步骤:
子步骤S21,从所述公共关系数据集合中获得与所述目标应用程序相关联的文稿;
在具体实现中,可以在PR数据集合的所有文稿的结构化信息中匹配目标应用程序的匹配因子(包括目标应用程序的名称、功能等),若匹配成功,则该文稿为与目标应用程序关联的文稿,若匹配不成功,则该文稿为与目标应用程序不相关的文稿。
子步骤S22,对所述相关联的文稿进行分词处理,获得所述文稿的有效词,并统计所述有效词的词频;
获得与目标应用程序关联的文稿以后,可以对每个关联的文稿的全文进行分词处理,得到分词结果,然后去掉分词结果中的诸如“的”“了”“是”等停用词,得到每一个关联的文稿的有效词。
需要说明的是,可以采用通用的分词方法,例如,基于字符串匹配、基于理解、基于统计等分词方法对关联的文稿进行全文分词,得到分词结果,本申请实施例对具体的分词方式不作限制。
得到每个关联的文稿的有效词以后,可以以有效词为单位,统计所有关联的文稿的每一个有效词的出现的频率,得到每个有效词的词频。
子步骤S23,将词频排序在前M的有效词,作为所述目标应用程序的实体知识;
得到每个有效词的词频以后,可以按照词频对有效词进行排序,并获取排序在前M(M为正整数)的有效词,作为该目标应用程序的实体知识,即作为知识图谱中的节点。例如,A音乐app的实体知识可以为B音乐、C音乐、张三等。
子步骤S24,识别所述实体知识的实体类型;
当确定目标应用程序的实体知识时,可以在预设字典中查找与该实体知识对应的实体类型。在具体实现中,预设字典可以包括投资人列表、开发者列表、著名微博大V列表、同行业应用程序列表等。
例如,若目标应用程序为A音乐app,其中一个实体知识为张三,则可以在上述投资人列表、开发者列表、著名微博大V列表、同行业应用程序列表等字典中查找张三,最后在投资人列表中找到“张三”,则得到“张三”的实体类型为投资人。
又如,若目标应用程序为A音乐app,其中一个实体知识为B音乐app,则可以在上述投资人列表、开发者列表、著名微博大V列表、同行业应用程序列表等字典中查找B音乐app,最后在同行业应用程序列表中找到B音乐app,则得到B音乐app的实体类型为竞争对手。
在实际中,该预设字典存储在SNS数据库中。
需要说明的是,上述识别实体类型的方式仅仅是本申请实施例的一种示例,本领域技术人员采用其他方式均是可以的,本申请实施例对此不作限制。
子步骤S25,生成所述目标应用程序、所述实体类型以及所述实体知识的映射关系;
得到目标应用程序的实体知识以及实体类型以后,可以生成三者的映射关系,例如,参照上例,生成“A音乐app——投资人——张三”的映射关系。又如,生成“A音乐app——竞争对手——B音乐app”的映射关系。
子步骤S26,将所有映射关系组织成所述目标应用程序的知识图谱。
得到所有实体知识的映射关系以后,组织所有的映射关系,可以得 到目标应用程序的知识图谱。
需要说明的是,后续得到的映射关系,若在知识图谱中存在,则放弃该映射关系,若不存在,则在知识图谱中增加该映射关系。
步骤208,以单个目标应用程序为主键,将所述知识图谱、所述访问行为特征、所述公共关系特征、所述社交属性特征,拼合成所述目标应用程序的特征数据表;
获得目标应用程序的公共关系特征、访问行为特征、社交属性特征以及知识图谱以后,可以以目标应用程序为主键,将其公共关系特征、访问行为特征、社交属性特征以及知识图谱进行合并,得到描述目标应用程序多个维度的特征信息的特征数据宽表。
在具体实现中,在生成特征数据表的过程中,还可以考虑目标应用程序的属性信息、每一关联文稿的结构化信息等的因素,将目标应用程序的属性信息、每一关联文稿的结构化信息等信息添加到特征数据表中。
例如,A音乐app的特征数据表为:A音乐+12(相关的文稿的数量)+3327(相关的文稿的评论数)+58(相关的文稿的转发数)+……。
步骤209,基于所述特征数据表,预估所述目标应用程序的期望授信金额;
得到目标应用程序的特征数据表以后,可以根据该特征数据表,预估目标应用程序的期望授信金额(或称期望投资金额)。
在本申请实施例的一种优选实施例中,步骤209可以包括如下子步骤:
子步骤S31,获取在先获得投资的已授信应用程序的授信数据,作为案例集合;
在具体实现中,SNS集合中还可以包括在先获得投资的已授信应用 程序的授信数据,该授信数据可以包括投资金额。
可以从SNS集合中获得在先获得投资的已授信应用程序的授信数据,作为案例集合。
子步骤S32,根据所述案例集合以及所述特征数据表,生成训练样本;
具体来说,可以依据特征数据表从案例集合中提取相对应格式的宽表作为特征变量,将已授信应用程序的投资金额作为目标变量,得到训练样本。
子步骤S33,对所述训练样本进行模型训练,生成预测模型;
得到训练样本以后,可以采用回归模型对训练样本进行监督学习,生成预测模型,该预测模型用于预测目标应用程序的期望授信金额。
子步骤S34,针对所述预测模型,计算所述目标应用程序的期望授信金额。
在具体实现中,可以将目标应用程序的特征数据表输入预测模型进行运算,最后得到该目标应用程序的期望授信金额。
步骤210,对所述期望授信金额进行对数运算以及归一化处理,得到所述目标应用程序的信用评分。
具体而言,可以首先对期望授信金额进行log对数运算,并将得到的运算结果进行归一化运算,得到信用评分。作为一种示例,该信用评分的范围可以为0-100。
在另一种实施方式中,还可以根据期望授信金额或信用评分获得目标应用程序的信用评级,具体可以根据预设的信用评级与投资金额的关联关系,获得与该期望投资金额对应的信用评级,例如,期望投资金额为100万,该100万在五星评级的范围内,则将该目标app的信用评级确定为五星。
本申请实施例通过社交属性特征(app及相关人员的社交能力)、公共关系特征(运营推广能力)以及访问行为特征(app的当前状况)等几个方面的整合和自学习打分,生成可应用于行业纵向以及整个app市场横向比较的信用评估体系,从而获得更具参考意义的项目信用系数。
需要说明的是,对于方法实施例,为了简单描述,故将其都表述为一系列的动作组合,但是本领域技术人员应该知悉,本申请实施例并不受所描述的动作顺序的限制,因为依据本申请实施例,某些步骤可以采用其他顺序或者同时进行。其次,本领域技术人员也应该知悉,说明书中所描述的实施例均属于优选实施例,所涉及的动作并不一定是本申请实施例所必须的。
参照图3,示出了本申请一种对应用程序进行项目评估的系统实施例的结构框图,所述系统可以包括如下模块:
异构数据获取模块301,用于获取多种异构数据;
特征信息获取模块302,用于分别从所述异构数据中获取待评估的目标应用程序的特征信息;
项目信用系数获取模块303,用于基于所述目标应用程序的特征信息,获取所述目标应用程序的项目信用系数。
在本申请实施例的一种优选实施例中,所述异构数据获取模块301可以包括如下子模块:
异构数据获取子模块,用于分别从预设的多个数据资源站点中获取对应的异构数据;
组织子模块,用于分别将所述异构数据组织成异构数据集合。
在本申请实施例的一种优选实施例中,所述异构数据至少包括:日志数据、公共关系数据以及社交网络服务数据;所述异构数据集合至少 包括:日志数据集合、公共关系数据集合以及社交网络服务数据集合。
在本申请实施例的一种优选实施例中,所述特征信息至少包括:访问行为特征、公共关系特征、社交属性特征;
所述特征信息获取模块302可以包括如下子模块:
目标应用程序确定子模块,用于确定待评估的目标应用程序;
访问特征获取子模块,用于从所述日志数据集合中获取所述目标应用程序的访问行为特征;
公共关系特征获取子模块,用于从所述公共关系数据集合中获取所述目标应用程序的公共关系特征;
社交特征获取子模块,用于从所述社交网络服务数据集合中获取所述目标应用程序的社交属性特征。
在本申请实施例的一种优选实施例中,所述目标应用程序确定子模块包括:
访问次数获取单元,用于获取所述日志数据集合中记录的,每个应用程序在预设时间段内的访问次数;
排序单元,用于基于所述访问次数,对所述应用程序进行排序;
确定单元,用于将排序在前的N个应用程序确定为待评估的目标应用程序,其中,N为正整数。
在本申请实施例的一种优选实施例中,所述访问行为特征至少包括:所述目标应用程序的日均独立访客量,和/或,日活跃用户数,和/或,日均平均使用时长;
和/或,
所述公共关系特征至少包括:与所述目标应用程序关联的文稿的数量,和/或,所述与所述目标应用程序关联的文稿的评论数和/或转发数;
和/或,
所述社交属性特征至少包括:所述目标应用程序被下载的次数,和/或,所述目标应用程序的官方微博中的粉丝数和/或大V粉丝数和/或关注数和/或官方微博的所有评论数和/或转发数,和/或,所述目标应用程序的相关人员名称,和/或,所述相关人员的微博的粉丝数和/或大V粉丝数和/或关注数和/或所述相关人员的微博的所有评论数和/或转发数。
在本申请实施例的一种优选实施例中,所述相关人员包括投资者和/或开发者和/或运维推广者。
在本申请实施例的一种优选实施例中,所述项目信用系数包括期望授信金额,所述项目信用系数获取模块303可以包括如下子模块:
知识图谱创建子模块,用于创建所述目标应用程序的知识图谱;
特征合并子模块,用于以单个目标应用程序为主键,将所述知识图谱、所述访问行为特征、所述公共关系特征、所述社交属性特征,拼合成所述目标应用程序的特征数据表;
金额预估子模块,用于基于所述特征数据表,预估所述目标应用程序的期望授信金额。
在本申请实施例的一种优选实施例中,所述知识图谱创建子模块包括:
关联文稿获取单元,用于从所述公共关系数据集合中获得与所述目标应用程序相关联的文稿;
文稿分词单元,用于对所述相关联的文稿进行分词处理,获得所述文稿的有效词,并统计所述有效词的词频;
知识确定单元,用于将词频排序在前M的有效词,作为所述目标应用程序的实体知识,其中,M为正整数;
类型识别单元,用于识别所述实体知识的实体类型;
实体映射单元,用于生成所述目标应用程序、所述实体类型以及所 述实体知识的映射关系;
知识图谱构建单元,用于将所有映射关系组织成所述目标应用程序的知识图谱。
在本申请实施例的一种优选实施例中,所述金额预估子模块包括:
案例获取单元,用于获取在先获得投资的已授信应用程序的授信数据,作为案例集合;
样本获取单元,用于根据所述案例集合以及所述特征数据表,生成训练样本;
模型训练单元,用于对所述训练样本进行模型训练,生成预测模型;
授信金额计算单元,用于针对所述预测模型,计算所述目标应用程序的期望授信金额。
在本申请实施例的一种优选实施例中,所述项目信用系数还包括信用评分,所述系统还包括:
信用评分获取模块,用于对所述期望授信金额进行对数运算以及归一化处理,得到所述目标应用程序的信用评分。
对于系统实施例而言,由于其与上述方法实施例基本相似,所以描述的比较简单,相关之处参见方法实施例的部分说明即可。
本说明书中的各个实施例均采用递进的方式描述,每个实施例重点说明的都是与其他实施例的不同之处,各个实施例之间相同相似的部分互相参见即可。
本领域内的技术人员应明白,本申请实施例的实施例可提供为方法、装置、或计算机程序产品。因此,本申请实施例可采用完全硬件实 施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且,本申请实施例可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本申请实施例是参照根据本申请实施例的方法、终端设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序操作指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序操作指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理终端设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理终端设备的处理器执行的操作指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序操作指令也可存储在能引导计算机或其他可编程数据处理终端设备以特定方式工作的计算机可读存储器中,使得存储在该计算机可读存储器中的操作指令产生包括操作指令装置的制造品,该操作指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序操作指令也可装载到计算机或其他可编程数据处理终端设备上,使得在计算机或其他可编程终端设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程终端设备上执行的操作指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
尽管已描述了本申请实施例的优选实施例,但本领域内的技术人员一旦得知了基本创造性概念,则可对这些实施例做出另外的变更和修改。所以,所附权利要求意欲解释为包括优选实施例以及落入本申请实施例范围的所有变更和修改。
最后,还需要说明的是,在本文中,诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来,而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者终端设备不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者终端设备所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括所述要素的过程、方法、物品或者终端设备中还存在另外的相同要素。
以上对本申请所提供的一种对应用程序进行项目评估的方法及系统进行了详细介绍,本文中应用了具体个例对本申请的原理及实施方式进行了阐述,以上实施例的说明只是用于帮助理解本申请的方法及其核心思想;同时,对于本领域的一般技术人员,依据本申请的思想,在具 体实施方式及应用范围上均会有改变之处,综上所述,本说明书内容不应理解为对本申请的限制。

Claims (22)

  1. 一种对应用程序进行项目评估的方法,其特征在于,所述方法包括:
    获取多种异构数据;
    分别从所述异构数据中获取待评估的目标应用程序的特征信息;
    基于所述目标应用程序的特征信息,获取所述目标应用程序的项目信用系数。
  2. 根据权利要求1所述的方法,其特征在于,所述获取多种异构数据的步骤包括:
    分别从预设的多个数据资源站点中获取对应的异构数据;
    分别将所述异构数据组织成异构数据集合。
  3. 根据权利要求2所述的方法,其特征在于,所述异构数据至少包括:日志数据、公共关系数据以及社交网络服务数据;所述异构数据集合至少包括:日志数据集合、公共关系数据集合以及社交网络服务数据集合。
  4. 根据权利要求3所述的方法,其特征在于,所述特征信息至少包括:访问行为特征、公共关系特征、社交属性特征;
    所述分别从所述异构数据中获取待评估的目标应用程序的特征信息的步骤包括:
    确定待评估的目标应用程序;
    从所述日志数据集合中获取所述目标应用程序的访问行为特征;
    从所述公共关系数据集合中获取所述目标应用程序的公共关系特征;
    从所述社交网络服务数据集合中获取所述目标应用程序的社交属性特征。
  5. 根据权利要求4所述的方法,其特征在于,所述确定待评估的目标应用程序的步骤包括:
    获取所述日志数据集合中记录的,每个应用程序在预设时间段内的访问次数;
    基于所述访问次数,对所述应用程序进行排序;
    将排序在前的N个应用程序确定为待评估的目标应用程序,其中,N为正整数。
  6. 根据权利要求4或5所述的方法,其特征在于,所述访问行为特征至少包括:所述目标应用程序的日均独立访客量,和/或,日活跃用户数,和/或,日均平均使用时长;
    和/或,
    所述公共关系特征至少包括:与所述目标应用程序关联的文稿的数量,和/或,所述与所述目标应用程序关联的文稿的评论数和/或转发数;
    和/或,
    所述社交属性特征至少包括:所述目标应用程序被下载的次数,和/或,所述目标应用程序的官方微博中的粉丝数和/或大V粉丝数和/或关注数和/或官方微博的所有评论数和/或转发数,所述目标应用程序的相关人员名称,和/或,所述相关人员的微博的粉丝数和/或大V粉丝数和/或关注数和/或所述相关人员的微博的所有评论数和/或转发数。
  7. 根据权利要求6所述的方法,其特征在于,所述相关人员包括投资者和/或开发者和/或运维推广者。
  8. 根据权利要求3所述的方法,其特征在于,所述项目信用系数包括期望授信金额,所述基于所述目标应用程序的特征信息,获取所述目标应用程序的项目信用系数的步骤包括:
    创建所述目标应用程序的知识图谱;
    以单个目标应用程序为主键,将所述知识图谱、所述访问行为特征、所述公共关系特征、所述社交属性特征,拼合成所述目标应用程序的特征数据表;
    基于所述特征数据表,预估所述目标应用程序的期望授信金额。
  9. 根据权利要求8所述的方法,其特征在于,所述创建所述目标应用程序的知识图谱的步骤包括:
    从所述公共关系数据集合中获得与所述目标应用程序相关联的文稿;
    对所述相关联的文稿进行分词处理,获得所述文稿的有效词,并统计所述有效词的词频;
    将词频排序在前M的有效词,作为所述目标应用程序的实体知识,其中,M为正整数;
    识别所述实体知识的实体类型;
    生成所述目标应用程序、所述实体类型以及所述实体知识的映射关系;
    将所有映射关系组织成所述目标应用程序的知识图谱。
  10. 根据权利要求8所述的方法,其特征在于,所述基于所述特征数据表,预估所述目标应用程序的期望授信金额的步骤包括:
    获取在先获得投资的已授信应用程序的授信数据,作为案例集合;
    根据所述案例集合以及所述特征数据表,生成训练样本;
    对所述训练样本进行模型训练,生成预测模型;
    针对所述预测模型,计算所述目标应用程序的期望授信金额。
  11. 根据权利要求8或9或10所述的方法,其特征在于,所述项目信用系数还包括信用评分,所述方法还包括:
    对所述期望授信金额进行对数运算以及归一化处理,得到所述目标 应用程序的信用评分。
  12. 一种对应用程序进行项目评估的系统,其特征在于,所述系统包括:
    异构数据获取模块,用于获取多种异构数据;
    特征信息获取模块,用于分别从所述异构数据中获取待评估的目标应用程序的特征信息;
    项目信用系数获取模块,用于基于所述目标应用程序的特征信息,获取所述目标应用程序的项目信用系数。
  13. 根据权利要求12所述的系统,其特征在于,所述异构数据获取模块包括:
    异构数据获取子模块,用于分别从预设的多个数据资源站点中获取对应的异构数据;
    组织子模块,用于分别将所述异构数据组织成异构数据集合。
  14. 根据权利要求13所述的系统,其特征在于,所述异构数据至少包括:日志数据、公共关系数据以及社交网络服务数据;所述异构数据集合至少包括:日志数据集合、公共关系数据集合以及社交网络服务数据集合。
  15. 根据权利要求14所述的系统,其特征在于,所述特征信息至少包括:访问行为特征、公共关系特征、社交属性特征;
    所述特征信息获取模块包括:
    目标应用程序确定子模块,用于确定待评估的目标应用程序;
    访问特征获取子模块,用于从所述日志数据集合中获取所述目标应用程序的访问行为特征;
    公共关系特征获取子模块,用于从所述公共关系数据集合中获取所述目标应用程序的公共关系特征;
    社交特征获取子模块,用于从所述社交网络服务数据集合中获取所述目标应用程序的社交属性特征。
  16. 根据权利要求15所述的系统,其特征在于,所述目标应用程序确定子模块包括:
    访问次数获取单元,用于获取所述日志数据集合中记录的,每个应用程序在预设时间段内的访问次数;
    排序单元,用于基于所述访问次数,对所述应用程序进行排序;
    确定单元,用于将排序在前的N个应用程序确定为待评估的目标应用程序,其中,N为正整数。
  17. 根据权利要求15或16所述的系统,其特征在于,所述访问行为特征至少包括:所述目标应用程序的日均独立访客量,和/或,日活跃用户数,和/或,日均平均使用时长;
    和/或,
    所述公共关系特征至少包括:与所述目标应用程序关联的文稿的数量,和/或,所述与所述目标应用程序关联的文稿的评论数和/或转发数;
    和/或,
    所述社交属性特征至少包括:所述目标应用程序被下载的次数,和/或,所述目标应用程序的官方微博中的粉丝数和/或大V粉丝数和/或关注数和/或官方微博的所有评论数以及转发数,和/或,所述目标应用程序的相关人员名称,和/或,所述相关人员的微博的粉丝数和/或大V粉丝数和/或关注数和/或所述相关人员的微博的所有评论数和/或转发数。
  18. 根据权利要求17所述的系统,其特征在于,所述相关人员包括投资者和/或开发者和/或运维推广者。
  19. 根据权利要求14所述的系统,其特征在于,所述项目信用系数包括期望授信金额,所述项目信用系数获取模块包括:
    知识图谱创建子模块,用于创建所述目标应用程序的知识图谱;
    特征合并子模块,用于以单个目标应用程序为主键,将所述知识图谱、所述访问行为特征、所述公共关系特征、所述社交属性特征,拼合成所述目标应用程序的特征数据表;
    金额预估子模块,用于基于所述特征数据表,预估所述目标应用程序的期望授信金额。
  20. 根据权利要求19所述的系统,其特征在于,所述知识图谱创建子模块包括:
    关联文稿获取单元,用于从所述公共关系数据集合中获得与所述目标应用程序相关联的文稿;
    文稿分词单元,用于对所述相关联的文稿进行分词处理,获得所述文稿的有效词,并统计所述有效词的词频;
    知识确定单元,用于将词频排序在前M的有效词,作为所述目标应用程序的实体知识,其中,M为正整数;
    类型识别单元,用于识别所述实体知识的实体类型;
    实体映射单元,用于生成所述目标应用程序、所述实体类型以及所述实体知识的映射关系;
    知识图谱构建单元,用于将所有映射关系组织成所述目标应用程序的知识图谱。
  21. 根据权利要求19所述的系统,其特征在于,所述金额预估子模块包括:
    案例获取单元,用于获取在先获得投资的已授信应用程序的授信数据,作为案例集合;
    样本获取单元,用于根据所述案例集合以及所述特征数据表,生成训练样本;
    模型训练单元,用于对所述训练样本进行模型训练,生成预测模型;
    授信金额计算单元,用于针对所述预测模型,计算所述目标应用程序的期望授信金额。
  22. 根据权利要求19或20或21所述的系统,其特征在于,所述项目信用系数还包括信用评分,所述系统还包括:
    信用评分获取模块,用于对所述期望授信金额进行对数运算以及归一化处理,得到所述目标应用程序的信用评分。
PCT/CN2017/077503 2016-03-30 2017-03-21 一种对应用程序进行项目评估的方法及系统 WO2017167071A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610194798.7A CN107292463A (zh) 2016-03-30 2016-03-30 一种对应用程序进行项目评估的方法及系统
CN201610194798.7 2016-03-30

Publications (1)

Publication Number Publication Date
WO2017167071A1 true WO2017167071A1 (zh) 2017-10-05

Family

ID=59963435

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/077503 WO2017167071A1 (zh) 2016-03-30 2017-03-21 一种对应用程序进行项目评估的方法及系统

Country Status (3)

Country Link
CN (1) CN107292463A (zh)
TW (1) TW201737072A (zh)
WO (1) WO2017167071A1 (zh)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109960650A (zh) * 2018-09-04 2019-07-02 中国平安人寿保险股份有限公司 基于大数据的应用程序评估方法、装置、介质及电子设备
CN110781311A (zh) * 2019-09-18 2020-02-11 上海生腾数据科技有限公司 一种企业一致行动人运算系统及方法
CN110990637A (zh) * 2019-10-14 2020-04-10 平安银行股份有限公司 网络图谱的构建方法及其装置
CN111104584A (zh) * 2018-10-25 2020-05-05 北京国双科技有限公司 信息价值度的确定方法、装置、存储介质和处理器
CN111625655A (zh) * 2020-05-12 2020-09-04 埃睿迪信息技术(北京)有限公司 基于知识图谱的归并以及分类方法、装置以及存储介质
CN112734598A (zh) * 2019-10-29 2021-04-30 北京国双科技有限公司 建立外包平台模型的方法、外包方查找方法及装置
CN113362158A (zh) * 2021-05-31 2021-09-07 中国银联股份有限公司 一种信用评估方法、装置及计算机可读存储介质

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110046981B (zh) * 2018-01-15 2022-03-08 腾讯科技(深圳)有限公司 一种信用评估方法、装置及存储介质
CN110163460B (zh) * 2018-03-30 2023-09-19 腾讯科技(深圳)有限公司 一种确定应用分值的方法及设备
CN109242573A (zh) * 2018-09-26 2019-01-18 湖南人文科技学院 App的评价方法、装置、设备及存储介质
CN109472312A (zh) * 2018-11-14 2019-03-15 平安科技(深圳)有限公司 应用程序评估方法、装置、计算机设备及存储介质
CN109885767B (zh) * 2019-02-18 2021-02-05 上海海事大学 一种基于GitHub的软件资产推荐的方法及系统
CN111178615B (zh) * 2019-12-24 2023-10-27 成都数联铭品科技有限公司 一种企业风险识别模型的构建方法及系统
TWI736233B (zh) * 2020-04-23 2021-08-11 兆豐國際商業銀行股份有限公司 貸前調查系統以及貸前調查方法
CN112270014B (zh) * 2020-10-16 2022-06-10 维沃移动通信有限公司 应用程序控制方法、装置及电子设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7669180B2 (en) * 2004-06-18 2010-02-23 International Business Machines Corporation Method and apparatus for automated risk assessment in software projects
CN102479150A (zh) * 2010-11-25 2012-05-30 神州数码信息系统有限公司 一种软件可信评估的方法
CN103208039A (zh) * 2012-01-13 2013-07-17 株式会社日立制作所 软件项目风险评价方法及装置
CN105389250A (zh) * 2014-09-05 2016-03-09 北京畅游天下网络技术有限公司 追踪软件质量的方法和系统

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102222040A (zh) * 2011-06-09 2011-10-19 西北工业大学 基于多属性熵权合成的软件可信等级评估方法
CN102880501B (zh) * 2012-07-24 2016-05-25 北京奇虎科技有限公司 应用推荐的实现方法、装置和系统
IN2013MU03907A (zh) * 2013-12-13 2015-07-31 Tata Consultancy Services Ltd
CN103678281B (zh) * 2013-12-31 2016-10-19 北京百度网讯科技有限公司 对文本进行自动标注的方法和装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7669180B2 (en) * 2004-06-18 2010-02-23 International Business Machines Corporation Method and apparatus for automated risk assessment in software projects
CN102479150A (zh) * 2010-11-25 2012-05-30 神州数码信息系统有限公司 一种软件可信评估的方法
CN103208039A (zh) * 2012-01-13 2013-07-17 株式会社日立制作所 软件项目风险评价方法及装置
CN105389250A (zh) * 2014-09-05 2016-03-09 北京畅游天下网络技术有限公司 追踪软件质量的方法和系统

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109960650A (zh) * 2018-09-04 2019-07-02 中国平安人寿保险股份有限公司 基于大数据的应用程序评估方法、装置、介质及电子设备
CN109960650B (zh) * 2018-09-04 2024-04-02 中国平安人寿保险股份有限公司 基于大数据的应用程序评估方法、装置、介质及电子设备
CN111104584A (zh) * 2018-10-25 2020-05-05 北京国双科技有限公司 信息价值度的确定方法、装置、存储介质和处理器
CN110781311A (zh) * 2019-09-18 2020-02-11 上海生腾数据科技有限公司 一种企业一致行动人运算系统及方法
CN110781311B (zh) * 2019-09-18 2024-02-27 上海合合信息科技股份有限公司 一种企业一致行动人运算系统及方法
CN110990637A (zh) * 2019-10-14 2020-04-10 平安银行股份有限公司 网络图谱的构建方法及其装置
CN110990637B (zh) * 2019-10-14 2022-09-20 平安银行股份有限公司 网络图谱的构建方法及其装置
CN112734598A (zh) * 2019-10-29 2021-04-30 北京国双科技有限公司 建立外包平台模型的方法、外包方查找方法及装置
CN111625655A (zh) * 2020-05-12 2020-09-04 埃睿迪信息技术(北京)有限公司 基于知识图谱的归并以及分类方法、装置以及存储介质
CN113362158A (zh) * 2021-05-31 2021-09-07 中国银联股份有限公司 一种信用评估方法、装置及计算机可读存储介质

Also Published As

Publication number Publication date
TW201737072A (zh) 2017-10-16
CN107292463A (zh) 2017-10-24

Similar Documents

Publication Publication Date Title
WO2017167071A1 (zh) 一种对应用程序进行项目评估的方法及系统
AU2016298250B2 (en) Classifying user behavior as anomalous
Shen et al. Linking named entities in tweets with knowledge base via user interest modeling
Schintler et al. Big data for policy analysis: The good, the bad, and the ugly
CN104574192B (zh) 在多个社交网络中识别同一用户的方法及装置
Chianese et al. Cultural heritage and social pulse: a semantic approach for CH sensitivity discovery in social media data
US10956469B2 (en) System and method for metadata correlation using natural language processing
US11606630B1 (en) Providing content related to a selected channel for presentation to a user via a client device
Liu et al. Do rumors diffuse differently from non-rumors? a systematically empirical analysis in sina weibo for rumor identification
JP2022519208A (ja) ソーシャル・データ・ストリーム上におけるリアルタイム・イベント検出
SG192380A1 (en) Social media data analysis system and method
CN109255000B (zh) 一种标签数据的维度管理方法及装置
US20130246463A1 (en) Prediction and isolation of patterns across datasets
CN111046237A (zh) 用户行为数据处理方法、装置、电子设备及可读介质
Rani et al. A survey of tools for social network analysis
Roedler et al. Profile matching across online social networks based on geo-tags
JP2021093163A (ja) ディープラーニングに基づく文書類似度測定モデルを利用した重複文書探知方法およびシステム
CN110720097A (zh) 图数据库中元组和边的功能性等价
US9336311B1 (en) Determining the relevancy of entities
Heravi et al. Tweet location detection
CN115470489A (zh) 检测模型训练方法、检测方法、设备以及计算机可读介质
CN112685572B (zh) 一种异构数据融合方法及装置
CN115238163A (zh) 基于文献数据的信息推送方法及装置、存储介质、终端
Huang et al. Web content adaptation for mobile device: A fuzzy-based approach
Wang et al. Positive and unlabeled learning for mobile app user and server interaction prediction

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

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

Ref document number: 17773101

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 17773101

Country of ref document: EP

Kind code of ref document: A1