Application type identification method and device of application program and electronic equipment

Abstract

The invention relates to the technical field of deep learning, in particular to an application type identification method and device of an application program and electronic equipment, and the application type is intelligently identified by automatically utilizing a preset search engine to retrieve function description of a target application program and combining a large-scale pre-training model and a preset cue word. The efficiency and accuracy of application classification are remarkably improved, the possible problems of low efficiency and error proneness in manual identification are effectively avoided, and a rapid and accurate automatic application classification process is realized.

Description

Technical Field

[0001] The present application relates to the field of deep learning technology, and specifically to a method, device and electronic device for identifying the application type of an application. Background Art

[0002] In the era of mobile Internet, with the explosive growth of various applications, users jumping and interacting between different applications has become the norm. In order to improve user experience and enhance user stickiness, many applications have begun to adopt cross-application jump technology, that is, directly pulling up the target application from the current application through specific links (such as URL Scheme, Universal Links, App Links, etc.). This technology not only simplifies the user's operation process, but also makes it possible for cooperation and traffic sharing between applications.

[0003] However, as the application market grows, the number of application package names (i.e., the unique identifier of the application, such as com.example.app) has increased dramatically, which has brought great challenges to application operators in user behavior analysis, accurate push and subsequent operation. Especially when it is necessary to formulate different operation strategies according to the type of the pulled application (such as video, music, novel, game, etc.), the traditional method based on manual identification of application package names and classification is not only inefficient but also prone to errors.

[0004] It can be seen that the method based on manual identification of application package names and classification in the prior art has the problems of low classification efficiency and prone to errors. Summary of the invention

[0005] The purpose of the present application is to provide an application type identification method, device and electronic device for an application program, so as to solve the problem that the method based on manual identification of application package names and classification in the prior art has low classification efficiency and is prone to errors.

[0006] In order to achieve the above purpose, the technical solution adopted in this application is as follows:

[0007] According to one aspect of an embodiment of the present application, a method for identifying an application type of an application is provided, comprising: determining a target application to be identified; using a preset search engine to search for functions of the target application to obtain target search results, wherein the target search results are used to describe the functions of the target application; calling a preset large model, and identifying the application type of the target application based on the target search results and preset prompt words, to obtain a target application type corresponding to the target application, wherein the prompt words are used to indicate candidate application types of the application.

[0008] According to the above technical means, by automatically using a preset search engine to retrieve the functional description of the target application, and combining a large pre-trained model with preset prompt words to intelligently identify the application type, the efficiency and accuracy of application classification are significantly improved, effectively avoiding the inefficiency and error-prone problems that may exist in manual identification, and realizing a fast and accurate automated application classification process.

[0009] Further, a preset big model is called to identify the application type of the target application according to the target search results and preset prompt words to obtain the target application type corresponding to the target application, including: preprocessing the text corresponding to the target search results through the big model to obtain multiple words; converting each of the multiple words into an embedding vector to obtain a set of embedding vectors corresponding to the multiple words; extracting features of all the embedding vectors in the embedding vector set through the neural network structure of the big model to obtain text features, wherein the text features are used to characterize the content described by the target search results; calling the classifier of the big model to calculate the probability value corresponding to each candidate application type in the prompt word according to the text features and the knowledge learned by the big model during the training process; traversing the probability values ​​corresponding to all candidate application types through the big model, and selecting the candidate application type with the highest probability as the target application type corresponding to the target application.

[0010] Based on the above-mentioned technical means, the target search results are deeply processed and feature extracted through a large model, and the knowledge accumulated during the training process is combined to accurately calculate and select the candidate application type that best matches the actual type of the target application, thereby achieving efficient and accurate automatic application type identification, significantly improving the accuracy and efficiency of classification.

[0011] Further, a preset search engine is used to search for functions of the target application to obtain target search results, including: using the target application package name of the target application to combine with any one of the preset multiple search keywords, thereby constructing multiple search statements, wherein the target application package name is a unique identifier of the target application; inputting the multiple search statements into the search engine for searching to obtain multiple initial search results corresponding to the multiple search statements; performing data cleaning on the multiple initial search results to obtain initial text data; extracting target text data related to the application type from the initial text data; and aggregating all target text data to obtain target search results.

[0012] According to the above-mentioned technical means, by carefully constructing a search statement based on the target application package name and preset search keywords, we can widely search and filter out accurate information related to the functions of the target application program, effectively filter out non-critical content, and thus gather an information collection that is highly focused on the application type, namely the target search results. This process greatly improves the data quality and efficiency of subsequent application type identification.

[0013] Furthermore, the multiple search keywords include application details, application type, function description and function introduction.

[0014] According to the above-mentioned technical means, by using multiple search keywords (such as application details, application type, function description and function introduction) to construct search statements, it is possible to more comprehensively cover all aspects of the target application, thereby ensuring that the search results obtained are richer and more accurate. This not only improves the efficiency of data cleaning and extraction, but also enhances the reliability and accuracy of subsequent application type identification.

[0015] Further, after obtaining the target application type corresponding to the target application, the method includes: saving the target application package name and the corresponding target application type of the target application in a preset classification information database.

[0016] According to the above-mentioned technical means, after obtaining the target application type corresponding to the target application, it is saved together with the target application package name in a preset classification information database, thereby realizing effective management and storage of the application and its classification information, and providing a convenient and efficient basis for subsequent data query, analysis and application management.

[0017] Furthermore, before using a preset search engine to search for functions of the target application and obtaining target search results, it includes: using a preset classification information database to determine whether the target application type of the target application can be identified, wherein the classification information database is used to store application types of applications; when it is determined that the target application type of the target application can be identified based on the classification information database, the application type corresponding to the application package name that is the same as the target application package name in the classification information database is used as the target application type corresponding to the target application.

[0018] According to the above technical means, before searching, the preset classification information database is first queried to determine whether the application type of the target application can be directly identified. If it already exists, the application type is directly extracted, avoiding unnecessary search processes, significantly improving the recognition efficiency, and also reducing the frequency of using the search engine, saving resources.

[0019] Furthermore, a preset classification information database is used to determine whether the target application type of the target application can be identified, including: querying the classification information database to determine whether there is an application package name in the classification information database that is identical to the target application package name of the target application; if there is an application package name in the classification information database that is identical to the target application package name, then it is determined that the application type of the target application can be identified; otherwise, it is determined that the application type of the target application cannot be identified.

[0020] According to the above technical means, by querying the preset classification information database, it is possible to quickly determine whether the application type of the target application can be directly identified, thereby effectively reducing unnecessary search operations. This pre-check mechanism improves the efficiency and accuracy of the overall identification process.

[0021] According to another aspect of an embodiment of the present application, an application type identification device is also provided, including: an application determination module, used to determine a target application to be identified; a content search module, used to search the target application using a preset search engine to obtain a target search result, wherein the target search result is used to describe the function of the target application; a type identification module, used to call a preset large model, and perform application type identification on the target application according to the target search result and a preset prompt word, to obtain a target application type corresponding to the target application, wherein the prompt word is used to indicate a candidate application type of the application.

[0022] According to another aspect of the embodiments of the present application, there is also provided an electronic device, including a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; wherein the memory is used to store computer programs; and the processor is used to execute the application type identification method steps of the application program in any of the above embodiments by running the computer program stored in the memory.

[0023] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the application type identification method steps of the application program in any of the above embodiments when running.

[0024] Beneficial effects of this application:

[0025] This application significantly improves the efficiency and accuracy of application classification by automatically using a preset search engine to retrieve the functional description of the target application, and combines a large pre-trained model with preset prompt words to intelligently identify the application type, effectively avoiding the inefficiency and error-prone problems that may exist in manual identification, and realizing a fast and accurate automated application classification process. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0027] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0028] Figure 1 It is a schematic diagram of a hardware environment of an optional application type identification method of an application program provided in an embodiment of the present application;

[0029] Figure 2 It is a flowchart of an optional application type identification method of an application provided in an embodiment of the present application;

[0030] Figure 3 It is a structural block diagram of an application type identification system of an optional application provided in an embodiment of the present application;

[0031] Figure 4 It is a structural block diagram of an optional application type identification device for an application provided in an embodiment of the present application;

[0032] Figure 5 It is a structural block diagram of an optional electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0033] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0034] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0035] According to one aspect of an embodiment of the present application, a method for identifying an application type of an application is provided. Optionally, in this embodiment, the method for identifying an application type of the above application can be applied to a hardware environment consisting of a terminal and a server. The server is connected to the terminal via a network, and can be used to provide services to the terminal or a client installed on the terminal. A database can be set on the server or independently of the server to provide data storage services for the server.

[0036] The above network may include but is not limited to at least one of the following: wired network, wireless network. The above wired network may include but is not limited to at least one of the following: wide area network, metropolitan area network, local area network, and the above wireless network may include but is not limited to at least one of the following: WIFI (Wireless Fidelity), Bluetooth. The terminal may not be limited to a PC, a mobile phone, a tablet computer, etc.

[0037] The application type identification method of the application in the embodiment of the present application can be executed by the server, or by the terminal, or by both the server and the terminal. The terminal executing the application type identification method of the application in the embodiment of the present application can also be executed by the client installed thereon.

[0038] Taking the method for identifying the application type of the application program in this embodiment executed by the terminal and the server together as an example, see Figure 1 , Figure 1 is a schematic diagram of a hardware environment of an optional application type identification method of an application provided in an embodiment of the present application, such as Figure 1 As shown, the hardware environment of the application type identification method of the application includes: a terminal 102, and a server 104 connected to the terminal 102 via a network. The terminal 102 is used to display the application type of the target application; the server 104 is used to execute the application type identification method of the application of the embodiment of the present application, and classify and identify the application type of the target application.

[0039] The application type identification method of the application in this embodiment can be applied to scenarios such as user behavior analysis and precise push by application operators. In particular, when the application operator needs to formulate different operation strategies according to the type of the pulled application (such as video, music, novel, game, etc.), the traditional method based on manual identification of application package names and classification is not only inefficient but also prone to errors.

[0040] To solve the above problem, this embodiment provides an application type identification method for an application running on the above server. Figure 2 , Figure 2FIG. 1 is a flow chart of an optional application type identification method of an application program provided in an embodiment of the present application. Figure 2 As shown, the application type identification method of the application in the embodiment of the present application specifically includes the following steps:

[0041] Step S201, determining a target application to be identified;

[0042] Step S202, searching for the function of the target application using a preset search engine to obtain a target search result, wherein the target search result is used to describe the function of the target application;

[0043] Step S203, calling a preset large model, identifying the application type of the target application according to the target search result and the preset prompt word, and obtaining the target application type corresponding to the target application, wherein the prompt word is used to indicate the candidate application type of the application.

[0044] Through the above steps S201 to S203, the functions of the target application to be identified are searched by using a preset search engine to obtain the target search results, and the preset large model is called to identify the application type of the target application according to the target search results and the preset prompt words to obtain the target application type corresponding to the target application. This solves the problem of low classification efficiency and prone to errors in the method based on manual identification of application package names and classification in the related art. By automatically using the preset search engine to retrieve the function description of the target application, and combining the large pre-trained model with the preset prompt words to intelligently identify the application type, the efficiency and accuracy of application classification are significantly improved, and the inefficiency and error-prone problems that may exist in manual identification are effectively avoided, realizing a fast and accurate automated application classification process.

[0045] Combine the following Figure 2 The application type identification method of the application program in the embodiment of the present application is explained.

[0046] In the technical solution of step S201, a target application to be identified is determined.

[0047] The target application refers to the target application that is directly launched from the current application through a specific link (such as URL Scheme, Universal Links, App Links, etc.). Here, application launching refers to the process in which the system automatically launches the corresponding application and enters its homepage or specific page after clicking an application icon on the client. For example, when a user chooses to jump to an external application from the client running the current application, the external application is the target application.

[0048] As an optional embodiment, determining the target application to be identified includes: obtaining the target application package name of the target application to be identified from the application package name list; and verifying the target application according to the target application package name through a preset classification information database, wherein the verification is used to confirm that the target application cannot be classified using the classification information database.

[0049] In specific practice, the target application package name is a unique identifier of the target application program, and the application package name and application type of the application program are stored in a preset classification information database.

[0050] It is understandable that the above steps constitute an efficient and structured method to identify and handle target applications that cannot be directly classified through the preset classification information database. By obtaining the package name of the application to be identified from the application package name list and attempting to verify it using the classification information database, this process not only ensures the automation of the classification attempt, but also clearly identifies applications that are not covered by the classification information database. This approach provides a clear basis for subsequent targeted processing measures (such as manual classification, database updates, etc.), which helps to improve the efficiency and accuracy of overall application management.

[0051] In the technical solution of step S202, a preset search engine is used to search for functions of the target application to obtain target search results.

[0052] A search engine refers to a system that collects information from the Internet using a specific computer program according to a certain search strategy, organizes and processes the information, provides retrieval services to users, and displays information related to the user's search to the user. In this embodiment, the search engine can be Baidu, Google, Bing, etc. The target search result is used to describe the function of the target application. In this embodiment, the function of the target application is searched using a preset search engine, and the target search result is output. As an optional embodiment, the function of the target application is searched using a preset search engine to obtain the target search result, including: using the target application package name of the target application and combining any one of the preset multiple search keywords to construct multiple search statements, wherein the target application package name is the unique identifier of the target application; inputting multiple search statements into the search engine for search, and obtaining multiple initial search results corresponding to the multiple search statements; performing data cleaning on the multiple initial search results to obtain initial text data; extracting target text data related to the application type from the initial text data; and aggregating all target text data to obtain the target search result.

[0053] The target application package name is a unique identifier of the target application, and an application package name such as com.hainanscsj.zxfd. In specific practice, the application package name is stored in an application package name list. The search keyword is used to indicate which content to search specifically about the target application. In this embodiment, the search keyword is used to indicate a specific search for content related to the function or application type of the target application. Since the specific search result is HTML content, there are irrelevant content such as HTML tags or advertisements in the content. In this embodiment, the irrelevant content is removed by data cleaning.

[0054] As an optional embodiment, the multiple search keywords include application details, application type, function description and function introduction. In specific practice, the search keyword can also be other words related to application functions or application types. It can be understood that by using multiple search keywords (such as application details, application type, function description and function introduction) to construct a search statement, it is possible to more comprehensively cover all aspects of the target application, thereby ensuring that the search results obtained are richer and more accurate, which not only improves the efficiency of data cleaning and extraction, but also enhances the reliability and accuracy of subsequent application type identification.

[0055] In practice, if the target application package name is com.hainanscsj.zxfd, and the search keyword includes application details, then in this embodiment, the target application package name and the search keyword are combined to construct a search statement, which is specifically com.hainanscsj.zxfd application details. The search statement is actually a search strategy for the functions of the target application, and the search engine uses the search strategy to search for relevant content of the target application.

[0056] After obtaining the search statement, the search statement is concatenated with the query link of the preset search engine to form a URL for initiating an HTTP request, and the initial search results corresponding to different search statements are obtained through the HTTP request. If the preset search engine is Bing and the search statement is com.hainanscsj.zxfd application details, then the URL is: https: / / www.bing.com / search?q=com.hainanscsj.zxfd application details.

[0057] In specific practice, each search statement may have one search result or multiple search results. When there are multiple search results, the first N search results are taken, where N is a positive integer greater than 0. In this embodiment, N is preferably 3. In specific practice, when there are multiple search statements, the first N search results corresponding to each search statement can be taken, and then merged to obtain multiple initial search results.

[0058] Remove irrelevant content such as HTML tags or advertisements in the HTML content corresponding to multiple initial search results to obtain multiple initial text data, extract valuable target text data related to the application type such as title or description text in each initial text, and aggregate multiple target text data to obtain target search results. For example, the application package name is com.hainanscsj.zxfd, and the example of using Bing search results (target search results) is as follows: Zhenxiang Restaurant Red Envelope Edition is a fun restaurant simulation business mobile game with exquisite cartoon pictures. Here you can open a restaurant of your own, make money while opening a restaurant, and manage your own restaurant well. There are many recipes that can be unlocked in the latest official version of Zhenxiang Restaurant, and you need to receive many customers. Various business tasks are waiting for you to complete. Continuously upgrade the level of the restaurant to get more rewards, and continue to manage to create the best restaurant.

[0059] It can be understood that by carefully constructing search statements based on the target application package name and preset search keywords, we can widely search and filter out accurate information related to the functions of the target application, effectively filter out non-critical content, and thus gather an information collection that is highly focused on the application type, namely the target search results. This process greatly improves the data quality and efficiency of subsequent application type identification.

[0060] As an optional embodiment, before using a preset search engine to search for functions of a target application and obtaining target search results, the method includes: using a preset classification information database to determine whether a target application type of the target application can be identified, wherein the classification information database is used to store application types of applications; when it is determined that the target application type of the target application can be identified based on the classification information database, using the application type corresponding to the application package name that is identical to the target application package name in the classification information database as the target application type corresponding to the target application.

[0061] The preset classification information database stores the application package name and application type of the application. It is understandable that before searching, the preset classification information database is first queried to determine whether the application type of the target application can be directly identified. If it already exists, the application type is directly extracted, avoiding unnecessary search processes, significantly improving identification efficiency, and also reducing the frequency of use of the search engine, saving resources.

[0062] As an optional embodiment, a preset classification information database is used to determine whether the target application type of the target application can be identified, including: querying the classification information database to determine whether there is an application package name in the classification information database that is identical to the target application package name of the target application; if there is an application package name in the classification information database that is identical to the target application package name, then it is determined that the application type of the target application can be identified, otherwise, it is determined that the application type of the target application cannot be identified.

[0063] It is understandable that by querying the preset classification information database, it is possible to quickly determine whether the application type of the target application can be directly identified, thereby effectively reducing unnecessary search operations. This pre-check mechanism improves the efficiency and accuracy of the overall identification process.

[0064] As an optional embodiment, when it is determined that the target application type of the target application cannot be identified according to the classification information database, the target application package name of the target application is stored in an application package name list. The application package name list is used to store the application package name of the target application that cannot be identified using the classification information database.

[0065] It is understandable that when the type of the target application cannot be determined based on the classification information database, storing the package name of the target application in the application package name list can accumulate and retain the identification information of unclassified or newly emerging applications, providing a data basis for subsequent analysis, updating of classification models or manual classification, thereby gradually improving the accuracy and coverage of application identification.

[0066] In the technical solution of step S203, a preset large model is called to identify the application type of the target application according to the target search result and the preset prompt word, so as to obtain the target application type corresponding to the target application.

[0067] In this embodiment, the preset large model, namely the Large Language Model (LLM), is an important concept in the field of artificial intelligence, especially natural language processing (NLP). These models are usually based on deep learning technology, especially models based on Transformer structure, such as the GPT (Generative Pre-trained Transformer) series, which learns complex features such as language grammar, semantics, and contextual relationships by pre-training on massive text data. In this embodiment, the specific large model used is no longer specifically limited. The preset prompt word is used to indicate the candidate application type of the application. In this embodiment, the prompt word (prompt) is for example:

[0068] You are an app package name classifier, and you classify app package names into one of the six categories based on the following search results: game (games), novel (novels, reading, audiobooks, audiobooks, audio novels, literary works, comics, etc.), music (children's songs, music, etc.), video (short plays, movies, TV shows, live broadcasts, videos, etc.), feed (information flow, news, social networking, search, browsers, etc.), tool (video editing, ringtone production, monitoring, weather, calendar, picture editing, tools, etc.);

[0069] Search result text: {search_text};

[0070] Application package name: {app_name}.

[0071] In actual application, the search result text in the above prompt is replaced with the text of the target search result, and the application package name is replaced with the target application package name corresponding to the target application, so as to obtain the prompt corresponding to the processed target application. The prompt corresponding to the processed target application is input into the preset large model for classification to obtain the target application type. The target application type can be any one of the following candidate application types: game (game category), novel (novel, reading, listening to books, audio books, audio novels, literary works, comics, etc.), music (children's songs, music, etc.), video (short drama, movie, film and television, live broadcast, video, etc.), feed (information flow, news, social, search, browser, etc.), tool (video editing, ringtone production, monitoring, weather, calendar, picture editing, tools, etc.).

[0072] As an optional embodiment, a preset big model is called to identify the application type of the target application according to the target search results and preset prompt words to obtain the target application type corresponding to the target application, including: preprocessing the text corresponding to the target search results through the big model to obtain multiple words; converting each of the multiple words into an embedding vector to obtain a set of embedding vectors corresponding to the multiple words; extracting features of all the embedding vectors in the embedding vector set through the neural network structure of the big model to obtain text features, wherein the text features are used to characterize the content described by the target search results; calling the classifier of the big model to calculate the probability value corresponding to each candidate application type in the prompt word according to the text features and the knowledge learned by the big model during the training process; traversing the probability values ​​corresponding to all candidate application types through the big model, and selecting the candidate application type with the highest probability as the target application type corresponding to the target application.

[0073] In this embodiment, the preset large model is generated based on the probability model. In the classification task, the most suitable target application type is determined by calculating the probability of each candidate application type word. Since the large model itself has been trained with a large-scale corpus and a specific prompt word, it can output the candidate application type with the highest probability, which is the target application type. In this embodiment, the candidate application types include: game (game category), novel (novel, reading, listening to books, audio books, audio novels, literary works, comics, etc.), music (children's songs, music, etc.), video (short drama, movie, film and television, live broadcast, video, etc.), feed (information flow, news, social, search, browser, etc.), tool (video editing, ringtone production, monitoring, weather, calendar, picture editing, tools, etc.).

[0074] According to the content of the target search result and the candidate application types in the above step S202, by calling the preset large model API, it can be determined that the application with the application package name com.hainanscsj.zxfd has the highest probability of being game (game type) among the candidate application types, so the application type of the application with the application package name com.hainanscsj.zxfd is game (game type).

[0075] It can be understood that by deeply processing and extracting features of the target search results through a large model, combined with the knowledge accumulated during the training process, the candidate application type that best matches the actual type of the target application is accurately calculated and selected, thereby achieving efficient and accurate automatic application type identification and significantly improving the accuracy and efficiency of classification.

[0076] As an optional embodiment, the application type results are unified into a standardized output. Specifically, the response text returned by the API request of the large model is received, the classification information is extracted from the text, and the application type results are unified into a standardized output using string matching. For example, "game", "game category", "game", and "Game" are standardized to "game". In the above example, the output result of the application type of the application with the application package name com.hainanscsj.zxfd is game. If the standardized result cannot be obtained or the classification does not belong to one of the preset six candidate application types, a retry is added, that is, the large model is re-called for classification.

[0077] It is understandable that by standardizing the output of application type results, the consistency and processability of the data are ensured, while string matching is used to improve the accuracy and efficiency of data processing. For results that cannot be directly standardized or applications that do not belong to the preset category, the large model is called for reclassification through the retry mechanism, which enhances the accuracy and robustness of the classification and effectively avoids omissions or errors in data processing.

[0078] As an optional embodiment, after obtaining the target application type corresponding to the target application, the method includes: saving the target application package name and the corresponding target application type of the target application in a preset classification information database.

[0079] It can be understood that after obtaining the target application type corresponding to the target application, it is saved together with the target application package name in a preset classification information database, thereby achieving effective management and storage of the application and its classification information, providing a convenient and efficient basis for subsequent data query, analysis and application management.

[0080] According to another aspect of the embodiment of the present application, there is also provided an application type identification system for implementing the application type identification method of the above application, see Figure 3 , Figure 3 is a structural block diagram of an optional application type identification system provided in an embodiment of the present application, such as Figure 3 As shown, the application type identification system of the application includes: a client, a server and a classification information database.

[0081] The client is used to obtain the target application to be identified, and pass the target application package name of the target application to the server. After receiving the target application package name, the server stores the target application package name in the application package name list for subsequent offline classification, which is used to determine the target application type of the target application to be identified. The execution process of offline classification is specifically as follows: obtain the target application package name in the application package name list; use a preset search engine to search for the target application package name to obtain a target search result; based on the target search result and a preset prompt word (prompt), call a preset large model to classify and identify the application type of the target application to obtain the target application type; and save the target application type in a classification information database. Among them, the target application to be identified is Figure 3 The target search result is used to describe the function of the target application, and the prompt word is used to indicate the candidate application type of the application.

[0082] It can be understood that the system can ensure that each application can be uniquely and accurately identified by accurately capturing the unique identity of each target application (i.e., the target application package name). When traditional methods cannot directly determine the application type, the system can call the preset large model to deeply analyze the rich information contained in the target application package name, such as the application name, function description, etc., so as to quickly and accurately infer the specific type of the target application. This not only improves the efficiency and accuracy of application type identification, but also greatly reduces the burden of manual classification.

[0083] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0084] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM (Read-Only Memory) / RAM (Random Access Memory), a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods of each embodiment of the present application.

[0085] According to another aspect of the embodiment of the present application, there is also provided an application type identification device for implementing the above-mentioned application type identification method. Figure 4 , Figure 4 is a structural block diagram of an optional application type identification device provided in an embodiment of the present application, such as Figure 4 As shown, the device may include:

[0086] An application determination module 401 is used to determine a target application to be identified;

[0087] The content search module 402 is used to search the target application using a preset search engine to obtain a target search result, wherein the target search result is used to describe the function of the target application;

[0088] The type identification module 403 is used to call the preset large model, identify the application type of the target application according to the target search result and the preset prompt word, and obtain the target application type corresponding to the target application, wherein the prompt word is used to indicate the candidate application type of the application.

[0089] It should be noted that the application determination module 401 in this embodiment can be used to execute the above step S201, the content search module 402 in this embodiment can be used to execute the above step S202, and the type identification module 403 in this embodiment can be used to execute the above step S203.

[0090] Regarding the application type identification device of the application in this embodiment, the specific manner in which its application determination module 401, content search module 402 and type identification module 403 execute the application type identification method of the above-mentioned application has been described in detail in the embodiment of the application type identification method related to the application, and will not be elaborated here.

[0091] It can be understood that the technical solution provided in this embodiment, each module in the application type identification device of the application, automatically uses a preset search engine to retrieve the functional description of the target application, and combines a large pre-trained model with preset prompt words to intelligently identify the application type, which significantly improves the efficiency and accuracy of application classification, effectively avoids the inefficiency and error-prone problems that may exist in manual identification, and realizes a fast and accurate automated application classification process.

[0092] The device in this embodiment, in addition to the above modules, may also include a module for executing any method in the embodiment of the method for identifying the application type of any of the above applications.

[0093] It should be noted that the above modules and corresponding steps implement the same examples and application scenarios, but are not limited to the contents disclosed in the above embodiments. It should be noted that the above modules, as part of the device, can be run in the implementation of Figure 1 The hardware environment of the method shown may be implemented by software or hardware, wherein the hardware environment includes a network environment.

[0094] According to another aspect of an embodiment of the present application, an electronic device for implementing the application type identification method of the above-mentioned application program is also provided. The electronic device may be a server, a terminal, or a combination thereof.

[0095] According to another embodiment of the present application, an electronic device is also provided. Figure 5 , Figure 5 is a structural block diagram of an optional electronic device provided in an embodiment of the present application, such as Figure 5 As shown, the electronic device may include: a processor 1501 , a communication interface 1502 , a memory 1503 and a communication bus 1504 , wherein the processor 1501 , the communication interface 1502 , and the memory 1503 communicate with each other via the communication bus 1504 .

[0096] Memory 1503, used for storing computer programs;

[0097] The processor 1501 is used to implement the following steps when executing the program stored in the memory 1503:

[0098] Step S201, determining a target application to be identified;

[0099] Step S202, searching for the function of the target application using a preset search engine to obtain a target search result, wherein the target search result is used to describe the function of the target application;

[0100] Step S203, calling a preset large model, identifying the application type of the target application according to the target search result and the preset prompt word, and obtaining the target application type corresponding to the target application, wherein the prompt word is used to indicate the candidate application type of the application.

[0101] It can be understood that the technical solution provided in this embodiment, the processor of the electronic device, significantly improves the efficiency and accuracy of application classification by automatically using a preset search engine to retrieve the functional description of the target application, and combines a large pre-trained model with preset prompt words to intelligently identify the application type, effectively avoiding the inefficiency and error-prone problems that may exist in manual identification, and realizing a fast and accurate automated application classification process.

[0102] Optionally, in this embodiment, the above-mentioned communication bus can be a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus. The communication interface is used for communication between the above-mentioned electronic device and other devices.

[0103] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0104] The above-mentioned processor can be a general-purpose processor, which can include but not be limited to: CPU (Central Processing Unit), NP (Network Processor), etc.; it can also be DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0105] An embodiment of the present application further provides a computer-readable storage medium, the storage medium including a stored program, wherein the method steps of the above method embodiment are executed when the program is run.

[0106] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media that can store program codes, such as a USB flash drive, a ROM, a RAM, a mobile hard disk, a magnetic disk, or an optical disk.

[0107] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0108] If the integrated units in the above embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in the above computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling one or more computer devices (which can be personal computers, servers or network devices, etc.) to execute all or part of the steps of the methods of each embodiment of the present application.

[0109] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0110] In the several embodiments provided in the present application, it should be understood that the disclosed client can be implemented in other ways. Among them, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0111] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution provided in this embodiment.

[0112] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0113] The above are only preferred implementations of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for identifying an application type of an application, characterized in that: include: Determine the target application to be identified; Using a preset search engine to search for the function of the target application to obtain a target search result, wherein the target search result is used to describe the function of the target application; The preset large model is called to identify the application type of the target application according to the target search result and the preset prompt word to obtain the target application type corresponding to the target application, wherein the prompt word is used to indicate the candidate application type of the application.

2. The method for identifying the application type of an application according to claim 1, characterized in that: The calling of the preset macro model, identifying the application type of the target application according to the target search result and the preset prompt word, and obtaining the target application type corresponding to the target application includes: Preprocessing the text corresponding to the target search result by using the large model to obtain multiple words; Convert each of the multiple words into an embedding vector to obtain a set of embedding vectors corresponding to the multiple words; Perform feature extraction on all embedded vectors in the embedded vector set through the neural network structure of the large model to obtain text features, wherein the text features are used to characterize the content described by the target search result; Calling the classifier of the large model to calculate the probability value corresponding to each candidate application type in the prompt word according to the text features and the knowledge learned by the large model during the training process; The probability values ​​corresponding to all candidate application types are traversed through the large model, and the candidate application type with the highest probability is selected as the target application type corresponding to the target application program.

3. The method for identifying the application type of an application according to claim 1, characterized in that: The using a preset search engine to search for the function of the target application to obtain target search results includes: Use the target application package name of the target application to combine with any one of the preset multiple search keywords respectively, so as to construct multiple search statements, wherein the target application package name is a unique identifier of the target application; Inputting the multiple search statements into the search engine to perform a search, and obtaining multiple initial search results corresponding to the multiple search statements; Performing data cleaning on the multiple initial search results to obtain initial text data; Extracting target text data related to the application type from the initial text data; All target text data are aggregated to obtain the target search result.

4. The method for identifying the application type of an application according to claim 3, characterized in that: The multiple search keywords include application details, application type, function description and function introduction.

5. The method for identifying the application type of an application according to claim 1, characterized in that: After obtaining the target application type corresponding to the target application, the following steps are included: The target application package name of the target application program and the corresponding target application type are stored in a preset classification information database.

6. The method for identifying the application type of an application according to claim 1, characterized in that: Before using a preset search engine to search for the function of the target application to obtain the target search result, the method includes: Using a preset classification information database, determining whether the target application type of the target application can be identified, wherein the classification information database is used to store application types of the application; When a target application type capable of identifying the target application is determined according to the classification information database, an application type corresponding to an application package name identical to the target application package name in the classification information database is used as the target application type corresponding to the target application.

7. The method for identifying the application type of an application according to claim 6, characterized in that: The using a preset classification information database to determine whether the target application type of the target application can be identified includes: Querying the classification information database to determine whether there is an application package name identical to the target application package name of the target application in the classification information database; If the classification information database contains an application package name identical to the target application package name, it is determined that the application type of the target application can be identified; otherwise, it is determined that the application type of the target application cannot be identified.

8. An application type identification device for an application, characterized in that: include: An application determination module, used to determine a target application to be identified; A content search module, used to search the target application using a preset search engine to obtain a target search result, wherein the target search result is used to describe the function of the target application; The type identification module is used to call a preset large model, identify the application type of the target application according to the target search results and preset prompt words, and obtain the target application type corresponding to the target application, wherein the prompt word is used to indicate the candidate application type of the application.

9. An electronic device comprising a processor, a communication interface, a memory and a communication bus, wherein: The processor, the communication interface and the memory communicate with each other via the communication bus, wherein: The memory is used to store computer programs; The processor is configured to execute the application type identification method of any one of claims 1 to 7 by running the computer program stored in the memory.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, wherein the computer program is configured to execute the method for identifying the application type of the application program described in any one of claims 1 to 7 when running.

Images