Application recommendation method and device, computer device and storage medium

Abstract

The application relates to an application program recommendation method and device, computer equipment and a storage medium. The application relates to the technical field of artificial intelligence. The method comprises the following steps: acquiring current environment information of a user, current state information of the user, current application program demand information of the user and a plurality of to-be-recommended application program information; based on the current environment information, the current application program demand information of the user and the current state information of the user, analyzing a target application program theme preferred by the user at present, and screening initial target application program information corresponding to the target application program theme from the application program information; based on the initial target application program information, establishing an application program recommendation graph network, and based on the application program recommendation graph network, recommending target application program information from the initial target application program information corresponding to the target application program theme. The method can improve the accuracy of recommending application program information to the user.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to an application recommendation method, apparatus, computer device, and storage medium. Background Technology

[0002] With the diversification of applications, various categories and types of applications have emerged. Even within the same application, the themes and categories vary, and users' needs for applications differ due to their different environments and personal circumstances. Therefore, how to intelligently recommend applications to users is a current research focus.

[0003] Traditional application recommendation methods respond to users' personalized application requests on the client side, combining this with contextual information from the textual information of various applications corresponding to those requests to recommend applications to the user. However, this method only considers the textual information associated with the application, resulting in low accuracy in recommending applications to the user. Summary of the Invention

[0004] Therefore, it is necessary to provide an application recommendation method, apparatus, computer device, computer-readable storage medium, and computer program product to address the aforementioned technical problems.

[0005] Firstly, this application provides an application recommendation method. The method includes:

[0006] The system obtains the user's current environment information, the user's current state information, the user's current application needs information, and information on multiple applications to be recommended.

[0007] Based on the current environment information, the user's current application requirements information, and the user's current state information, the target application theme currently preferred by the user is analyzed, and the initial target application information corresponding to the target application theme is filtered from each application information.

[0008] Based on the initial target application information, an application recommendation graph network is established, and based on the application recommendation graph network, target application information is recommended from the initial target application information corresponding to the target application topic.

[0009] Optionally, obtaining information on multiple applications to be recommended includes:

[0010] Obtain the user's historical application information; the historical application information includes multiple initial application information and the browsing frequency of each initial application information;

[0011] Filter the initial application information corresponding to browsing frequencies greater than a preset browsing frequency threshold, and use them as the first candidate application information;

[0012] Using an application similarity identification strategy, first candidate application information that meets the application similarity criteria is selected from the application information database as second candidate application information, and each of the first candidate application information and each of the second candidate application information is selected as application information to be recommended.

[0013] Optionally, the step of analyzing the user's current preferred target application topic based on the current environment information, the user's current application needs information, and the user's current state information includes:

[0014] Based on the current environment information, the user's current application requirements information, and the user's current state information, the current application state corresponding to the user is analyzed through a state analysis strategy, and the target application theme is determined based on the current application state corresponding to the user.

[0015] Optionally, the step of analyzing the current application state corresponding to the user based on the current environment information, the user's current application requirement information, and the user's current state information, through a state analysis strategy, includes:

[0016] Extract environmental factor information of different categories from the current environmental information, and identify the first application status factor value corresponding to each category of environmental factor information;

[0017] Based on the user status information, the user's personal emotional state and the degree value of the personal emotional state are identified, and a second application status factor corresponding to the user's personal emotional state is selected from each historical application status factor. Based on the degree value of the personal emotional state, the second application status factor value of each second application status factor is determined.

[0018] Based on the user's current application requirement information, the user's third application status factor value is determined, and based on each of the first application status factor values, each of the second application status factor values, and each of the third application status factor values, the user's corresponding current application status is determined.

[0019] Optionally, determining the target application topic based on the current application state corresponding to the user includes:

[0020] Obtain the application status factors for each of multiple application topics, as well as the overall application status factor value for each application topic;

[0021] Based on the application status factors of each application topic and the application status factors of the current application state corresponding to the user, among the application topics, the application topics that are the same as the application status factors of the current application state corresponding to the user are selected as the initial target application topics.

[0022] The weighted summation of each application state factor value corresponding to the current application state of the user is performed to obtain the comprehensive application state factor value corresponding to the current application state of the user, and the factor difference between the comprehensive application state factor value of the user and the comprehensive application state factor value of each initial target application theme is calculated respectively.

[0023] Select the initial target application topics whose factor differences are lower than the preset factor difference threshold as the target application topics.

[0024] Optionally, the application information includes vendor information, list information, and category information. The step of establishing an application recommendation graph network based on the initial target application information includes:

[0025] For each pair of initial target application information, a first similarity factor is determined based on the number of common manufacturer information in the manufacturer information of the two initial target application information; a second similarity factor is determined based on the number of common category information of the two initial target application information; and a third similarity factor is determined based on the number of common list information of the two initial target application information.

[0026] Linear weighting is applied to the first similarity factor, the second similarity factor, and the third similarity factor of each of the two initial target application information to obtain the application recommendation graph network corresponding to each of the initial target application information.

[0027] Optionally, before performing linear weighting on the first similarity factor, the second similarity factor, and the third similarity factor of each of the two initial target application information, the method further includes:

[0028] Obtain the first initial weight value of the first similarity factor of each of the two initial target application information, the second initial weight value of the second similarity factor of each of the two initial target application information, and the third initial weight value of the third similarity factor of each of the two initial target application information, and obtain the topic content information of the target application topic;

[0029] Based on the topic content information of the target application topic, the first initial weight value, the second initial weight value, and the third initial weight value are adjusted respectively through a weight adjustment function to obtain the first weight value of the first similarity factor of each of the two initial target application information, the second weight value of the second similarity factor of each of the two initial target application information, and the third weight value of the third similarity factor of each of the two initial target application information.

[0030] Optionally, the step of recommending target application information based on the application recommendation graph network, from each of the initial target application information corresponding to the target application topic, includes:

[0031] The application recommendation graph network takes the two initial target application information and, based on the first weight value, the second weight value, and the third weight value, performs a weighted fusion process on the first similarity factor, the second similarity factor, and the third similarity factor of the two initial target application information to obtain the similarity of the two initial target application information.

[0032] Among the initial target application information, the initial target application information with an average similarity greater than a first similarity threshold is selected as the first target application information, and the application status information corresponding to each first target application information is identified.

[0033] Calculate the application status information corresponding to each first target application information and its similarity to the user's current application status information, and filter the first target application information that is greater than a second similarity threshold as the target application information.

[0034] Secondly, this application also provides an application recommendation device. The device includes:

[0035] The acquisition module is used to acquire information about the user's current environment, the user's current state, the user's current application needs, and information about multiple applications to be recommended.

[0036] The topic determination module is used to analyze the target application topic currently preferred by the user based on the current environment information, the user's current application requirement information, and the user's current state information, and to filter the initial target application information corresponding to the target application topic from each application information.

[0037] The recommendation module is used to establish an application recommendation graph network based on the initial target application information, and to recommend target application information from the initial target application information corresponding to the target application topic based on the application recommendation graph network.

[0038] Optionally, the acquisition module is specifically used for:

[0039] Obtain the user's historical application information; the historical application information includes multiple initial application information and the browsing frequency of each initial application information;

[0040] Filter the initial application information corresponding to browsing frequencies greater than a preset browsing frequency threshold, and use them as the first candidate application information;

[0041] Using an application similarity identification strategy, first candidate application information that meets the application similarity criteria is selected from the application information database as second candidate application information, and each of the first candidate application information and each of the second candidate application information is selected as application information to be recommended.

[0042] Optionally, the topic determination module is specifically used for:

[0043] Based on the current environment information, the user's current application requirements information, and the user's current state information, the current application state corresponding to the user is analyzed through a state analysis strategy, and the target application theme is determined based on the current application state corresponding to the user.

[0044] Optionally, the topic determination module is specifically used for:

[0045] Extract environmental factor information of different categories from the current environmental information, and identify the first application status factor value corresponding to each category of environmental factor information;

[0046] Based on the user status information, the user's personal emotional state and the degree value of the personal emotional state are identified, and a second application status factor corresponding to the user's personal emotional state is selected from each historical application status factor. Based on the degree value of the personal emotional state, the second application status factor value of each second application status factor is determined.

[0047] Based on the user's current application requirement information, the user's third application status factor value is determined, and based on each of the first application status factor values, each of the second application status factor values, and each of the third application status factor values, the user's corresponding current application status is determined.

[0048] Optionally, the topic determination module is specifically used for:

[0049] Obtain the application status factors for each of multiple application topics, as well as the overall application status factor value for each application topic;

[0050] Based on the application status factors of each application topic and the application status factors of the current application state corresponding to the user, among the application topics, the application topics that are the same as the application status factors of the current application state corresponding to the user are selected as the initial target application topics.

[0051] The weighted summation of each application state factor value corresponding to the current application state of the user is performed to obtain the comprehensive application state factor value corresponding to the current application state of the user, and the factor difference between the comprehensive application state factor value of the user and the comprehensive application state factor value of each initial target application theme is calculated respectively.

[0052] Select the initial target application topics whose factor differences are lower than the preset factor difference threshold as the target application topics.

[0053] Optionally, the recommendation module is specifically used for:

[0054] For each pair of initial target application information, a first similarity factor is determined based on the number of common manufacturer information in the manufacturer information of the two initial target application information; a second similarity factor is determined based on the number of common category information of the two initial target application information; and a third similarity factor is determined based on the number of common list information of the two initial target application information.

[0055] Linear weighting is applied to the first similarity factor, the second similarity factor, and the third similarity factor of each of the two initial target application information to obtain the application recommendation graph network corresponding to each of the initial target application information.

[0056] Optionally, the device further includes:

[0057] The weight acquisition module is used to acquire the first initial weight value of the first similarity factor of each of the two initial target application information, the second initial weight value of the second similarity factor of each of the two initial target application information, and the third initial weight value of the third similarity factor of each of the two initial target application information, and to acquire the topic content information of the target application topic;

[0058] The weight adjustment module is used to adjust the first initial weight value, the second initial weight value, and the third initial weight value respectively based on the topic content information of the target application topic through a weight adjustment function, so as to obtain the first weight value of the first similarity factor of each of the two initial target application information, the second weight value of the second similarity factor of each of the two initial target application information, and the third weight value of the third similarity factor of each of the two initial target application information.

[0059] Optionally, the recommendation module is specifically used for:

[0060] The application recommendation graph network takes the two initial target application information and, based on the first weight value, the second weight value, and the third weight value, performs a weighted fusion process on the first similarity factor, the second similarity factor, and the third similarity factor of the two initial target application information to obtain the similarity of the two initial target application information.

[0061] Among the initial target application information, the initial target application information with an average similarity greater than a first similarity threshold is selected as the first target application information, and the application status information corresponding to each first target application information is identified.

[0062] Calculate the application status information corresponding to each first target application information and its similarity to the user's current application status information, and filter the first target application information that is greater than a second similarity threshold as the target application information.

[0063] Thirdly, this application provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method described in any one of the first aspects.

[0064] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0065] Fifthly, this application provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0066] The aforementioned application recommendation method, apparatus, computer device, storage medium, and computer program product acquire information about the user's current environment, current state, current application needs, and multiple applications to be recommended. Based on the current environment, current application needs, and current state, they analyze the user's current preferred target application topics and filter initial target application information corresponding to the target application topics from among the application information. Based on the initial target application information, they establish an application recommendation graph network and recommend target application information from among the initial target application information corresponding to the target application topics. By comprehensively analyzing the current environment, current state, and current application needs, the user's current preferred target application topics are determined. Based on the application information of these target application topics, an application recommendation graph network is established, and then target application information is recommended to the user through this network. This avoids recommending applications solely based on textual information, thus improving the accuracy of application recommendations. Attached Figure Description

[0067] Figure 1 This is a flowchart illustrating an application recommendation method in one embodiment;

[0068] Figure 2 A flowchart illustrating an application recommendation example in one embodiment;

[0069] Figure 3 This is a structural block diagram of an application recommendation device in one embodiment;

[0070] Figure 4 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0072] The application recommendation method provided in this application can be applied to terminals, servers, and systems including both terminals and servers, and is implemented through interaction between the terminal and the server. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, and tablets. The server can be a standalone server or a server cluster composed of multiple servers. The terminal determines the user's current preferred target application topic by comprehensively analyzing current environmental information, the user's current state information, and the user's current application needs information. Based on the application information of this target application topic, it establishes an application recommendation graph network, and then recommends target application information to the user through this application recommendation graph network. This avoids recommending applications to the user solely based on textual information about applications, thus improving the accuracy of application recommendations.

[0073] In one embodiment, such as Figure 1 As shown, an application recommendation method is provided. Taking the application of this method to a terminal as an example, the method includes the following steps:

[0074] Step S101: Obtain current environment information, user's current status information, user's current application requirements information, and information on multiple applications to be recommended.

[0075] In this embodiment, the terminal acquires current environmental information based on an environmental information acquisition program. This environmental information includes weather information, temperature information, time information, location information, and, if the user is in a vehicle, vehicle speed information and vehicle information (vehicle information refers to the type of vehicle, such as a car, bus, subway, ship, or airplane). With user authorization, the terminal, in response to user input, acquires the user's current state information, including mood, physical state, and the user's current environment (e.g., work, vacation, busy, leisure). In response to user input, the terminal acquires the user's current application request information, which is the user's request for application categories, and this request can be, but is not limited to, a single application category. Based on the user's historical application information and an application information database, the terminal acquires multiple applications to be recommended. The specific process for acquiring these applications will be described in detail later. These applications can be, but are not limited to, financial applications, lifestyle applications, and entertainment applications. Among them, financial applications may include, but are not limited to, wealth management applications, loan applications, financial product applications, and bank auxiliary applications; lifestyle applications may include, but are not limited to, food and beverage applications, medical applications, leisure applications, and navigation applications; and entertainment applications may include, but are not limited to, music applications, video applications, game applications, image applications, and document applications.

[0076] Step S102: Based on the current environment information, the user's current application requirements information, and the user's current state information, analyze the target application topics currently preferred by the user, and filter the initial target application information corresponding to the target application topics from each application information.

[0077] In this embodiment, the terminal analyzes application state factors in the current environmental information, analyzes application state factors corresponding to the user's state information, and predicts the current application state required by the user based on the application state factors corresponding to the user's current application needs. Based on this current application state, the terminal filters the target application theme corresponding to the current application state from various application themes. The target application theme may not be limited to one. The application state is a comprehensive state information obtained by combining multiple application state factors, including states such as relaxation, excitement, depression, tension, blankness, fear, and exhilaration. Application state factors are the information of each factor that constitutes the application state; for example, factors for a tense state include repression and quietness, while factors for a calm state include time, space, and relaxation. Then, the terminal filters the application information corresponding to the target application theme from the application information to be filtered, using this as the initial target application information. The current application needs information includes needs information containing application categories, such as application types, work assistance applications, life convenience applications, and entertainment applications. The specific process of analyzing the current application state will be explained in detail later.

[0078] Step S103: Based on the initial target application information, establish an application recommendation graph network, and based on the application recommendation graph network, recommend target application information from the initial target application information corresponding to the target application topic.

[0079] In this embodiment, the terminal establishes an application recommendation graph network based on the similarity factors among the selected initial target application information. Then, based on this application recommendation graph network, the terminal selects initial target application information from the initial target application information that has a similarity greater than a first similarity threshold and a similarity greater than a second similarity threshold with the user's current application state, and uses these as target application information. The specific process of establishing the application recommendation graph network and recommending target application information will be explained in detail later.

[0080] Based on the above scheme, by comprehensively analyzing the current environmental information, the user's current state information, and the user's current application needs information, the target application theme currently preferred by the user is determined. Based on the application information of each application in the target application theme, an application recommendation graph network is established. Then, the target application information is recommended to the user through the application recommendation graph network, which avoids recommending applications to the user based solely on the text information of the application information and improves the accuracy of recommending application information to the user.

[0081] Optionally, multiple application information to be recommended can be obtained, including: obtaining the user's historical application information and an application information database; the historical application information includes multiple initial application information and the browsing frequency of each initial application information; initial application information with browsing frequencies greater than a preset browsing frequency threshold is selected as first candidate application information; and application information similar to each first candidate application information is selected from the application information database as second candidate application information through an application similarity identification strategy, and each first candidate application information and each second candidate application information is selected as application information to be recommended.

[0082] In this embodiment, with user authorization, the terminal obtains the user's historical application information. This historical application information includes multiple initial application information entries and the browsing frequency of each initial application entry, as well as a list of the user's preferred applications, a list of applications browsed by the user, and a list of applications favorited by the user. The terminal obtains a database of all copyrighted applications within the merchant's applications, creating an application information database. The terminal presets a browsing frequency threshold and filters application information from the historical application information with browsing frequencies exceeding the preset threshold as first candidate application information. Then, for each first candidate application information entry, the terminal calculates the text similarity between the text information of that first candidate application information and the text information of each application information entry in the application database, and uses the application information in the application database with a similarity greater than the similarity threshold as second candidate application information. Similarly, through the above scheme, the terminal obtains second candidate application information entries with similarity greater than the similarity threshold for each first candidate application information entry. Finally, the terminal uses each first candidate application information entry and each second candidate application information entry as application information to be recommended. Specifically, when the application is a financial product application, the browsing frequency is the number of times a user views the webpage of that financial product application per unit of time (e.g., one day, 15 days, one month, etc.). When the application is an entertainment product application, the browsing frequency is the number of times a user plays various entertainment product applications per unit of time (e.g., one day, 15 days, one month, etc.). Examples include the number of times a music product is played in a day, the number of clicks on a video product in a day, and the number of visits to an image product in a day.

[0083] Based on the above scheme, by filtering frequently viewed application information from the user's history list and filtering application information with high text similarity to the frequently viewed application information from the database, the accuracy of the application information to be recommended is improved.

[0084] Optionally, based on current environment information, user's current application requirements information, and user's current state information, the target application theme currently preferred by the user can be analyzed, including: based on current environment information, user's current application requirements information, and user's current state information, using a state analysis strategy, analyzing the current application state corresponding to the user, and determining the target application theme based on the current application state corresponding to the user.

[0085] In this embodiment, the terminal analyzes application state factors in the current environment information, analyzes application state factors corresponding to user state information, and determines the values ​​of each application state factor (i.e., state analysis strategy) based on the application state factors corresponding to the user's current application needs information. The terminal then uses the user's current application state factors and their values ​​as the user's current application state. Next, based on the application state factor values ​​of the current application state and the application state factor values ​​of each application theme, the terminal determines the target application theme. The specific process of determining the application theme will be explained in detail later.

[0086] Based on the above scheme, by analyzing the current environment information, the user's current application needs information, and the user's current status information, the user's current application status is determined, thereby filtering target application topics and improving the accuracy of recommending target application topics to users.

[0087] Optionally, based on current environmental information, the user's current application needs information, and the user's current state information, a state analysis strategy is used to analyze the user's current application state, including: extracting different categories of environmental factor information from the current environmental information and identifying the first application state factor value corresponding to each category of environmental factor information; based on the user's state information, identifying the user's personal emotional state and the degree value of the personal emotional state, and filtering the second application state factors corresponding to the user's personal emotional state from each historical application state factor, and determining the second application state factor value of each second application state factor based on the degree value of the personal emotional state; based on the user's current application needs information, determining the user's third application state factor value, and determining the user's current application state based on each first application state factor value, each second application state factor value, and each third application state factor value.

[0088] In this embodiment, the terminal presets application state factors corresponding to the categories of various environmental factors, extracts different categories of environmental factor information from the current environmental information, and for each environmental factor, the terminal determines the corresponding application state factor based on the category of that environmental factor. The terminal then determines the first application state factor value corresponding to that environmental factor information. Here, the environmental factor refers to different sub-environmental information within the current environmental information. These environmental factors include weather, temperature, time, location, etc. The application state factors corresponding to each category of environmental factor are different. For example, the application state factor corresponding to the weather factor is the application climate factor, the application state factor corresponding to the temperature factor is the application rhythm factor, the application state factor corresponding to the time factor is the application emotion factor, and the application state factor corresponding to the location factor is the application regional factor. The application climate factor includes applications containing different climate types, the application rhythm factor includes applications containing different application rhythms, the application emotion factor includes applications containing different human emotional experiences, and the application regional factor includes applications containing different regional styles and characteristics.

[0089] With user authorization, the terminal responds to the user's emotional state input by acquiring the user's mood, physical state, and environmental state (e.g., work status, vacation status, busy status, leisure status, etc.). The terminal presets the emotional range corresponding to each emotional state for each user's state information, and analyzes the user's current emotional state based on each state information to obtain the user's personal emotional state. For example, the range corresponding to a depressed mood state is: mood 1-3 points, physical condition 5-7 points, and environmental state: working or busy; the range corresponding to an excited mood state is: mood 7-10 points, physical condition 5-7 points, and environmental state: vacation or leisure; the range corresponding to a relaxed mood state is: mood 5-7 points, physical condition 3-6 points, and environmental state: leisure; the range corresponding to a depressed mood state is: mood 0-1 points, physical condition 2-3 points, and environmental state: busy. If user A has a mood state of 3 points, a physical condition of 7 points, and is in a working environment, then the terminal determines that user A's current mood state is depressed. If user B has a mood state of 5 points, a physical condition of 6 points, and is in a leisure environment, then the terminal determines that user B's current mood state is relaxed. The terminal uses each state information as a degree value for the current mood state. The terminal presets application state factors corresponding to each emotional state. Based on the user's personal emotional state, it queries the second application state factor corresponding to that emotional state and normalizes each state information to obtain the values ​​of each sub-application state factor of the second application state factor. The terminal presets weight values ​​for each state information and performs a weighted summation of the sub-application state factor values ​​corresponding to each state information to obtain the second application state factor value. Then, based on the user's current application needs information, the terminal analyzes the application state factors corresponding to the current application needs information and determines the application state factor value corresponding to each application state factor to obtain the values ​​of each third application state factor. The terminal uses each first application state factor, each second application state factor, each third application state factor, and each first application state factor value, each second application state factor value, and each third application state factor value as the user's corresponding current application state.

[0090] Based on the above approach, by considering environmental factors and user emotional factors, the accuracy of application recommendations to users has been improved.

[0091] Optionally, based on the current application state corresponding to the user, the target application topic is determined, including: obtaining the application state factors of each application topic and the comprehensive application state factor value; based on the application state factors of each application topic and the application state factors of the current application state corresponding to the user, among the application topics, selecting application topics that have the same application state factors as the current application state corresponding to the user as initial target application topics; performing weighted summation on the application state factor values ​​of the current application state corresponding to the user to obtain the comprehensive application state factor value of the user corresponding to the current application state, and calculating the factor difference between the comprehensive application state factor value of the user and the comprehensive application state factor value of each initial target application topic; selecting the initial target application topics with factor differences lower than a preset factor difference threshold as target application topics.

[0092] In this embodiment, the terminal responds to the input operation of the staff, obtaining the application status factors contained in each application topic, and the comprehensive application status factor value of each application topic. The comprehensive application status factor value is obtained by weighted summation of all application status factor values. The terminal queries each application topic that contains the same application status factors as the current application state corresponding to the user, and uses this as the initial target application topic. The terminal presets weight values ​​for environmental information and user state information, and based on these weight values, performs weighted summation on the first application status factor values ​​corresponding to the environmental information and the second application status factor values ​​corresponding to the user state information to obtain the comprehensive application status factor value for the user's current application state. The terminal calculates the factor difference between the user's comprehensive application status factor value and the comprehensive application status factor values ​​of each initial target application topic, and presets a factor difference threshold. The terminal selects the initial target application topics with factor differences lower than the preset factor difference threshold as target application topics.

[0093] Based on the above scheme, the accuracy of the selected application topics is improved by filtering application topics using factor differences and application status factors.

[0094] Optionally, the application information includes vendor information, list information, and category information. Based on each initial target application information, an application recommendation graph network is established, including: for each pair of initial target application information, determining a first similarity factor based on the number of vendor information they share in each vendor information, determining a second similarity factor based on the number of categories they share, and determining a third similarity factor based on the number of lists they share. The first, second, and third similarity factors of each pair of initial target application information are then linearly weighted to obtain the application recommendation graph network corresponding to each initial target application information.

[0095] In this embodiment, the terminal extracts the manufacturer information, list information, and category information for each initial target application. The manufacturer information refers to the company that generated the initial target application information; there can be one or more such companies. The list information refers to the number of times the application information was added to the user's browsing list, and the time of each addition. The category information refers to the list to which the initial target application information belongs; there can be one or more such lists. For each pair of initial target application information, the terminal determines a first similarity factor based on the number of common manufacturer information in their respective manufacturer information. Secondly, the terminal determines a second similarity factor based on the number of common category information. Finally, the terminal determines a third similarity factor based on the number of common list information. The terminal performs linear weighting on the first, second, and third similarity factors of each pair of initial target application information to obtain the application recommendation graph network corresponding to each initial target application information. Specifically, the application's recommendation graph network can be a homogeneous network G(S), with the following weight parameters:

[0096] W(s1,s2)=PW*Num(s1→P→s2)+AW*Num(s1→A→s2)+MW*Num

[0097] (s1→M→s2)

[0098] In the above formula, s1 and s2 are two initial target application information, PW is the weight of P, AW is the weight of A, and MW is the weight of M; Num(s1→P→s2) is the number of common vendor information in the vendor information of the two initial target application information; Num(s1→A→s2) is the number of information of the same category in the two initial target application information; Num(s1→M→s2) is the number of information of the same list in the two initial target applications, where P is the vendor information factor, A is the category information factor, and M is the list information factor.

[0099] Based on the above scheme, by analyzing the vendor information factors of each pair of initial target application information, the list information factors of each pair of initial targets, and the list information factors of each pair of initial targets, the application recommendation graph network of the initial target application information is determined, thereby improving the accuracy of subsequent application recommendation information.

[0100] Optionally, before performing linear weighting on the first similarity factor, the second similarity factor, and the third similarity factor of each pair of initial target application information, the method further includes: obtaining the first initial weight value of the first similarity factor, the second initial weight value of the second similarity factor, and the third initial weight value of the third similarity factor of each pair of initial target application information, and obtaining the topic content information of the target application topic; based on the topic content information of the target application topic, adjusting the first initial weight value, the second initial weight value, and the third initial weight value respectively through a weight adjustment function to obtain the first weight value of the first similarity factor, the second weight value of the second similarity factor, and the third weight value of the third similarity factor of each pair of initial target application information.

[0101] In this embodiment, the terminal acquires a first initial weight value for a first similarity factor, a second initial weight value for a second similarity factor, and a third initial weight value for a third similarity factor for each pair of initial target application information. The first, second, and third initial weight values ​​are preset by the operator on the terminal. The terminal acquires the topic content information of the target application theme, and then, based on this topic content information, adjusts the first, second, and third initial weight values ​​using a weight adjustment function to obtain the first weight value for the first similarity factor, the second weight value for the second similarity factor, and the third weight value for the third similarity factor for each pair of initial target application information.

[0102] The weight adjustment function is as follows:

[0103]

[0104] In the above formula, G t R is the topic content information obtained by the terminal under different application topics t. t Indicates the recommendation method ρ through the application. t Recommended target application topic content information. PW, AW, and MW are the weights of P, A, and M, respectively. P represents the vendor information factor, A represents the category information factor, and M represents the list information factor. The search space of parameters PW, AW, and MW is uniformly distributed between 0 and 1.

[0105] Based on the above scheme, the recommendation accuracy of the application recommendation graph network is improved by adjusting the second weight value of the second similarity factor of each pair of initial target application information and the third weight value of the third similarity factor of each pair of initial target application information through the weight adjustment function.

[0106] Optionally, based on the application recommendation graph network, target application information is recommended from the initial target application information corresponding to the target application topic. This includes: taking each pair of initial target application information from the application recommendation graph network, and performing weighted fusion processing on the first similarity factor, the second similarity factor, and the third similarity factor of each pair of initial target application information based on the first weight value, the second weight value, and the third weight value, to obtain the similarity of each pair of initial target application information; among the initial target application information, selecting initial target application information whose average similarity with other initial target application information is greater than the first similarity threshold as the first target application information, and identifying the application status information corresponding to each first target application information; calculating the similarity between the application status information corresponding to each first target application information and the user's current application status information, and selecting the first target application information with a similarity greater than the second similarity threshold as the target application information.

[0107] In this embodiment, the terminal takes the initial target application information from each pair of application recommendation graph network and performs weighted fusion processing on the first similarity factor, the second similarity factor, and the third similarity factor of each pair of initial target application information based on a first weight value, a second weight value, and a third weight value to obtain the similarity of each pair of initial target application information. The terminal presets a first similarity threshold, and then selects initial target application information whose average similarity with other initial target application information is greater than the first similarity threshold as the first target application information. The terminal presets a second similarity threshold and identifies the application status information corresponding to each first target application information. The terminal calculates the similarity between the application status information corresponding to each first target application information and the user's current application status information, and selects first target application information whose similarity is greater than the second similarity threshold as the target application information.

[0108] Based on the above scheme, target application information is recommended to users through the application recommendation graph network, which avoids recommending applications to users based solely on textual information about the applications, and improves the accuracy of application information recommendations to users.

[0109] In one embodiment, such as Figure 2 As shown, an application recommendation method is provided. Taking the application of this method to a terminal as an example, the method includes the following steps:

[0110] Step S201: Obtain the user's current environment information, the user's current state information, and the user's current application requirements information.

[0111] Step S202: Obtain the user's historical application information.

[0112] Step S203: Filter the initial application information corresponding to browsing frequencies greater than the preset browsing frequency threshold as the first candidate application information.

[0113] Step S204: Using an application similarity identification strategy, first candidate application information that meets the application similarity criteria is selected from the application information database as second candidate application information, and each first candidate application information and each second candidate application information is selected as application information to be recommended.

[0114] Step S205: Extract environmental factor information of different categories of current environmental information respectively, and identify the first application state factor value corresponding to each category of environmental factor information.

[0115] Step S206: Based on user status information, identify the user's personal emotional state and the degree value of the personal emotional state, and filter the second application status factors corresponding to the user's personal emotional state from each historical application status factor. Based on the degree value of the personal emotional state, determine the second application status factor value of each second application status factor.

[0116] Step S207: Based on the user's current application requirement information, determine the user's third application status factor value, and based on each first application status factor value, each second application status factor value, and each third application status factor value, determine the user's corresponding current application status.

[0117] Step S208: Obtain the application status factors of each application topic and the comprehensive application status factor value of each application topic.

[0118] Step S209: Based on the application status factors of each application topic and the application status factors of the current application state corresponding to the user, select the application topics that are the same as the application status factors of the current application state corresponding to the user as the initial target application topics.

[0119] Step S210: Perform weighted summation on the application state factor values ​​of the current application state corresponding to the user to obtain the comprehensive application state factor value of the user corresponding to the current application state, and calculate the factor difference between the comprehensive application state factor value of the user and the comprehensive application state factor value of each initial target application theme.

[0120] Step S211: Select the initial target application topics corresponding to factor differences that are lower than the preset factor difference threshold as the target application topics.

[0121] Step S212: For each pair of initial target application information, based on the number of common manufacturer information in the manufacturer information of each pair of initial target application information, determine the first similarity factor of the two initial target application information; based on the number of common category information of the two initial target application information, determine the second similarity factor of the two initial target application information; and based on the number of common list information of the two initial target application information, determine the third similarity factor of the two initial target application information.

[0122] Step S213: Obtain the first initial weight value of the first similarity factor of each pair of initial target application information, the second initial weight value of the second similarity factor of each pair of initial target application information, and the third initial weight value of the third similarity factor of each pair of initial target application information, and obtain the topic content information of the target application topic.

[0123] Step S214: Based on the topic content information of the target application topic, the first initial weight value, the second initial weight value, and the third initial weight value are adjusted respectively through the weight adjustment function to obtain the first weight value of the first similarity factor of each pair of initial target application information, the second weight value of the second similarity factor of each pair of initial target application information, and the third weight value of the third similarity factor of each pair of initial target application information.

[0124] Step S215: Perform linear weighting on the first similarity factor, the second similarity factor, and the third similarity factor of each pair of initial target application information to obtain the application recommendation graph network corresponding to each initial target application information.

[0125] Step S216: Based on the first weight value, the second weight value, and the third weight value, the initial target application information of each pair of initial target application information is weighted and fused to obtain the similarity of each pair of initial target application information.

[0126] Step S217: Among the initial target application information, filter out the initial target application information whose average similarity with other initial target application information is greater than a first similarity threshold, and use it as the first target application information, and identify the application status information corresponding to each first target application information.

[0127] Step S218: Calculate the application status information corresponding to each first target application information and its similarity to the user's current application status information, and filter the first target application information that is greater than the second similarity threshold as the target application information.

[0128] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0129] Based on the same inventive concept, this application also provides an application recommendation apparatus for implementing the application recommendation method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more application recommendation apparatus embodiments provided below can be found in the limitations of the application recommendation method described above, and will not be repeated here.

[0130] In one embodiment, such as Figure 3 As shown, an application recommendation device is provided, including: an acquisition module 310, a topic determination module 320, and a recommendation module 330, wherein:

[0131] The acquisition module 310 is used to acquire the user's current environment information, the user's current state information, the user's current application needs information, and information on multiple applications to be recommended;

[0132] The topic determination module 320 is used to analyze the target application topic currently preferred by the user based on the current environment information, the user's current application requirement information, and the user's current state information, and to filter the initial target application information corresponding to the target application topic from each application information.

[0133] The recommendation module 330 is used to establish an application recommendation graph network based on the initial target application information, and recommend target application information based on the application recommendation graph network among the initial target application information corresponding to the target application topic.

[0134] Optionally, the acquisition module 310 is specifically used for:

[0135] Obtain the user's historical application information; the historical application information includes multiple initial application information and the browsing frequency of each initial application information;

[0136] Filter the initial application information corresponding to browsing frequencies greater than a preset browsing frequency threshold, and use them as the first candidate application information;

[0137] Using an application similarity identification strategy, first candidate application information that meets the application similarity criteria is selected from the application information database as second candidate application information, and each of the first candidate application information and each of the second candidate application information is selected as application information to be recommended.

[0138] Optionally, the topic determination module 320 is specifically used for:

[0139] Based on the current environment information, the user's current application requirements information, and the user's current state information, the current application state corresponding to the user is analyzed through a state analysis strategy, and the target application theme is determined based on the current application state corresponding to the user.

[0140] Optionally, the topic determination module 320 is specifically used for:

[0141] Extract environmental factor information of different categories from the current environmental information, and identify the first application status factor value corresponding to each category of environmental factor information;

[0142] Based on the user status information, the user's personal emotional state and the degree value of the personal emotional state are identified, and a second application status factor corresponding to the user's personal emotional state is selected from each historical application status factor. Based on the degree value of the personal emotional state, the second application status factor value of each second application status factor is determined.

[0143] Based on the user's current application requirement information, the user's third application status factor value is determined, and based on each of the first application status factor values, each of the second application status factor values, and each of the third application status factor values, the user's corresponding current application status is determined.

[0144] Optionally, the topic determination module 320 is specifically used for:

[0145] Obtain the application status factors for each of multiple application topics, as well as the overall application status factor value for each application topic;

[0146] Based on the application status factors of each application topic and the application status factors of the current application state corresponding to the user, among the application topics, the application topics that are the same as the application status factors of the current application state corresponding to the user are selected as the initial target application topics.

[0147] The weighted summation of each application state factor value corresponding to the current application state of the user is performed to obtain the comprehensive application state factor value corresponding to the current application state of the user, and the factor difference between the comprehensive application state factor value of the user and the comprehensive application state factor value of each initial target application theme is calculated respectively.

[0148] Select the initial target application topics whose factor differences are lower than the preset factor difference threshold as the target application topics.

[0149] Optionally, the recommendation module 330 is specifically used for:

[0150] For each pair of initial target application information, a first similarity factor is determined based on the number of common manufacturer information in the manufacturer information of the two initial target application information; a second similarity factor is determined based on the number of common category information of the two initial target application information; and a third similarity factor is determined based on the number of common list information of the two initial target application information.

[0151] Linear weighting is applied to the first similarity factor, the second similarity factor, and the third similarity factor of each of the two initial target application information to obtain the application recommendation graph network corresponding to each of the initial target application information.

[0152] Optionally, the device further includes:

[0153] The weight acquisition module is used to acquire the first initial weight value of the first similarity factor of each of the two initial target application information, the second initial weight value of the second similarity factor of each of the two initial target application information, and the third initial weight value of the third similarity factor of each of the two initial target application information, and to acquire the topic content information of the target application topic;

[0154] The weight adjustment module is used to adjust the first initial weight value, the second initial weight value, and the third initial weight value respectively based on the topic content information of the target application topic through a weight adjustment function, so as to obtain the first weight value of the first similarity factor of each of the two initial target application information, the second weight value of the second similarity factor of each of the two initial target application information, and the third weight value of the third similarity factor of each of the two initial target application information.

[0155] Optionally, the recommendation module 330 is specifically used for:

[0156] The application recommendation graph network takes the two initial target application information and, based on the first weight value, the second weight value, and the third weight value, performs a weighted fusion process on the first similarity factor, the second similarity factor, and the third similarity factor of the two initial target application information to obtain the similarity of the two initial target application information.

[0157] Among the initial target application information, the initial target application information with an average similarity greater than a first similarity threshold is selected as the first target application information, and the application status information corresponding to each first target application information is identified.

[0158] Calculate the application status information corresponding to each first target application information and its similarity to the user's current application status information, and filter the first target application information that is greater than a second similarity threshold as the target application information.

[0159] The modules in the aforementioned application recommended device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0160] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 4 As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a recommended application method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad located on the computer device casing, or an external keyboard, touchpad, or mouse.

[0161] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0162] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in any one of the first aspects.

[0163] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0164] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0165] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0166] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0167] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0168] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An application recommendation method, characterized in that, The method includes: The system obtains the user's current environment information, the user's current state information, the user's current application needs information, and information on multiple applications to be recommended. Based on the current environment information, the user's current application requirements information, and the user's current state information, the target application theme currently preferred by the user is analyzed, and the initial target application information corresponding to the target application theme is filtered from each application information. Based on the initial target application information, an application recommendation graph network is established, and based on the application recommendation graph network, target application information is recommended from the initial target application information corresponding to the target application topic. The application information includes vendor information, list information, and category information. The step of establishing an application recommendation graph network based on the initial target application information includes: For every two initial target application information sets, a first similarity factor is determined based on the number of common vendor information among the vendor information of the two initial target application information sets. Based on the number of identical category information of the two initial target application information, a second similarity factor is determined for the two initial target application information; based on the number of identical list information of the two initial target application information, a third similarity factor is determined for the two initial target application information. Linear weighting is applied to the first similarity factor, the second similarity factor, and the third similarity factor of each of the two initial target application information to obtain the application recommendation graph network corresponding to each of the initial target application information. Before performing linear weighting on the first similarity factor, the second similarity factor, and the third similarity factor of the two initial target application information, the process further includes: Obtain the first initial weight value of the first similarity factor of each of the two initial target application information, the second initial weight value of the second similarity factor of each of the two initial target application information, and the third initial weight value of the third similarity factor of each of the two initial target application information, and obtain the topic content information of the target application topic; Based on the topic content information of the target application topic, the first initial weight value, the second initial weight value, and the third initial weight value are adjusted respectively through a weight adjustment function to obtain the first weight value of the first similarity factor of each of the two initial target application information, the second weight value of the second similarity factor of each of the two initial target application information, and the third weight value of the third similarity factor of each of the two initial target application information.

2. The method according to claim 1, characterized in that, The process of obtaining information on multiple applications to be recommended includes: Obtain the user's historical application information; the historical application information includes multiple initial application information and the browsing frequency of each initial application information; Filter the initial application information corresponding to browsing frequencies greater than a preset browsing frequency threshold, and use them as the first candidate application information; Using an application similarity identification strategy, first candidate application information that meets the application similarity criteria is selected from the application information database as second candidate application information, and each of the first candidate application information and each of the second candidate application information is selected as application information to be recommended.

3. The method according to claim 1, characterized in that, The step of analyzing the user's current preferred target application topics based on the current environment information, the user's current application needs information, and the user's current state information includes: Based on the current environment information, the user's current application requirements information, and the user's current state information, the current application state corresponding to the user is analyzed through a state analysis strategy, and the target application theme is determined based on the current application state corresponding to the user.

4. The method according to claim 3, characterized in that, The process of analyzing the current application state of the user based on the current environment information, the user's current application requirements information, and the user's current state information, using a state analysis strategy, includes: Extract environmental factor information of different categories from the current environmental information, and identify the first application status factor value corresponding to each category of environmental factor information; Based on the user status information, the user's personal emotional state and the degree value of the personal emotional state are identified, and a second application status factor corresponding to the user's personal emotional state is selected from each historical application status factor. Based on the degree value of the personal emotional state, the second application status factor value of each second application status factor is determined. Based on the user's current application requirement information, the user's third application status factor value is determined, and based on each of the first application status factor values, each of the second application status factor values, and each of the third application status factor values, the user's corresponding current application status is determined.

5. The method according to claim 4, characterized in that, Determining the target application topic based on the current application state corresponding to the user includes: Obtain the application status factors for each of multiple application topics, as well as the overall application status factor value for each application topic; Based on the application status factors of each application topic and the application status factors of the current application state corresponding to the user, among the application topics, the application topics that are the same as the application status factors of the current application state corresponding to the user are selected as the initial target application topics. The weighted summation of each application state factor value corresponding to the current application state of the user is performed to obtain the comprehensive application state factor value corresponding to the current application state of the user, and the factor difference between the comprehensive application state factor value of the user and the comprehensive application state factor value of each initial target application theme is calculated respectively. Select the initial target application topics whose factor differences are lower than the preset factor difference threshold as the target application topics.

6. The method according to claim 1, characterized in that, The method of recommending target application information based on the application recommendation graph network, from each of the initial target application information corresponding to the target application topic, includes: The application recommendation graph network takes the two initial target application information and, based on the first weight value, the second weight value, and the third weight value, performs a weighted fusion process on the first similarity factor, the second similarity factor, and the third similarity factor of the two initial target application information to obtain the similarity of the two initial target application information. Among the initial target application information, the initial target application information with an average similarity greater than a first similarity threshold is selected as the first target application information, and the application status information corresponding to each first target application information is identified. Calculate the application status information corresponding to each first target application information and its similarity to the user's current application status information, and filter the first target application information that is greater than a second similarity threshold as the target application information.

7. An application recommendation device, characterized in that, The device includes: The acquisition module is used to acquire information about the user's current environment, the user's current state, the user's current application needs, and information about multiple applications to be recommended. The topic determination module is used to analyze the target application topic currently preferred by the user based on the current environment information, the user's current application requirement information, and the user's current state information, and to filter the initial target application information corresponding to the target application topic from each application information. The recommendation module is used to establish an application recommendation graph network based on the initial target application information, and recommend target application information in the initial target application information corresponding to the target application topic based on the application recommendation graph network. The recommendation module is further configured to, for each pair of initial target application information, determine a first similarity factor based on the number of common vendor information in the vendor information of the two initial target application information, determine a second similarity factor based on the number of common category information of the two initial target application information, and determine a third similarity factor based on the number of common list information of the two initial target application information; and perform linear weighting on the first similarity factor, the second similarity factor, and the third similarity factor of each pair of initial target application information to obtain the application recommendation graph network corresponding to each pair of initial target application information. Before performing linear weighted processing on the first similarity factor, the second similarity factor, and the third similarity factor of each of the two initial target application information, the device further includes: a weight acquisition module, used to acquire a first initial weight value of the first similarity factor, a second initial weight value of the second similarity factor, and a third initial weight value of the third similarity factor of each of the two initial target application information, and to acquire the topic content information of the target application topic; and a weight adjustment module, used to adjust the first initial weight value, the second initial weight value, and the third initial weight value respectively through a weight adjustment function based on the topic content information of the target application topic, to obtain the first weight value of the first similarity factor, the second weight value of the second similarity factor, and the third weight value of the third similarity factor of each of the two initial target application information.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images