An application recommendation method, system, device, and storage medium

Abstract

This application discloses an application recommendation method, system, device, and storage medium. The method involves acquiring user data, including user behavior data and / or user information data; generating a user feature tag set based on the user data, the user feature tag set including several feature tags used to classify and characterize user attributes; and determining a set of applications to recommend to the user based on the user feature tag set and an application list. By integrating user interest information or daily usage data to form the feature tag set, and applying this user feature tag set to application recommendations in app stores, the method eliminates the need for users to manually search for specific applications to download. Instead, intelligent recommendations help applications find the users who need them. This method helps improve the efficiency of user application downloads and increases the click-through rate and download volume of applications in in-vehicle infotainment app stores.

Description

Technical Field

[0001] This application relates to the field of automotive technology, and in particular to application recommendation methods, systems, devices, and storage media. Background Technology

[0002] With the rapid development of mobile internet technology, various applications on in-vehicle infotainment systems have gradually become an indispensable part of users' driving experience and a major channel for service providers to offer various services to users. As a result, a large number of applications have been introduced into in-vehicle app stores. With the significant increase in the number of applications, however, due to the lack of effective recommendation algorithms, users can only actively search for and download applications. In this process, users need to spend a lot of time filtering and trying out applications, rather than applications actively identifying users who need them and recommending them to them, resulting in low application usage rates in app stores.

[0003] Therefore, how to recommend applications that meet users' needs and preferences to increase the click-through rate and download volume of applications in the vehicle's app store is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] To address the aforementioned issues, this application provides an application recommendation method, system, device, and storage medium to mine user preferences and recommend relevant applications to users, thereby increasing the click-through rate and download volume of applications in the vehicle infotainment app store.

[0005] To address the above problems, the technical solutions provided in this application are as follows:

[0006] The first aspect of this application provides an application recommendation method, including:

[0007] Acquire user data, including user behavior data and / or user information data;

[0008] A user feature tag set is generated based on the user data, and the user feature tag set includes several feature tags used to classify and characterize user attributes;

[0009] The set of applications to be recommended to the user is determined based on the user feature tag set and the application list. The application list includes several application sets classified according to user cluster attributes, where the user cluster attribute is a set of users with the same user feature tags.

[0010] Optionally, before determining the set of applications to recommend to the user based on the user feature tag set and the application list, the method further includes:

[0011] Obtain application data of the target application to be classified, wherein the application data is used to characterize the application attributes;

[0012] Based on the application data, determine the application list to which the target application's application data belongs, and update the application list.

[0013] Optionally, determining the application list based on the application data includes:

[0014] The initial category of the target application is determined based on the application data;

[0015] Obtain a reference rating for the target application, adjust the initial category of the application based on the reference rating, and obtain a sub-category of the target application. The reference rating consists of ratings of the target application by several reference users and the user cluster attributes of the reference users. The reference users are users whose ratings of the target application are not lower than a preset threshold.

[0016] Obtain the initial application set corresponding to the sub-category, wherein the initial application set includes several applications that conform to the sub-category;

[0017] The target application is added to the initial application set to obtain the application list.

[0018] Optionally, determining the set of applications to recommend to the user based on the user feature tag set and the application list includes:

[0019] The user cluster attribute of the user is determined based on the user feature tag set;

[0020] Retrieve the collection of applications corresponding to each user cluster attribute in the application list;

[0021] The system retrieves applications from various application sets that meet preset rating requirements, and selects a set of these applications as recommendations to the user.

[0022] Optionally, before determining the application list based on the application data, the method further includes:

[0023] Based on the set of user feature labels, users with the same feature labels are clustered to form user cluster attributes.

[0024] Optionally, generating a set of user feature tags based on the user data includes:

[0025] The acquired user data is processed into metrics to obtain metric data;

[0026] The indicator data is labeled to obtain a set of user feature tags.

[0027] Optionally, the user data includes updated data acquired within a preset time period, and the step of generating a user feature tag set based on the user data includes:

[0028] A set of user feature tags is generated based on the updated data.

[0029] A second aspect of this application provides an application recommendation system, comprising:

[0030] User data acquisition unit, used to acquire user data, the user data including user behavior data and / or user information data;

[0031] The user feature tag set generation unit is used to generate a user feature tag set based on the user data. The user feature tag set includes several feature tags used to classify and characterize user attributes.

[0032] The application set determination unit is used to determine the application set to be recommended to the user based on the user feature tag set and the application list. The application list includes several application sets classified according to user cluster attributes, where the user cluster attribute is a set of users with the same user feature tags.

[0033] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the application recommendation method described in any one of the first aspects above.

[0034] A fourth aspect of this application provides a computer-readable storage medium storing instructions that, when executed on a terminal device, cause the terminal device to perform the application recommendation method as described in any of the preceding first aspects.

[0035] Compared with the prior art, this application has the following beneficial effects:

[0036] The process involves acquiring user data, including user behavior data and / or user information data; generating a user feature tag set based on the user data, the feature tag set including several feature tags used to classify and characterize user attributes; and determining a set of applications to recommend to the user based on the user feature tag set and an application list. This method integrates user interest information or daily usage data to form the feature tag set, which is then used in application recommendations within the app store. Instead of requiring users to manually search for specific applications to download, intelligent recommendations help applications find the users who need them. This method helps improve the efficiency of user application downloads and increases the click-through rate and download volume of applications in the in-vehicle app store. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in this embodiment or the prior art, the drawings used in the description of the embodiment or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a flowchart of an application recommendation method provided in an embodiment of this application;

[0039] Figure 2 This is a schematic diagram of the user feature tag set provided in the embodiments of this application;

[0040] Figure 3 This is a structural diagram of an application recommendation system provided in an embodiment of this application. Detailed Implementation

[0041] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0042] To facilitate understanding of the technical solutions provided in the embodiments of this application, the background technology involved in the embodiments of this application will be described below.

[0043] As mentioned earlier, in recent years, with the rapid development of the mobile internet industry, the amount of information carried by the internet has also experienced explosive growth. Various mobile internet information carriers provide users with diverse ways to access information, but this has also led to information overload. Users have shifted from actively searching for internet content to passively receiving large amounts of internet subscriptions and push notifications, which has also increased the cost of information acquisition for users. In specific business scenarios, analyzing user behavior habits and providing content and services that match user preferences has become a core requirement. To address these user needs, recommendation methods and systems in the internet field have emerged.

[0044] The method provided in this application embodiment can be implemented in a server (cloud server). During the application process, the ECU (Electronic Control Unit) can control the in-vehicle touch screen to display the application store interface. It can select the corresponding target application in the application store for download according to the user's selection command. The selection command can be the user's voice command or finger touch screen selection command.

[0045] To address this issue, this application provides an application recommendation method, system, device, and storage medium. The method involves acquiring information data; generating a feature tag set based on the acquired information data; and determining a set of applications to recommend to the user based on the user's feature tag set and an application list. By integrating user interest information or daily usage data to form the feature tag set, and applying this user's feature tag set to application recommendations in the app store, the method eliminates the need for users to manually search for specific applications to download. Instead, intelligent recommendations help applications find the users who need them. This method helps improve the efficiency of user application downloads and increases the click-through rate and download volume of applications in the in-vehicle app store.

[0046] To facilitate understanding of the application recommendation methods provided in the embodiments of this application, the following scenario examples of this application are described.

[0047] The following embodiment illustrates an application recommendation method provided in this application. See also... Figure 1 The figure is a flowchart of an application recommendation method provided in an embodiment of this application. The execution subject of this method is a cloud server. Further, the subject can be an application recommendation system in the cloud server. The method includes:

[0048] S101: Obtain user data.

[0049] User data includes user behavior data and / or user information data. User information data includes basic user information. Basic information may include customer registration information, such as age (e.g., those born in the 1990s or 2000s), family composition (e.g., a family of three, a family of two), identity information (e.g., white-collar workers, students), gender, etc. User behavior data may include transaction information, access information, hobby information, and consultation information, which reflect user needs and preferences.

[0050] In practical applications, user data from various platforms can be integrated using the user's mobile phone number. This includes browsing and clicking data from the car owner's app, browsing data from forums, and the different types of software and their frequency of operation on the car's infotainment system. User surveys can also be sent to users via mobile app to collect their needs.

[0051] S102: Generate a set of user feature tags based on the user data.

[0052] Based on the information data obtained from the above steps, generate N corresponding feature tags, and combine the obtained feature tags into a feature tag set, which includes several feature tags corresponding to the current user.

[0053] In one possible implementation, the generation of the feature label set based on the acquired user data includes A1-A2:

[0054] A1: Perform index processing on the acquired user data to obtain index data.

[0055] Establish a standard indicator system and process user data according to it to form indicator data, thereby achieving quantitative characteristics.

[0056] The indicator system standards can be divided into three dimensions based on the theme: product, user, and behavior.

[0057] The product dimension focuses on the product itself, analyzing its fundamental attributes after launch, including activity levels and average usage time per user. The user dimension focuses on people, analyzing their basic attributes and usage patterns, primarily used for customer operations. The behavior dimension focuses on behavior, also known as scenario analysis, analyzing product performance under different working conditions or scenarios. This is used to optimize user experiences, revise testing standards, and monitor abnormal events such as collisions and scrapes.

[0058] The three dimensions mentioned above can be divided into two categories from a statistical analysis perspective: statistical and detailed. Statistical analysis takes big data as its starting point, statistically analyzing the performance of the topic within the context of big data, removing differences, and better reflecting the average attributes of the topic. Detailed indicators are used for individual analysis, identifying abnormal states, analyzing causes, and can be used for monitoring user, product, and event anomalies. They can also be used for tagging and support statistical analysis.

[0059] Statistical indicators can be further divided into atomic indicators, derived indicators, and distribution indicators. Atomic indicators are indicators that can be directly aggregated from signals. Derived indicators are calculated from basic indicators and can be used for analysis. Distribution indicators are special; they are not specific values ​​but rather data distribution charts. Detailed indicators can also be distinguished as static or dynamic. Static indicators refer to basic, immutable, or unchanging indicators, such as vehicle production date, vehicle purchase date, vehicle features, gender, occupation, and phone number. Dynamic indicators are indicators that change cumulatively over time, such as usage frequency, mileage, number of failures, and number of visits. The standard division of the indicator system in this application embodiment can be referred to Table 1, which is a schematic table illustrating the standard division of the indicator system provided in this application embodiment.

[0060] Table 1. Schematic diagram of the index system standard division provided in the embodiments of this application.

[0061]

[0062]

[0063] A2: Tag the indicator data to obtain a set of user feature tags.

[0064] Based on the several indicator data obtained in the above steps, the indicator data is tagged according to the pre-established labeling standards to form label data for generating labels, which can be understood by professionals and processed by machines.

[0065] For example, if the acquired indicator data is an analysis of activity rates in different time periods, the pre-set labeling standards can include six time periods: 6:00 AM to 10:00 AM, 10:00 AM to 2:00 PM, 2:00 PM to 6:00 PM, 6:00 PM to 10:00 PM, 10:00 PM to 2:00 AM, and 2:00 AM to 6:00 AM. Based on the indicator data within each time period, such as usage frequency and usage time, labels are generated. When the vehicle usage frequency is high between 6:00 PM and 10:00 PM, the feature label "prefers evening usage" is generated. In practical applications, if the usage frequency in multiple time periods is higher than the preset threshold, multiple feature labels related to the activity rate of each time period can be generated. Furthermore, each feature label can be ranked according to the usage frequency of each time period and represented in a hierarchical manner, such as "prefers evening usage (Level 1)" or "prefers early morning usage (Level 2)".

[0066] In practical applications, based on expert experience, machine learning, or actual application needs, key feature tags that fit a specific business scenario can be selected from the feature tags obtained in the above steps to form a user "profile." The specific business scenario can be set according to actual needs and can be adaptively adjusted during application. Feature tags can be manually acquired or deleted; the adaptive adjustment process is not restricted here. For example, the specific scenario could be a non-repair scenario, where the user does not need repair applications recommended. In this case, only feature tags related to the user's interests, travel preferences, and identity information can be acquired, without acquiring repair and maintenance-related tags for the current vehicle user. This directly avoids generating repair-related applications when recommending applications to the user, reducing unnecessary resource waste. The generated feature tag set (user profile) can be found in [reference needed]. Figure 2 , Figure 2 This is a schematic diagram of the user feature tag set provided in the embodiments of this application. It includes the user's interests, travel preferences, basic information, and maintenance information. The interests include a preference for financial management, a preference for popular music, and a preference for road trips. The maintenance information includes high fuel consumption, a preference for 4S stores, and a 3-star maintenance frequency. The travel preferences include fuel-saving experts, daily commuting, and high frequency of travel on weekends. The basic information includes being born in the 1990s, male, a family of three, and a white-collar youth.

[0067] S103: Determine the set of applications to recommend to the user based on the user feature tag set and the application list.

[0068] Based on the "profile" formed by these feature tags, i.e. the feature tag set, the set of applications needed by the current user is determined. This set of applications can include several applications, and the arrangement order and display method of the application set can be edited and set in actual application, i.e., adaptive adjustments can be made without any restrictions.

[0069] In one possible implementation, a set of applications to be recommended to the user is determined based on the user feature tag set and the application list. The application list includes several application sets categorized according to user clusters, where each user cluster is a set of users with the same user feature tags, including B1-B3.

[0070] B1. Determine the user cluster attribute of the user based on the user feature tag set.

[0071] User clusters are used to represent groups of users with the same demographic attributes. User cluster attributes are the demographic attributes corresponding to the current user group. For example, a user cluster attribute of "post-90s women" can be generated.

[0072] In practical applications, user feature tag sets can be clustered using clustering algorithms, collaborative filtering recommendation algorithms, etc., to divide user groups into different clusters. A cluster is a set of samples generated by clustering. Samples within the same cluster are similar to each other, but different from samples in other clusters. That is, users with the same feature tags are clustered to form user clusters, which are sets of users with the same user feature tags. In other words, in this application, when a user has 5 feature tags, there may be 5 or fewer corresponding user cluster attributes. For example, the current user's feature tags are female, born in the 1990s, prefers pop music, and prefers puzzle games. This user may correspond to four user cluster attributes: female born in the 1990s, prefers pop music, and prefers puzzle games. Each feature tag may directly correspond to a user cluster attribute, or several feature tags may combine to correspond to one user cluster attribute; the number of combinations is not limited in this application. The algorithm is not limited in this application; similar recommendation algorithms are acceptable. This patent emphasizes innovation in the algorithm call process and can accommodate other atomic algorithms.

[0073] Based on the previously obtained user profile, i.e. the user's set of user feature tags, obtain the user cluster attribute corresponding to the current user feature tag set. For example, if the current user set includes the basic information tags "post-90s" and "female", the user cluster attribute is determined to be "post-90s female" based on the current tags. Similarly, the user cluster attribute can also be determined based on a feature tag. For example, if the current feature tag is "prefers financial management", the user cluster attribute is determined to be "people who prefer financial management" based on the feature tag.

[0074] B2. Retrieve the application set corresponding to each user cluster attribute in the application list.

[0075] The application list includes several application sets categorized based on user cluster attributes. These user cluster attributes are attributes of user sets sharing the same user characteristic labels. The application sets included in the list can be categorized as follows: applications targeting the post-90s generation, applications targeting the post-80s generation, applications targeting women, applications targeting men, applications targeting those who prefer puzzle games, applications targeting those who prefer combat games, etc. Among these, post-90s, post-80s, women, men, preference for puzzle games, and preference for combat games are all user cluster attributes. Each set includes several applications that conform to the user cluster attributes.

[0076] The application sets in the application list can be preset by the system. The system obtains the user feature tag set of each user in the user sample based on the user samples of the car application store in recent years. Users with the same feature tags are clustered to determine the user cluster attributes. For example, the user samples from 2015 to 2022 are processed and clustered to generate several user cluster attributes, such as white-collar workers, students, women born in the 1990s, women born in the 1980s, and men born in the 1980s. The generated user cluster attributes determine the names of the application sets in the application list, such as the white-collar application set, the student application set, the women born in the 1990s application set, the women born in the 1980s application set, and the men born in the 1980s application set.

[0077] Based on this, the system retrieves the settings attributes of various applications in the app store, such as chat tools, puzzle games, and battle games. It then populates the applications into the corresponding application sets according to these settings, generating sets such as chat tool application sets, puzzle game application sets, and battle game application sets. For example, if application A's settings attribute is card game, application A is added to the card game application set.

[0078] In one possible implementation, the various sets can be filtered and extracted based on the application's defined attributes. For example, the applications in the student application set can be categorized into categories such as chat tools, puzzle games, and combat games. The student application set can then be further subdivided based on the currently generated categories to obtain a student puzzle game set.

[0079] In one possible implementation, several user cluster attribute sets can be combined to represent different categories. For example, the intersection of two application sets—one for female users born in the 1990s and the other for puzzle games—can be used to create a set of puzzle games for female users born in the 1990s, which is then stored in the application list. This means the application list can simultaneously contain these three sets, facilitating subsequent queries based on application subcategories. The combination of these application sets can be adjusted according to actual needs. For instance, if the application subcategory is determined to be "card games for male white-collar workers born in the 1970s," but there is no directly corresponding application set in the application list, the system can retrieve the applications from the male white-collar workers' application set, the card game application set, and the white-collar worker application set, extracting the applications that appear in all three sets and defining them as the "card games for male white-collar workers born in the 1970s" set. This generates the application set directly corresponding to this subcategory.

[0080] For example, a target user's user feature tag set contains four feature tags: A, B, C, and D. Based on these four feature tags, user clusters corresponding to these feature tags are obtained, such as cluster A, cluster B, cluster C, and cluster D. Each cluster has its own corresponding user cluster attributes. Then, based on the four determined user cluster attributes, application sets corresponding to each user cluster attribute are obtained. These application sets can be used to display a list of applications preferred by the user within that set, such as cluster A set, cluster B set, cluster C set, and cluster D set.

[0081] B3. Obtain applications from each application set that meet the preset requirements in terms of application rating, and determine the obtained applications as the application set to recommend to the user.

[0082] Based on the application sets obtained in the above steps, select several applications to generate an application set recommended to the current target user.

[0083] In practical applications, applications that meet preset requirements are obtained from the application set corresponding to each cluster. These applications are then combined to form an application set recommended to the target user. Specifically, the system's recommended application list can be combined with adjustments to the application rankings within the sub-category table. The top-ranked application from each user cluster's corresponding application set can be extracted to form the recommended application list. Alternatively, applications with ratings higher than a preset threshold from each user cluster's corresponding application set can be used to form the recommended application list. For example, applications with ratings higher than 8 points from the four application sets corresponding to the current user can be used to form the recommended application list.

[0084] The preset requirements can be set to a rating of 8 or higher, or based on the app's ranking within the app set corresponding to the user group cluster, such as selecting the top two ranked apps. The display order of the apps extracted from each app set can be adjusted adaptively according to actual needs and is not limited here.

[0085] In one possible implementation, before determining the set of applications to recommend to the user based on the user feature tag set and the application list, the method further includes:

[0086] S1021. Obtain the application data of the target application to be classified.

[0087] The target application is the application currently to be classified, and the application data is data used to characterize the application attributes. The system acquires the application data of the current target application and subsequently determines the initial category of the application based on the current application data.

[0088] S1022. Determine the application list to which the application data of the target application belongs based on the application data, and update the application list.

[0089] In one possible implementation, determining the application list to which the target application's application data belongs based on the application data, and updating the application list, includes C1-C4:

[0090] C1. Determine the initial category of the target application based on the application data.

[0091] After a new app is launched, the publisher will initially categorize it based on the app's attributes, such as positioning it as a puzzle game. The game category positioned here is the initial category of the current game.

[0092] C2. Obtain a reference score for the target application, adjust the initial category of the application based on the reference score, and obtain the subcategories of the target application.

[0093] The reference rating is the rating given to the target application by a number of reference users, and the reference users are those whose ratings of the target application are not lower than a preset threshold.

[0094] During operation, the system can obtain the user cluster attributes of users whose ratings are higher than a preset threshold, and further adjust the initial category of the current game. For example, if the majority of users who rate app A above 8 points are women born in the 1990s, then the initial category of the app will be updated and adjusted based on this user cluster attribute, positioning it as a puzzle game for women born in the 1990s.

[0095] C3. Obtain the initial application set corresponding to the sub-category.

[0096] The application list includes several application sets categorized according to user cluster attributes. Based on the subcategories obtained in the previous steps, the application list is queried to obtain the initial application set corresponding to the subcategories. For example, if the current subcategorie is "Puzzle Games for Post-90s Women," the "Puzzle Games for Post-90s Women" set in the application list is queried, and this set is determined as the initial application set for the current category. In practical applications, if when querying the current subcategorie "Puzzle Games for Post-90s Women," it is found that there is no directly corresponding application set in the application list, the subcategorie can be split. Based on the split content, the corresponding application sets in the application list are obtained. Applications appearing in both sets are then selected to generate the initial application set directly corresponding to the current subcategorie. For example, "Puzzle Games for Post-90s Women" is split into "Post-90s Women" and "Puzzle Games." The applications in the "Post-90s Women" application set include a, d, e, and f, while the applications in the "Puzzle Games" application set include d, e, g, t, h, and l. Applications d and e, which appear in both sets, are selected to generate the initial application set corresponding to the subcategorie "Puzzle Games for Post-90s Women." This set includes applications d and e.

[0097] The initial application set includes several applications that conform to the subcategories. Following the steps above, the application set A corresponding to the current subcategory "Puzzle Games for Post-90s Women" is obtained, and this application set includes game a, game b, and game c.

[0098] C4. Add the target application to the initial application set to obtain the application list.

[0099] Since the target application has been identified as belonging to this category, the target application is added to application set A to generate an updated application set. The current application set is located in the application list, and an updated application list is also generated along with the update of the application set.

[0100] In one possible implementation, the information data includes updated data, which includes user information update data acquired within a preset time period, and / or application information update data. The feature tag set is generated based on the acquired information data, and the feature tag set can be generated based on the updated data during the application process.

[0101] This application provides an application recommendation method. The method involves acquiring information data; generating a feature tag set based on the acquired information data; and determining a set of applications to recommend to the user based on the user's feature tag set and an application list. By integrating user interest information or daily usage data to form the feature tag set, and applying this user's feature tag set to application recommendations in the app store, the method eliminates the need for users to manually search for and download specific applications. Instead, intelligent recommendations help applications find the users who need them. This method helps improve the efficiency of user application downloads and increases the click-through rate and download volume of applications in the in-vehicle app store.

[0102] The above are some specific implementations of the application recommendation method provided in the embodiments of this application. Based on this, this application also provides a corresponding system for application recommendation. The system provided in the embodiments of this application will be described below from the perspective of functional modularity. Figure 3 This is a structural diagram of an application recommendation system provided in an embodiment of this application.

[0103] The system includes:

[0104] User data acquisition unit 201 is used to acquire user data, which includes user behavior data and / or user information data;

[0105] User feature tag set generation unit 202 is used to generate a user feature tag set based on the user data. The feature tag set includes several feature tags used to classify and characterize user attributes.

[0106] The application set determination unit 203 is used to determine the application set to be recommended to the user based on the user feature tag set and the application list. The application list includes several application sets classified according to user cluster attributes. The user cluster attribute is a set of users with the same user feature tags.

[0107] Optionally, the system further includes:

[0108] The cluster generation unit is used to determine the user cluster attributes by performing clustering processing on users with the same feature labels based on the user feature label set.

[0109] Optionally, the system further includes:

[0110] An application data acquisition unit is used to acquire application data of a target application to be classified, wherein the application data is data used to characterize application attributes.

[0111] The application list determination unit is used to determine the application list to which the application data of the target application belongs based on the application data, and to update the application list.

[0112] This application also provides corresponding devices and computer storage media for implementing the application recommendation method provided in this application.

[0113] The device includes a memory and a processor. The memory stores instructions or code, and the processor executes the instructions or code to cause the device to perform the application recommendation method described in any embodiment of this application.

[0114] The computer storage medium stores code, and when the code is executed, the device running the code implements the application recommendation method described in any embodiment of this application.

[0115] This application also provides a vehicle that includes the application recommendation method provided in the first aspect of the embodiments of this application.

[0116] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems or apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and relevant parts can be referred to the method section.

[0117] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0118] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0119] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0120] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An application recommendation method characterized by comprising: The application comprises the following steps: acquiring user data, wherein the user data comprises user behavior data in a vehicle scene and / or user information data, and the user behavior data in the vehicle scene comprises vehicle use time and vehicle use frequency; generating a user feature label set according to the user data, wherein the user feature label set comprises a plurality of feature labels for classifying and representing user attributes; determining an application set recommended to a user according to the user feature label set and an application list, wherein the application list comprises a plurality of application sets classified according to user cluster attributes, and the user cluster attributes are attributes of a user set with the same user feature label; the step of generating the user feature label set according to the user data comprises the following steps: performing index processing on the acquired user data according to a vehicle networking data index system to obtain index data, wherein the vehicle networking data index system comprises product dimensions, user dimensions and behavior dimensions; performing label processing on the index data to obtain the user feature label set; the step of performing label processing on the index data to obtain the user feature label set comprises the following steps: when the vehicle use frequency in a plurality of time periods is higher than a preset threshold, generating a plurality of feature labels about time period activity rates, and performing hierarchical representation on each feature label according to the vehicle use frequency in each time period; the step of determining the application set recommended to the user according to the user feature label set and the application list comprises the following steps: adjusting the application ranking in a fine classification category, taking out an application with the first ranking in the application set corresponding to each user cluster attribute to form a recommended application list, or obtaining an application with a score higher than a preset threshold in the application set corresponding to each user cluster attribute to form a recommended application list.

2. The method of claim 1, wherein, before the step of determining the application set recommended to the user according to the user feature label set and the application list, the method further comprises the following steps: acquiring application data of a target application to be classified, wherein the application data is data for representing attributes of the target application; determining an application list to which the application data of the target application belongs according to the application data, and updating the application list.

3. The method of claim 2, wherein, the step of determining the application list to which the application data of the target application belongs according to the application data, and updating the application list, comprises the following steps: determining an initial category of the target application according to the application data; acquiring score information of the target application, adjusting the initial category of the application according to the score information to obtain a fine classification category of the target application, wherein the score information comprises scores of the target application given by a plurality of reference users and user cluster attributes of the reference users, and the reference users are users with a score not lower than a preset threshold for the target application; acquiring an initial application set corresponding to the fine classification category, wherein the initial application set comprises a plurality of applications conforming to the fine classification category; adding the target application to the initial application set to obtain the application list.

4. The method of claim 1, wherein, the step of determining the application set recommended to the user according to the user feature label set and the application list comprises the following steps: determining user cluster attributes of the user according to the user feature label set; acquiring an application set corresponding to each user cluster attribute in the application list; The applications in each application set are obtained according to the application scores meeting preset requirements, and the obtained applications are determined as the application set recommended to the user.

5. The method of claim 2, wherein, Before the application data of the target application is determined according to the application data, the method further includes: According to the user feature label set, a user cluster attribute is determined through clustering processing of users with the same feature label.

6. The method of claim 1, wherein, The user data includes updated data obtained within a preset time period, and the user feature label set is generated according to the user data. The user feature label set is generated according to the updated data.

7. An application recommendation system characterized by, The system includes: A user data obtaining unit is configured to obtain user data, wherein the user data includes user behavior data and / or user information data in a vehicle scenario, and the user behavior data in the vehicle scenario includes a use time and a use frequency; A user feature label set generating unit is configured to generate a user feature label set according to the user data, wherein the user feature label set includes a plurality of feature labels for classifying and representing user attributes; An application set determining unit is configured to determine an application set recommended to a user according to the user feature label set and an application list, wherein the application list includes a plurality of application sets classified according to user cluster attributes, and the user cluster attributes are a user set with the same user feature label; The user feature label set is generated according to the user data, including: The obtained user data is index processed according to a vehicle networking data index system to obtain index data, wherein the vehicle networking data index system includes product dimensions, user dimensions, and behavior dimensions; The index data is label processed to obtain the user feature label set; The index data is label processed to obtain the user feature label set, including: When the use frequencies in a plurality of time periods are all higher than a preset threshold, a plurality of feature labels about time period activity rates are generated, each feature label is hierarchically represented according to the use frequency of each time period and the sorting; The application set recommended to the user is determined according to the user feature label set and the application list, including: The application ranking in the fine classification table is adjusted, the application with the first ranking in the application set corresponding to each user cluster attribute is taken out, and the application list recommended is composed, or the application with a score higher than a preset threshold in the application set corresponding to each user cluster attribute is obtained, and the application list recommended is composed.

8. An electronic device, comprising: It includes: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the application recommendation method according to any one of claims 1-6 when executing the computer program.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores instructions, and when the instructions run on the terminal device, the terminal device executes the application recommendation method according to any one of claims 1-6.

Images