Appreciation Score Determination Method, Device, Electronic Device, and Storage Medium

Abstract

The present disclosure provides an appreciation score determination method, apparatus, electronic device, and storage medium, relating to the field of Internet technologies. The method includes: obtaining behavior feature information in multiple dimensions based on the behavior information of a target user in a target application; where each dimension represents a usage dimension of the target user for the target application; respectively obtaining a dimension evaluation value corresponding to each of the multiple dimensions according to the behavior feature information in the multiple dimensions; and determining an appreciation score of the target user for the target application based on the obtained multiple dimension evaluation values. Through the above solution, the appreciation ability of the target user for the target application can be accurately obtained.

Description

Background Art

[0002] With the continuous development of Internet technology, various application programs emerge in an endless stream. Currently, for any application program, in order to determine the appreciation ability of users on this application program, it is usually evaluated based on the login duration of users in the application program; specifically, the longer the login duration of a user, the stronger the appreciation ability of the user on the application program; conversely, the shorter the login duration of a user, the weaker the appreciation ability of the user on the application program.

[0003] However, the login duration of users cannot accurately reflect the appreciation ability of users on the application program, making the above evaluation method of users' appreciation ability inaccurate. Summary of the Invention

[0004] Embodiments of the present disclosure provide an appreciation score determination method, device, electronic device, and storage medium for accurately evaluating the appreciation ability of users on an application program.

[0005] In a first aspect, embodiments of the present disclosure provide an appreciation score determination method, including:

[0006] Based on the behavior information of a target user in a target application, obtain behavior feature information of multiple dimensions; where each dimension represents a usage dimension of the target user for the target application;

[0007] Respectively obtain the dimension evaluation values corresponding to the multiple dimensions according to the behavior feature information of the multiple dimensions;

[0008] Based on the obtained multiple dimension evaluation values, determine the appreciation score of the target user on the target application.

[0009] In a second aspect, embodiments of the present disclosure further provide an appreciation score determination device, including:

[0010] A division module for obtaining behavior feature information of multiple dimensions based on the behavior information of a target user in a target application; where each dimension represents a usage dimension of the target user for the target application;

[0011] An evaluation module for respectively obtaining the dimension evaluation values corresponding to the multiple dimensions according to the behavior feature information of the multiple dimensions;

[0012] A determination module for determining the appreciation score of the target user on the target application based on the obtained multiple dimension evaluation values.

[0013] In a third aspect, an embodiment of the present disclosure further provides an electronic device, including a memory and a processor. A computer program that can run on the processor is stored on the memory. When the computer program is executed by the processor, the processor is caused to implement the steps of any one of the appreciation score determination methods in the first aspect.

[0014] In a fourth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium. A computer program is stored in the computer-readable storage medium. When the computer program is executed by a processor, the steps of any one of the appreciation score determination methods in the first aspect are implemented.

[0015] In a fifth aspect, an embodiment of the present disclosure provides a computer program product, which includes computer instructions. The computer instructions are stored in a computer-readable storage medium. When a processor of an electronic device reads the computer instructions from the computer-readable storage medium, the processor executes the computer instructions, causing the electronic device to execute the steps of any one of the above-mentioned appreciation score determination methods.

[0016] The appreciation score determination method provided by the embodiment of the present disclosure has at least the following beneficial effects:

[0017] According to the solution provided by the embodiment of the present disclosure, by analyzing various behavior information of a target user in a target application, behavior feature information in multiple dimensions can be obtained. Then, the behavior feature information in multiple dimensions is respectively evaluated, and according to the evaluation values in multiple dimensions obtained by the evaluation, the appreciation score of the target user for the target application is determined. In this way, by evaluating the behavior feature information in multiple dimensions, the behavior information of the target user in the target application can be comprehensively evaluated, and thus the appreciation ability of the target user on the target application can be accurately obtained.

[0018] Other features and advantages of the present disclosure will be described in the following specification, and part of them will become obvious from the specification, or be understood by implementing the present disclosure. The objectives and other advantages of the present disclosure can be achieved and obtained by the structures specifically pointed out in the written specification, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present disclosure. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0020] Figure 1 It is a schematic diagram of an application scenario of an appreciation score determination method provided by an embodiment of the present disclosure;

[0021] Figure 2 Flow chart of a method for determining appreciation score provided by an embodiment of the present disclosure;

[0022] Figure 3 Flow chart of another method for determining appreciation score provided by an embodiment of the present disclosure;

[0023] Figure 4 Schematic diagram of 27 - level grids provided by an embodiment of the present disclosure;

[0024] Figure 5 Flow chart of another method for determining appreciation score provided by an embodiment of the present disclosure;

[0025] Figure 6 Schematic diagram of transitions between grids of different levels provided by an embodiment of the present disclosure;

[0026] Figure 7 Schematic diagram of an apparatus for determining appreciation score provided by an embodiment of the present disclosure;

[0027] Figure 8 Schematic diagram of another apparatus for determining appreciation score provided by an embodiment of the present disclosure;

[0028] Figure 9 Schematic diagram of another apparatus for determining appreciation score provided by an embodiment of the present disclosure;

[0029] Figure 10 Schematic diagram of the structure of an electronic device provided by an embodiment of the present disclosure. Detailed implementation manners

[0030] In order to make the objectives, technical solutions and advantages of the present disclosure clearer, the present disclosure will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present disclosure without creative efforts shall fall within the protection scope of the present disclosure. The data involved in the present disclosure may be data authorized by users or fully authorized by all parties. The implementation manners / embodiments of the present disclosure may be combined with each other.

[0031] The following explains some terms related to the embodiments of the present disclosure.

[0032] Application: Abbreviated as App, it refers to a computer program installed on a terminal device that can complete one or more services and generally needs to cooperate with a server to run. Common applications are mainly divided into two categories: one is pre-installed system applications, such as text messages, photos, memos, etc.; the other is third-party applications, such as information applications, shopping applications, social applications, etc. The target application in the embodiments of the present disclosure can be one of the third-party applications.

[0033] Degree of duration: It refers to the amount of time of touching a thing. In the embodiments of the present disclosure, it refers to the amount of time a user logs in to an application.

[0034] Degree of breadth: It refers to the scope degree of a thing. In the embodiments of the present disclosure, it refers to the breadth of a user's use of an application, which can be specifically measured by the corresponding behavior information of the user in the application. For example: when the application is a music application, the breadth can represent the number of music genre categories and the number of music language categories that the user listens to in the music application, etc.

[0035] Degree of depth: It refers to the degree of touching the essence of a thing, that is, the degree of development of a thing to a higher stage. In the embodiments of the present disclosure, it refers to the depth of a user's use of an application, which can be specifically measured by the corresponding behavior information of the user in the application. For example: when the application is a music application, the depth can represent the popularity degree of the music singers that the user listens to in the music application, including the general type, the niche type, etc.

[0036] Transition: When the level of any one dimension among a user's multiple dimensions increases, it is a transition of the user. For example, if one dimension is breadth, including three levels: low, medium, and high, then when the user changes from low breadth to medium breadth, it can be called a transition.

[0037] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein.

[0038] In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those clearly listed steps or units, but may include other steps or units that are not clearly listed or are inherent to these processes, methods, products, or devices.

[0039] Next, the design concept of the embodiments of the present disclosure will be introduced.

[0040] As described above, for any application, in order to determine the user's appreciation ability on the application, it is usually evaluated based on the user's login duration in the application; specifically, the user with a longer login duration has a stronger appreciation ability on the application; conversely, the user with a shorter login duration has a weaker appreciation ability on the application.

[0041] Considering that the user's behavior information in the application not only includes login information, but also includes various other behavior information, which is specifically related to the functions provided by the application; for example, if the application is an audio and video application, the user's behavior information in the audio and video application includes, but is not limited to, login information, audio and video playback information, audio and video collection information, audio and video download information, etc. Evaluating the user's appreciation ability of the application only through the user's login duration is not comprehensive.

[0042] Therefore, the above evaluation method cannot accurately evaluate the user's appreciation ability on the application. In view of this, the embodiments of the present disclosure provide an appreciation score determination method, device, electronic device, and storage medium. By obtaining the behavioral feature information of multiple dimensions through the behavioral information of the target user in the target application, and then evaluating the behavioral feature information of multiple dimensions, the behavioral information of the target user in the target application can be comprehensively evaluated, and then the appreciation ability of the target user on the target application can be accurately obtained.

[0043] The application scenario of the embodiments of the present disclosure will be exemplarily introduced below with reference to the accompanying drawings.

[0044] Reference Figure 1 , which is a schematic diagram of the application scenario of the appreciation score determination method provided by the embodiments of the present disclosure. This application scenario includes multiple terminal devices 100 and a server 200, and the multiple terminal devices 100 and the server 200 can be respectively connected through a wired or wireless communication network.

[0045] Among them, the terminal device 100 is an electronic device used by the user, and this electronic device includes, but is not limited to, desktop computers, mobile phones, computers, smart home appliances, intelligent voice interaction devices, vehicle-mounted terminals and other electronic devices. Various applications can be installed on the terminal device, such as news applications, shopping applications, social applications, audio and video applications, and so on. The server 200 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, as well as big data and artificial intelligence platforms.

[0046] In an alternative embodiment, an audio-video application can be installed in the terminal device 100. The target user can use the terminal device 100 to log in to the audio-video application and perform various operations in the audio-video application, including but not limited to playing audio and video, favoriting audio and video, downloading audio and video, etc. The server 200 can be the background server of the audio-video application and can obtain various behavioral characteristic information of the target user in the target application, including but not limited to login information, audio and video playing information, audio and video favoriting information, audio and video downloading information, etc.

[0047] In order to comprehensively evaluate the appreciation ability of the target user on the audio-video application, the server 200 can evaluate the above behavioral information from multiple dimensions; specifically, obtain behavioral characteristic information in multiple dimensions based on the above behavioral information, evaluate the behavioral characteristic information in multiple dimensions respectively, obtain the dimension evaluation values corresponding to each of the multiple dimensions, and then based on the obtained multiple dimension evaluation values, the appreciation score of the target user on the target application can be accurately obtained, and this appreciation score can represent the appreciation ability.

[0048] The above application scenarios are only shown for the convenience of understanding the spirit and principle of the present disclosure, and the embodiments of the present disclosure are not limited in this regard. On the contrary, the embodiments of the present disclosure can be applied to any applicable scenario.

[0049] The method for determining the appreciation score according to the embodiments of the present application will be introduced below in conjunction with the accompanying drawings and specific embodiments.

[0050] Reference Figure 2 , the embodiments of the present disclosure provide a method for determining an appreciation score, which can be applied to a server, such as Figure 1 the server 200 shown, and this method for determining the appreciation score can include the following steps S201 - S203:

[0051] S201, based on the behavioral information of the target user in the target application, obtain behavioral characteristic information in multiple dimensions; where each dimension represents a usage dimension of the target user for the target application.

[0052] Among them, the target application can be any application program. The target user can use the terminal device installed with the target application to log in to the target application and perform various operations, thereby generating various behavioral information, including login information, operation information, etc.; the terminal device can send this behavioral information to the server for storage by the server.

[0053] In S201, the server can obtain the behavior information of the target user in the target application within a set historical time period. The set historical time period can be set as needed, such as recent days, recent months, recent months, etc., which is not limited here; further, the obtained behavior information is analyzed to obtain behavior feature information of multiple dimensions. The multiple dimensions can be set as needed, and the behavior feature information contained in each of the multiple dimensions can be determined according to the specific content of the behavior information, which is not limited in the embodiments of the present disclosure.

[0054] In a possible implementation, the multiple dimensions may include multiplicity, breadth, and depth; then in S201 above, based on the behavior information of the target user in the target application, the behavior characteristic information of multiplicity, the behavior characteristic information of breadth, and the behavior characteristic information of depth may be obtained.

[0055] Among them, the plurality of behavioral characteristic information, the breadth of behavioral characteristic information, and the depth of behavioral characteristic information can be determined according to the specific content of the behavioral information.

[0056] Optionally, the target application is a media resource playback application, such as an audio or video application, and the target user's behavior information in the target application includes at least: login information and media resource operation information; wherein the login information may include the time point and login duration of each target user's login to the target application, and the media resources include but are not limited to audio, video, etc., and the operation information includes but is not limited to the playback information, collection information, download information, etc. of each media resource.

[0057] On this basis, the above-mentioned obtaining the multi-degree behavior characteristic information, the breadth behavior characteristic information and the depth behavior characteristic information based on the behavior information of the target user in the target application may include the following steps A1-A3:

[0058] A1. Determine the login duration information of the target user in the target application based on the login information, and use the login duration information as multiple behavioral feature information.

[0059] In this step, the login duration information of the target user in the set historical time period may be determined based on the login information of the target user in the set historical time period, for example, the number of login hours and login days in the past month.

[0060] A2. Based on the media resource operation information, determine the type information of the media resource played by the target user in the target application and the character type information in the played media resource.

[0061] Among them, the media resources can include multiple types, and the classification rules for the multiple types can be set as needed. For example, when the media resource is music, the type information of the music can include music genre categories, language categories, etc. The music genre categories can include, for example, pop, rock, jazz, folk, light music, classical, etc., and the language categories can be various languages, such as Chinese, English, French...; when the media resource is a TV or movie video, the type information of the video can include video style categories, language categories, etc. The video style categories can include, for example: modern, comedy, emotional, historical, war, science fiction, and so on.

[0062] The characters involved in the media resources can also include multiple types, and the classification rules for the multiple character types can also be set as needed. Taking the media resource being music as an example, the character can be the singer of the song. For example, all singers can be divided into multiple singer types according to the popularity of the singer. The embodiments of the present disclosure do not limit this.

[0063] The above classification rules for the media resource types and the classification rules for the character types in the media resources are only exemplary and can be specifically set as needed. The embodiments of the present disclosure do not limit this.

[0064] A3. Using the type information of the played media resource as the behavior feature information of the breadth, and using the character type information in the played media resource as the behavior feature information of the depth.

[0065] Exemplarily, still taking the media resource being music as an example, the multiplicity can represent the amount of the user's login duration; the breadth can represent the amount of the language categories and music genre categories of the music listened to by the user; the depth can represent whether the music listened to by the user belongs to the popular type or the niche type, which can be determined according to the singer type. For example: the music of popular singers belongs to the popular type, and the music of unpopular singers belongs to the niche type.

[0066] S202. Obtaining the dimension evaluation values corresponding to each of the multiple dimensions respectively according to the behavior feature information of the multiple dimensions.

[0067] Among them, for the behavior feature information of each dimension, according to the set calculation rules, the behavior feature information of this dimension can be calculated to obtain the dimension evaluation value of this dimension. In the following embodiments of the present disclosure, the behavior feature information of multiple dimensions will be further exemplified, and the calculation rules for the feature information of each dimension will be introduced exemplarily.

[0068] S203. Determining the appreciation score of the target user for the target application based on the obtained multiple dimension evaluation values.

[0069] In this step, multiple dimension evaluation values can be calculated according to the set calculation rules to obtain the appreciation score of the target user for the target application. For example, the multiple dimension evaluation values are weighted and summed to obtain the appreciation score.

[0070] Exemplarily, if the multiple dimensions include multiplicity, depth, and breadth, then the appreciation score = w1 * multiplicity evaluation value + w2 * depth evaluation value + w3 * breadth evaluation value, where w1, w2, and w3 are the weights of multiplicity, depth, and breadth respectively, and can be specifically set according to the actual situation, such as 0.3, 0.4, and 0.3 respectively, and there is no limitation here.

[0071] Optionally, in order to simplify the calculation process, in the above S203, the multiple dimension evaluation values can be respectively normalized so that the value range of each dimension evaluation value is 0 to 1. Among them, the normalization methods include but are not limited to max - min normalization, normal normalization, sigmoid normalization, etc.

[0072] For example, taking the max - min normalization method as an example, each dimension evaluation value can be normalized through the following formula (1):

[0073]

[0074] Where X′ is the normalized value, X is the dimension evaluation value, Xmin is the minimum dimension evaluation value, and Xmax is the maximum dimension evaluation value. The target user can be multiple users. Among the dimension evaluation values of multiple users in each dimension, there are the minimum dimension evaluation value and the maximum dimension evaluation value.

[0075] After obtaining the multiple normalized dimension evaluation values of each user, the appreciation score of the user for the target application can be determined based on the multiple normalized dimension evaluation values of each user.

[0076] It should be noted that considering that the behavior information of the target user in the target application will change continuously, the above steps S201 - S203 can be executed periodically, for example, once every few days, a month, or several months, so as to periodically determine the appreciation score of the target user on the target application.

[0077] In the embodiments of the present disclosure, by analyzing various behavior information of the target user in the target application, behavior feature information of multiple dimensions can be obtained, and then the behavior feature information of multiple dimensions is respectively evaluated, so that the behavior information of the target user in the target application can be comprehensively evaluated, and then the appreciation ability of the target user on the target application can be accurately obtained.

[0078] In some embodiments, the above-mentioned target users may include multiple users. After obtaining the multiple-dimensional evaluation values of each of the multiple users in S202 above, the multiple users may also be grouped based on the multiple-dimensional evaluation values. For example, Figure 3 As shown, in order to group the multiple users, the following steps S204 - S206 may be performed:

[0079] S204, divide each of the multiple dimensions into multiple levels according to a preset level rule, and construct multiple level grids, where each level grid is composed of the levels of the corresponding dimension.

[0080] Among them, the preset level rule for each dimension can be set as needed, and the preset level rules between multiple dimensions can be the same or different.

[0081] For example: the multiple dimensions include abundance, breadth, and depth. Each dimension is set with 3 levels: low, medium, and high. These 3 levels can respectively correspond to corresponding value ranges, which can be specifically set as needed. For example, for abundance, the range of the dimension evaluation value of abundance is 0 - 100. The dimension evaluation value less than 30 is taken as the low level, 30 - 60 as the medium level, and greater than 60 as the high level; the value ranges of the three levels of breadth and depth can also be set as needed.

[0082] In this way, each of the 3 dimensions corresponds to 3 levels, and 27 level grids can be constructed. Each level grid is composed of 1 level of each of the 3 dimensions. For example, 1 level grid corresponds to the low level of abundance, the medium level of breadth, and the high level of depth, as specifically shown in Figure 4 As shown.

[0083] It should be noted that the above 3 dimensions are described by taking 3 levels corresponding to each dimension as an example. Of course, each dimension can also correspond to N (for example, greater than 3) levels, and the number of finally formed level grids is N 3 pieces.

[0084] S205, obtain the levels of multiple dimensions corresponding to each of the multiple users according to the multiple-dimensional evaluation values of each of the multiple users.

[0085] As can be seen from the above, each level of each dimension corresponds to a corresponding value range. For each user, according to the multiple-dimensional evaluation values of this user, the level of each dimension evaluation value under the corresponding dimension can be determined. In this way, the levels of multiple dimensions corresponding to each of the multiple users can be obtained.

[0086] For example, the multi-dimensional evaluation values of a user include: the multiplicity evaluation value 40, the breadth evaluation value 30, and the depth evaluation value 70. Assume that the grading rules for these 3 dimensions are all: less than 30 is the low grade, 30 - 60 is the medium grade, and greater than 60 is the high grade. Then, for this user, it corresponds to the medium grade in terms of multiplicity, the medium grade in terms of breadth, and the high grade in terms of depth.

[0087] Optionally, in order to simplify the process of determining the grades of multiple dimensions of multiple users, the value range of the dimension evaluation value of each dimension can be specified as 0 - 1. By normalizing the multi-dimensional evaluation values of each user respectively, the value range of each dimension evaluation value is made to be 0 - 1. At the same time, multiple grades under each dimension can also be set according to the value range of 0 - 1. For example, it can be set as: less than 0.3 is the low grade, 0.3 - 0.6 is the medium grade, and greater than 0.6 is the high grade.

[0088] At this time, the step of obtaining the grades of multiple dimensions corresponding to each of the multiple users according to the multi-dimensional evaluation values of each user among the multiple users in step S205 above may include the following steps:

[0089] B1. Normalize the multi-dimensional evaluation values of each user among the multiple users respectively to obtain the multi-dimensional normalized values of each user among the multiple users.

[0090] When normalizing each dimension evaluation value of each user, it can be processed based on the maximum - minimum normalization method in formula (1) in the above embodiment, or it can also be processed based on normal normalization, sigmoid normalization, etc.

[0091] B2. Obtain the grades of multiple dimensions corresponding to each of the multiple users according to the multi-dimensional normalized values of each user among the multiple users.

[0092] This step is similar to the implementation process of step S205 above and will not be elaborated here.

[0093] S206. Determine the grade grid where the user is located according to the grades of multiple dimensions corresponding to any user.

[0094] Exemplarily, taking the above 27 grade grids as an example, assume that a user corresponds to the medium grade in terms of multiplicity, the medium grade in terms of breadth, and the high grade in terms of depth. Then, this user is located in the grade grid corresponding to the medium grade in terms of multiplicity, the medium grade in terms of breadth, and the high grade in terms of depth among the 27 grade grids.

[0095] In the above embodiments of the present disclosure, by constructing multiple hierarchical grids and dividing multiple users into corresponding hierarchical grids, the clustering of multiple users is realized, so as to manage multiple users hierarchically and achieve refined operation; it can also prompt users to make transitions between different hierarchical grids to improve the appreciation ability of users on the target application.

[0096] The following embodiments introduce the transition process of users between different hierarchical grids.

[0097] In some embodiments, as Figure 5 shown, after determining the respective hierarchical grids where multiple users are located in the above step S206, the following steps S207 - S209 can be executed:

[0098] S207, obtain the user transition probability from any first - level grid to any second - level grid among multiple hierarchical grids within a set historical time period.

[0099] Since the behavioral information of users in the target application is constantly changing, the corresponding multi - dimensional evaluation values may also change, resulting in possible changes in the hierarchical grids where users are located. Therefore, by obtaining the change situation of the hierarchical grids where multiple users are located within a set historical time period, the user transition probability from any hierarchical grid (i.e., the first - level grid) to other hierarchical grids (i.e., the second - level grid) can be determined. Among them, the set historical time period can be set as needed, such as recent days, recent month, recent several months, etc., and is not limited here.

[0100] In the first optional implementation manner, the above S207 may include the following steps:

[0101] C1. Determine the user proportion of users transitioning from any first - level grid to any second - level grid within a set historical time period.

[0102] C2. Use the user proportion as the user transition probability from the first - level grid to the second - level grid.

[0103] Exemplarily, assuming that the hierarchical grids where multiple users are located are determined once every month, the set historical time period can be the recent month. Assuming that the recent month is December, then first obtain the hierarchical grids where multiple users are located in November, and then obtain the hierarchical grids where multiple users are located in December. In this way, the users who have made transitions in each hierarchical grid and the hierarchical grids they have transitioned to can be determined.

[0104] As Figure 6As shown in the figure, taking grid 1 in the 27-level grid as an example, grid 1 corresponds to a low degree level, a low breadth level, and a low depth level. Suppose in November, it is determined that grid 1 contains 100 users. In December, after determining the grids where multiple users are located again, it is found that 40 users in grid 1 have all jumped to grid 2 corresponding to a medium degree level, a medium breadth level, and a medium depth level, and 20 users have all jumped to grid 3 corresponding to a medium degree level, a low breadth level, and a medium depth level. Then the user conversion probability from grid 1 to grid 2 is 40%, and the user conversion probability from grid 1 to grid 3 is 20%. By analogy, the user conversion probability of each grid to other grids can be determined.

[0105] In the second optional implementation manner, in order to more accurately determine the user conversion probability of any first grid to any second grid, all users in the first grid can be divided into multiple user portraits based on their respective user portraits, and then the user conversion probabilities of different user portraits can be determined respectively. That is, the user conversion probability in S207 above includes the user conversion probabilities of different user portraits.

[0106] At this time, obtaining the user conversion probability of any first grid to any second grid in multiple grids within a set historical time period in S207 above can include the following steps:

[0107] D1. Determine the proportion of users with different user portraits who jump from any first grid to any second grid within the set historical time period.

[0108] D2. Use the proportion of users with different user portraits as the user conversion probability of different user portraits when the first grid jumps to the second grid.

[0109] Among them, the user portrait can be determined based on user age, gender, behavior information in the target application, etc. Therefore, different user portraits can be distinguished by age, gender, and specific behavior information.

[0110] Exemplarily, different user portraits are distinguished by age group. Taking grid 1 of the first level as an example, for example, grid 1 of the first level corresponds to a low degree level, a low breadth level, and a low depth level. Suppose in November, it is determined that grid 1 of the first level contains 100 users. Among these 100 users, there are 10 users under 20 years old, 60 users aged 20 - 30, 20 users aged 30 - 40, and 10 users over 40 years old.

[0111] In December, after determining the level grids where multiple users are located again, it is found that among the 60 users aged 20 to 30 in the first-level grid 1, 30 users have all jumped to the second-level grid 1 (for example, corresponding to a low degree level, a medium breadth level, and a medium depth level), and 15 users have all jumped to the second-level grid 2 (for example, corresponding to a medium degree level, a medium breadth level, and a low depth level); among the 30 users aged 30 to 40, 9 users have all jumped to the second-level grid 1, and 6 users have all jumped to the second-level grid 3 (for example, corresponding to a medium degree level, a medium breadth level, and a medium depth level). Users in other age groups may also jump to multiple second-level grids, which will not be listed one by one here.

[0112] It can be seen that among the 60 users aged 20 to 30 in the first-level grid 1, the conversion probability of users jumping to the second-level grid 1 is 30 / 60 = 50%, and the conversion probability of users jumping to the second-level grid 2 is 15 / 60 = 25%; among the 30 users aged 30 to 40, the conversion probability of users jumping to the second-level grid 1 is 9 / 30 = 30%, and the conversion probability of users jumping to the second-level grid 3 is 6 / 30 = 20%; and so on. The conversion probabilities of different user portraits in the first-level grid jumping to any second-level grid can be determined.

[0113] S208. For any first-level grid, use the second-level grid corresponding to the maximum user conversion probability between the first-level grid and any second-level grid as the jump-level grid of the first-level grid.

[0114] Based on the above S207, the conversion probabilities of users in any level grid (i.e., the first-level grid) jumping to other level grids (i.e., the second-level grid) can be determined. For a first-level grid, use the second-level grid corresponding to the maximum user conversion probability of this first-level grid as the jump-level grid of this first-level grid.

[0115] For example, in the above first optional implementation manner, the conversion probability of users in the first-level grid 1 jumping to the second-level grid 2 is 40%, the conversion probability of users jumping to the second-level grid 3 is 20%, and the conversion probabilities of users jumping to second-level grids other than the second-level grid 2 and the second-level grid 3 are all less than 20%. Then, use the second-level grid 2 as the jump-level grid of the first-level grid 1.

[0116] Optionally, when the above-mentioned user conversion probabilities include user conversion probabilities of different user portraits, when determining the transition level grid of any first-level grid in S208 above, for any first-level grid among multiple level grids, the second-level grid corresponding to the maximum user conversion probability of the corresponding user portrait between the first-level grid and any second-level grid can be used as the transition level grid of the corresponding user portrait within the first-level grid.

[0117] For example, in the second optional implementation manner above, for users with a user portrait of 20 to 30 years old in the first-level grid 1, the user conversion probability of transitioning to the second-level grid 1 is 30 / 60 = 50%, and the user conversion probability of transitioning to the second-level grid 2 is 15 / 60 = 25%; then the second-level grid 1 is used as the transition level grid for users aged 20 to 30.

[0118] S209. For any first user within the first-level grid, calculate the similarity between any second user within the transition level grid and the first user, and recommend the behavior information of the second user whose similarity reaches the similarity threshold to the first user.

[0119] In this step, in order to prompt the first user in any first-level grid to transition to its corresponding transition level grid, for any first user, calculate the similarity between any second user within the transition level grid and the first user, use the second user whose similarity reaches the similarity threshold as the similar user of the first user, and recommend the behavior information of the similar user to the first user. Among them, the similarity threshold can be set as needed and is not limited here. The similar users of each first user can be one or more.

[0120] Among them, when recommending the behavior information of the similar user to the first user, the behavior information that the first user has not occurred in the behavior information of the similar user can be recommended to the first user. For example, when the target application is a music application, the behavior information of the second user in the music application can include song playing information, song collection information, song download information, etc., and the songs that the first user has not played among the multiple songs played by the second user can be recommended to the first user.

[0121] It should be noted that after determining the similar users of the first user, the behavior information of the similar users can be continuously recommended to the first user until the first user transitions to the corresponding transition level grid.

[0122] Optionally, when calculating the similarity between any second user within the transition level grid and the first user in S209 above, the similarity between any second user within the transition level grid and the first user can be calculated based on the behavior information of the first user in the target application and the behavior information of any second user in the transition level grid in the target application.

[0123] Specifically, based on the behavior information of the first user in the target application, a user portrait of the first user can be obtained and represented as a first vector; similarly, based on the behavior information of any second user in the target application, a user portrait of any second user can be obtained and represented as a second vector; then, by calculating the similarity between the first vector and the second vector through a vector similarity algorithm, the similarity between any second user and the first user can be obtained; where the vector similarity algorithm can be, for example, cosine distance, Hamming distance, etc., which is not limited here.

[0124] It should be noted that when the above user conversion probability includes the user conversion probability of different user portraits, when recommending the behavior information of similar second users in the transition level grid to any first user in the first level grid in S209 above, for any first user with a user portrait in the first level grid, the similarity between the first user and any second user in the corresponding transition level grid can be calculated, and the behavior information of the second user whose similarity reaches the similarity threshold can be recommended to the first user.

[0125] In addition, since the users in multiple level grids are constantly changing, in S209 above, when calculating the similarity between any second user and the first user in the transition level grid for any first user in the first level grid, any first user may also be a new first user who has just entered the first level grid.

[0126] In the above embodiments of the present disclosure, based on the behavior feature information of the target user in multiple dimensions in the target application, dimension evaluation values in multiple dimensions are obtained. According to the dimension evaluation values in multiple dimensions, not only can the appreciation score of the target user for the target application be obtained, but also multiple level grids can be constructed to obtain multiple user groups. Then, any user in the target user is prompted to transition between different level grids (user groups) to improve the appreciation score of the user on the target application.

[0127] In the following embodiments, taking multiple dimensions including abundance, breadth, and depth as an example, an exemplary introduction to the evaluation process of the dimension evaluation values in multiple dimensions is given.

[0128] Taking the target application as a media resource playback application as an example, based on the above steps A1 - A3, the login duration information of the target user in the target application is used as the behavior feature information of abundance, the type information of the played media resources is used as the behavior feature information of breadth, and the character type information in the played media resources is used as the behavior feature information of depth.

[0129] Further, in the above S202, obtaining the dimension evaluation values corresponding to each of the multiple dimensions according to the behavioral characteristic information of multiple dimensions may include the following steps E1 - E3:

[0130] E1. Obtain the multi - degree evaluation value according to the login duration information of the target user in the target application.

[0131] Optionally, the login duration information may include the login hours and login days; according to the login hours and login days of the target user in the target application within the first set time period, the multi - degree evaluation value can be obtained. Among them, the first set time period may be the set historical time period in the above - mentioned embodiment, and can be specifically set as needed, such as recent days, recent month, recent several months, etc., which is not limited herein.

[0132] For example, if the first set time period is the recent month, the multi - degree evaluation value D1 can be calculated by the following formula (2):

[0133] D1 = lna * lnb (2)

[0134] Where a is the cumulative login hours in the recent month, and b is the cumulative login days in the recent month.

[0135] E2. Obtain the breadth evaluation value according to the type information of the media resources played by the target user in the target application.

[0136] Among them, the type information of the media resources may include the language category and style category of the media resources. Among them, the style category can be set according to the specific content of the media resources. For example, as mentioned above, when the media resource is music, the style category can be the music genre category; when the media resource is a TV or movie video, the style category can refer to the video style category. In the embodiments of the present disclosure, the breadth evaluation value can be calculated based on the type information of the media resources according to the set calculation method.

[0137] E3. Obtain the depth evaluation value according to the character type information in the media resources played by the target user in the target application.

[0138] Taking the media resource as music as an example, the character can be the singer of the song. For example, the character type information can be divided according to the popularity of the singer. In the embodiments of the present disclosure, the depth evaluation value can be calculated based on the character type information in the media resources according to the set calculation method.

[0139] The following gives an exemplary introduction to the calculation method of the breadth evaluation value in the above E2.

[0140] In some embodiments, the type information of the media resources in step E2 above may include language categories and style categories; then, obtaining the breadth evaluation value according to the type information of the media resources played by the target user in the target application in step E2 may include the following steps E21 - E23:

[0141] E21. Determine the first breadth evaluation value according to the language categories and style categories of the media resources played by the target user in the target application within the second set time period.

[0142] Optionally, the language categories of the media resources may include the number of language categories of the media resources played by the target user and the maximum language category proportion; the style categories of the media resources may include the number of style categories of the media resources played by the target user and the maximum style category proportion.

[0143] Based on this, step E21 may include: determining the first breadth evaluation value according to the number of language categories of the media resources played by the target user in the target application, the maximum language category proportion, the number of style categories, and the maximum style category proportion within the second set time period.

[0144] Among them, the second set time period is a set historical time period, which may be the same as the above first set time period. The maximum language category proportion refers to the ratio of the language category with the largest number of plays to the number of language categories of the media resources played in various languages within the second set time period. For example, if the language categories played within the second set time period include Chinese and English, the number of Chinese media resources is 100, and the number of English media resources is 50, then the maximum language category is Chinese, and the maximum language category proportion is 100 / 150 = 2 / 3. The maximum style category proportion is similar to the maximum language category proportion and will not be elaborated here.

[0145] Exemplarily, if the second set time period is the past month, then the first breadth evaluation value G1 can be calculated by the following formula (3):

[0146] G1 = lnq1 + ln(1.05 - t1) * 0.5 + lnq2 + ln(1.05 - t2) * 0.5 (3)

[0147] Among them, q1 is the number of language categories of the media resources played in the past month, t1 is the maximum language category proportion, q2 is the number of style categories of the media resources played in the past month, t2 is the maximum style category proportion, and 1.05 and 0.5 are both set parameters, which can also be replaced with other values and are not limited here.

[0148] E22. Determine the second breadth evaluation value according to the language categories and style categories of the media resources liked by the target user.

[0149] Among them, the language category and style category of the media resources liked by the target user can be determined according to each media resource marked as liked by the target user. For example, the target user can add the liked media resources to the like list.

[0150] In some optional embodiments, step E22 may include: determining a second breadth evaluation value according to the number of language categories of the media resources liked by the target user, the number of language categories of the acceptable media resources, the number of style categories of the liked media resources, and the number of style categories of the acceptable media resources.

[0151] Exemplarily, the second breadth evaluation value G2 is calculated by the following formula (4):

[0152] G2 = lnh1 + ln(1.05 - t3) * 0.5 + lnh2 * 0.5 + lnh3 + ln(1.05 - t4) * 0.5 + lnh4 * 0.5 (4)

[0153] Among them, h1 is the number of language categories of the liked media resources, t3 is the maximum proportion of the liked language category, h2 is the number of language categories of the acceptable media resources, h3 is the number of style categories of the liked media resources, t4 is the maximum proportion of the liked style category, h4 is the number of style categories of the acceptable media resources, and both 1.05 and 0.5 are set parameters, which can also be replaced with other values and are not limited here.

[0154] Optionally, when determining the language category of the media resources liked by the target user, the language category of the acceptable media resources, the style category of the liked media resources, and the style category of the acceptable media resources, the following methods may be included:

[0155] If the proportion of a language category in the media resource like list of the target user reaches the first preset ratio and the number of media resources of a language category reaches the first preset quantity, then determine a language category as the language category of the media resources liked by the target user;

[0156] If the proportion of a language category in the media resource like list of the target user does not reach the first preset ratio and the number of media resources of a language category does not reach the first preset quantity, then determine a language category as the language category of the acceptable media resources of the target user;

[0157] If the proportion of a style category in the media resource like list of the target user reaches the second preset ratio and the number of media resources of a style category reaches the second preset quantity, then determine a style category as the style category of the media resources liked by the target user;

[0158] If the proportion of a style category in the media resource like list of the target user does not reach the second preset ratio, and / or the number of media resources of a style category does not reach the second preset quantity, then determine a style category as the style category of the media resources acceptable to the target user.

[0159] Among them, the media resource like list of the target user includes each media resource added by the target user. The first preset ratio, the first preset quantity, the second preset ratio, and the second preset quantity can all be set as needed. For example, both the first preset ratio and the second preset ratio are 1%, and both the first preset quantity and the second preset quantity are 2. The embodiments of the present disclosure do not limit this.

[0160] E23. Determine the breadth evaluation value based on the first breadth evaluation value, the second breadth evaluation value, the number of media resources liked by the target user, and the playing duration of the media resources within the second set time period.

[0161] Among them, the number of media resources liked by the target user can be the quantity of each media resource in the above media resource like list.

[0162] For example, calculate the breadth evaluation value G through the following formula (5):

[0163]

[0164]

[0165] Among them, T is the playing duration of the media resources within the second set time period, m is the number of media resources liked by the target user, τ m is the time decay factor of the mth media resource added to the media resource like list, datediff(T now , T m ) refers to the number of days between the addition date of the mth media resource added to the media resource like list and the current date. 0.005 is a set parameter, which can make the decay rate of time be 0.85 in one year and 0.7 in two years. This set parameter can also be other values.

[0166] The following gives an exemplary introduction to the calculation method of the depth evaluation value in the above step E3.

[0167] In some embodiments, obtaining the depth evaluation value according to the character type information in the media resources played by the target user in the target application in the above step E3 may include the following steps E31 - E34:

[0168] E31. Determine the proportion evaluation value according to the respective quantity proportions of multiple character types in the media resources played by the target user in the target application.

[0169] Assume that multiple character types may include leading characters, middle characters, and trailing characters. This step may calculate the proportion evaluation value S1 through the following formula (7):

[0170] S1 = f1 * 0.05 + f2 * 0.1 + f3 * 0.85 (7)

[0171] where f1 is the proportion of the number of played leading characters, f2 is the proportion of the number of played middle characters, f3 is the proportion of the number of played trailing characters, and 0.05, 0.1, and 0.85 are all set parameters, which may also be replaced with other values. The embodiments of the present disclosure do not limit this.

[0172] E32. Determine the play evaluation value according to the proportion of the play times of each of the multiple character types played by the target user in the target application within the third set time period.

[0173] Among them, the third set time period is a set historical time period, which may be the same as the above-mentioned second set time period. According to the number of each of the multiple character types involved in the media resources played by the target user within the third set time period, the proportion of the play times of each of the multiple character types can be determined. For example, if the media resource is music and the multiple character types include leading singers, middle singers, and trailing singers, and the play times of these three character types are 100, 50, and 20 respectively, then the proportion of the play times of the leading singer is 100 / 170 = 10 / 17, the proportion of the play times of the middle singer is 50 / 170 = 5 / 17, and the proportion of the play times of the trailing singer is 20 / 170 = 2 / 17.

[0174] Exemplarily, if the third set time period is the recent one month, then this step may calculate the play evaluation value S2 through the following formula (8):

[0175] S1 = f4 * 0.05 + f5 * 0.1 + f6 * 0.85 (8)

[0176] where f4 is the proportion of the play times of the leading characters within the recent one month, f5 is the proportion of the play times of the middle characters within the recent one month, f6 is the proportion of the play times of the trailing characters within the recent one month, and 0.05, 0.1, and 0.85 are all set parameters, which may also be replaced with other values. The embodiments of the present disclosure do not limit this.

[0177] E33. Determine the like degree evaluation value according to the number of likes of each of the multiple character types by the target user in the target application.

[0178] Among them, the number of likes of each character type by the target user can be the number of times each character type involved in the media resources in the target user's media resource like list appears. For example, if the media resource is music and the multiple character types include the lead singer, the middle singer, and the tail singer, then according to the singers included in the target user's music like list, the number of lead singers, the number of middle singers, and the number of tail singers can be determined.

[0179] Exemplarily, this step can calculate the like degree evaluation value S3 through the following formula (9):

[0180] S3 = f7 * 0.05 + f8 * 0.1 + f9 * 0.85 (9)

[0181] Among them, f7 is the number of lead characters liked by the target user, f8 is the number of middle characters liked by the target user, f9 is the number of tail characters liked by the target user, and 0.05, 0.1, and 0.85 are all set parameters, which can also be replaced by other values, and the embodiments of the present disclosure do not limit this.

[0182] E34. Determine the depth evaluation value of the depth based on the proportion evaluation value, the play evaluation value, and the like degree evaluation value.

[0183] For example, this step can calculate the depth evaluation value S through the following formula (10):

[0184] S = S1 * 0.05 + S2 * 0.1 + S3 * 0.85 (10)

[0185] Among them, 0.05, 0.1, and 0.85 are all set parameters, which can also be replaced by other values, and the embodiments of the present disclosure do not limit this.

[0186] Optionally, the lead character, the middle character, and the tail character among the above multiple character types can be divided through the following steps 1)-3):

[0187] According to the number of plays and the number of likes corresponding to each of the multiple characters, respectively determine the performance evaluation values of each of the multiple characters.

[0188] Among them, the multiple characters are all the characters involved in the media resources in the target application. For example, for a music application, the multiple characters are all singers.

[0189] Optionally, the following operations are respectively performed on the multiple characters:

[0190] Determine the performance evaluation value of a person according to the number of plays and the number of likes corresponding to the person within the fourth set time period, the maximum and minimum number of plays among the number of plays corresponding to multiple persons, and the maximum and minimum number of likes among the number of likes corresponding to multiple persons.

[0191] Among them, the fourth set time period is a set historical time period, which can be the same as the above-mentioned third set time period. Taking the media resource as music and a person as a singer as an example, the number of plays corresponding to a singer is the number of songs of the singer among the multiple songs played within the fourth set time period, and the number of likes corresponding to a singer refers to the number of people who like the singer, such as the number of fans of the singer.

[0192] Exemplarily, if the fourth set time period is the recent month, then the performance evaluation value F of a person can be calculated by the following formula (11):

[0193]

[0194] Among them, r is the number of plays corresponding to a person within the recent month, p is the number of likes corresponding to the person, r min 、r max are respectively the minimum and maximum number of plays among the number of plays corresponding to multiple persons within the recent month, and p min 、p max are respectively the minimum and maximum number of likes among the number of likes corresponding to multiple persons.

[0195] 2) Arrange the performance evaluation values of multiple persons from high to low.

[0196] 3) Respectively take the persons ranked in the top N as the head persons, the persons ranked from the N + 1th to the Mth as the waist persons, and the persons ranked after the Mth as the tail persons; where N and M are positive integers, and M > N + 1.

[0197] Among them, the values of N and M can be set as needed. For example, take the persons ranked from 1 to 500 as the head persons, 501 to 8500 as the waist persons, and the rest as the tail persons.

[0198] In the above embodiments of the present disclosure, in the process of calculating the multi-degree evaluation value, the breadth evaluation value, and the depth evaluation value, each parameter in the behavioral feature information of each dimension involved can be set as needed, and is not limited to those shown in the above embodiments.

[0199] Based on the above definitions of the abundance evaluation value, breadth evaluation value, and depth evaluation value, when prompting the user to transition between different level grids, the user can be guided to increase the login duration and login days in the target application to improve abundance, increase the number of language categories and style categories of the media resources played by the user to improve breadth, and increase the playback proportion of the middle and tail figures in different style categories to improve depth.

[0200] Based on the same inventive concept, an embodiment of the present disclosure also provides an appreciation score determination device. The principle of the device for solving problems is similar to that of the method in the above embodiment. Therefore, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0201] As Figure 7 shown, an appreciation score determination device provided by an embodiment of the present disclosure includes a division module 71, an evaluation module 72, and a determination module 73.

[0202] The division module 71 is configured to obtain behavioral feature information of multiple dimensions based on the behavioral information of the target user in the target application; wherein each dimension represents a usage dimension of the target user for the target application.

[0203] The evaluation module 72 is configured to obtain respective dimension evaluation values corresponding to multiple dimensions according to the behavioral feature information of multiple dimensions respectively.

[0204] The determination module 73 is configured to determine the appreciation score of the target user for the target application based on the obtained multiple dimension evaluation values.

[0205] In the embodiment of the present disclosure, by analyzing various behavioral information of the target user in the target application, behavioral feature information of multiple dimensions can be obtained, and then the behavioral feature information of multiple dimensions is evaluated respectively, and the appreciation score of the target user for the target application is determined according to the obtained multiple dimension evaluation values. In this way, by evaluating the behavioral feature information of multiple dimensions, the behavioral information of the target user in the target application can be comprehensively evaluated, and then the appreciation ability of the target user for the target application can be accurately obtained.

[0206] Optionally, the target user includes multiple users. As Figure 8 shown, the device further includes a grid construction module 74, configured to:

[0207] Divide each of the multiple dimensions into multiple levels according to a preset level rule, and construct multiple level grids, and each level grid is composed of levels of the corresponding dimension;

[0208] Obtain the levels of multiple dimensions corresponding to each of the multiple users according to the multiple dimension evaluation values of each of the multiple users;

[0209] Determine the rank grid where the user is located according to the ranks of multiple dimensions corresponding to any user.

[0210] Optionally, when obtaining the ranks of multiple dimensions corresponding to each of multiple users according to the evaluation values of multiple dimensions of each user among the multiple users, the grid construction module 74 is further configured to:

[0211] Normalize the evaluation values of multiple dimensions of each user among the multiple users respectively to obtain the normalized values of multiple dimensions of each user among the multiple users;

[0212] Obtain the ranks of multiple dimensions corresponding to each of multiple users according to the normalized values of multiple dimensions of each user among the multiple users.

[0213] Optionally, the device further includes a grid transition module 75, configured to:

[0214] Obtain the user conversion probability of any first rank grid to any second rank grid among multiple rank grids within a set historical time period;

[0215] For any first rank grid, use the second rank grid corresponding to the maximum user conversion probability between the first rank grid and any second rank grid as the transition rank grid of the first rank grid;

[0216] For any first user within the first rank grid, calculate the similarity between any second user within the transition rank grid and the first user, and recommend the behavior characteristic information of the second user whose similarity reaches the similarity threshold to the first user.

[0217] Optionally, when obtaining the user conversion probability of any first rank grid to any second rank grid among multiple rank grids within a set historical time period, the grid transition module 75 is further configured to:

[0218] Determine the user proportion of the users who transition from any first rank grid to any second rank grid within a set historical time period;

[0219] Use the user proportion as the user conversion probability of the first rank grid to the second rank grid.

[0220] Optionally, the user conversion probability includes the user conversion probabilities of different user portraits;

[0221] Optionally, when obtaining the user conversion probability of any first rank grid to any second rank grid among multiple rank grids within a set historical time period, the grid transition module 75 is further configured to:

[0222] Determine the user proportion of different user portraits of the users who transition from any first rank grid to any of its second rank grids within a set historical time period;

[0223] Take the proportion of users with different user portraits as the user conversion probability of different user portraits for the transition of the first-level grid to the second-level grid.

[0224] Optionally, for any first-level grid, when the second-level grid corresponding to the maximum user conversion probability between the first-level grid and any second-level grid is used as the transition-level grid of the first-level grid, the grid transition module 75 is further configured to:

[0225] For any first-level grid among multiple levels of grids, take the second-level grid corresponding to the maximum user conversion probability of the corresponding user portrait between the first-level grid and any second-level grid as the transition-level grid of the corresponding user portrait within the first-level grid.

[0226] Optionally, for any first user within the first-level grid, when calculating the similarity between any second user within the transition-level grid and the first user and recommending the behavior feature information of the second user whose similarity reaches the similarity threshold to the first user, the grid transition module 75 is further configured to:

[0227] For any first user of any user portrait within the first-level grid, calculate the similarity between the first user and any second user within the corresponding transition-level grid, and recommend the behavior feature information of the second user whose similarity reaches the similarity threshold to the first user.

[0228] Optionally, when calculating the similarity between any second user within the transition-level grid and any first user within the first-level grid, the grid transition module 75 is further configured to:

[0229] For any first user within the first-level grid, calculate the similarity between any second user within the transition-level grid and the first user based on the behavior information of the first user in the target application and the behavior information of any second user in the transition-level grid in the target application.

[0230] Optionally, the multiple dimensions include multiplicity, breadth, and depth; the partitioning module 71 is further configured to:

[0231] Obtain the behavior feature information of multiplicity, the behavior feature information of breadth, and the behavior feature information of depth based on the behavior information of the target user in the target application.

[0232] Optionally, the behavior information of the target user in the target application at least includes: login information, media resource operation information;

[0233] Obtaining the behavior feature information of multiplicity, the behavior feature information of breadth, and the behavior feature information of depth based on the behavior information of the target user in the target application includes:

[0234] Determine the target user's login time in the target application based on the login information, and use the login time information as multiple behavioral feature information;

[0235] Based on the media resource operation information, determine type information of the media resource played by the target user in the target application and character type information in the played media resource;

[0236] The type information of the played media resource is used as the breadth behavior characteristic information, and the type information of the characters in the played media resource is used as the depth behavior characteristic information.

[0237] Alternatively, if Figure 9 As shown, the evaluation module 72 also includes:

[0238] The multi-degree evaluation submodule 721 is used to obtain a multi-degree evaluation value of the multi-degree according to the login time information of the target user in the target application;

[0239] The breadth evaluation submodule 722 is used to obtain a breadth evaluation value of the breadth according to the type information of the media resource played by the target user in the target application;

[0240] The depth assessment submodule 723 is used to obtain a depth assessment value of the depth according to the character type information in the media resources played by the target user in the target application.

[0241] Optionally, the login duration information includes login hours and login days; the multi-degree evaluation submodule 721 is also used for:

[0242] According to the number of login hours and the number of login days of the target user in the target application within the first set time period, a multi-degree evaluation value is obtained.

[0243] Optionally, the type information of the media resource includes a language category and a style category;

[0244] The breadth assessment submodule 722 is also used to:

[0245] Determining a first breadth evaluation value according to the language category and style category of the media resource played by the target user in the target application within the second set time period;

[0246] Determine the second breadth evaluation value based on the language category and style category of the media resources preferred by the target users;

[0247] The breadth evaluation value is determined based on the first breadth evaluation value, the second breadth evaluation value, the number of media resources that the target user likes, and the playback duration of the media resources within the second set time period.

[0248] Optionally, when determining the first breadth evaluation value according to the language category and style category of the media resources played by the target user in the target application within the second set time period, the breadth evaluation sub-module 722 is further configured to:

[0249] Determine the first breadth evaluation value according to the number of language categories, the maximum language category proportion, the number of style categories, and the maximum style category proportion of the media resources played by the target user in the target application within the second set time period.

[0250] Optionally, when determining the second breadth evaluation value according to the language category and style category of the media resources liked by the target user, the breadth evaluation sub-module 722 is further configured to:

[0251] Determine the second breadth evaluation value according to the number of language categories of the media resources liked by the target user, the number of acceptable language categories of the media resources, the number of style categories of the media resources liked by the target user, and the number of acceptable style categories of the media resources.

[0252] Optionally, the device further includes a preference category acquisition module, configured to:

[0253] If the proportion of a language category in the media resource like list of the target user reaches a first preset ratio, and the number of media resources of a language category reaches a first preset quantity, then determine a language category as the language category of the media resources liked by the target user;

[0254] If the proportion of a language category in the media resource like list of the target user does not reach the first preset ratio, and the number of media resources of a language category does not reach the first preset quantity, then determine a language category as the language category of the media resources acceptable to the target user;

[0255] If the proportion of a style category in the media resource like list of the target user reaches a second preset ratio, and the number of media resources of a style category reaches a second preset quantity, then determine a style category as the style category of the media resources liked by the target user;

[0256] If the proportion of a style category in the media resource like list of the target user does not reach the second preset ratio, and / or, the number of media resources of a style category does not reach the second preset quantity, then determine a style category as the style category of the media resources acceptable to the target user.

[0257] Optionally, the depth evaluation sub-module 723 is further configured to:

[0258] Determine the proportion evaluation value according to the respective quantity proportions of multiple character types in the media resources played by the target user in the target application;

[0259] Determine a playback evaluation value according to the proportion of the playback times of each of multiple character types played by a target user in a target application within a third set time period;

[0260] Determine a like degree evaluation value according to the number of likes of each of multiple character types by the target user in the target application;

[0261] Based on the proportion evaluation value, the playback evaluation value, and the like degree evaluation value, determine a depth evaluation value of depth.

[0262] Optionally, the multiple character types at least include a head character, a waist character, and a tail character; the apparatus further includes a character type acquisition module, configured to:

[0263] Determine the performance evaluation value of each of multiple characters according to the playback times and the number of likes corresponding to each of the multiple characters respectively;

[0264] Arrange the performance evaluation values of each of the multiple characters in descending order;

[0265] Take the top N characters as head characters respectively, the characters ranked from the (N + 1)-th to the M-th as waist characters respectively, and the characters ranked after the M-th as tail characters respectively; where N and M are positive integers, and M > N + 1.

[0266] Optionally, determine the performance evaluation value of each of multiple characters according to the playback information and the number of likes corresponding to each of the multiple characters respectively, and the character type acquisition module is further configured to:

[0267] For each of the multiple characters, perform the following operations respectively:

[0268] According to the playback times and the number of likes of a character within a fourth set time period, the maximum and minimum playback times among the playback times of multiple characters, and the maximum and minimum numbers of likes among the numbers of likes of multiple characters, determine the performance evaluation value of a character.

[0269] For the convenience of description, the above parts are divided into respective modules and described separately according to functions. Of course, when implementing the present application, the functions of each module can be implemented in the same or multiple software or hardware.

[0270] Those skilled in the art of the technical field to which the present application pertains can understand that various aspects of the present application can be implemented as a system, a method, or a program product. Therefore, various aspects of the present application can be specifically implemented in the following forms, namely: a complete hardware implementation manner, a complete software implementation manner (including firmware, microcode, etc.), or an implementation manner combining hardware and software aspects, which can be collectively referred to as "circuit", "module", or "system" here.

[0271] Regarding the device in the above embodiments, the specific implementation manners of each module have been described in detail in the embodiments related to the method, and will not be elaborated herein.

[0272] Based on the same inventive concept, an embodiment of the present disclosure further provides an electronic device. The principle of the electronic device for solving problems is similar to that of the method in the above embodiments. Therefore, the implementation of the electronic device can refer to the implementation of the method, and the repeated parts will not be described again.

[0273] Referring to Figure 10 As shown, the electronic device may include a processor 1002 and a memory 1001. The memory 1001 provides the program instructions and data stored in the memory 1001 to the processor 1002. In the embodiments of the present disclosure, the memory 1001 may be used to store the program for multimedia resource processing in the embodiments of the present disclosure.

[0274] The processor 1002 is configured to execute the method in any of the above method embodiments by calling the program instructions stored in the memory 1001. For example Figure 2 a method for determining appreciation scores provided by the embodiment shown.

[0275] In the embodiments of the present disclosure, the specific connection medium between the above-mentioned memory 1001 and the processor 1002 is not limited. In the embodiments of the present disclosure Figure 10 it is shown that the memory 1001 and the processor 1002 are connected through a bus 1003. The bus 1003 is represented by a thick line in Figure 10 The connection manners between other components are only for illustrative purposes and are not limiting. The bus 1003 may be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity of representation, Figure 10 only a thick line is used to represent it in

[0276] The memory may include a Read-Only Memory (ROM) and a Random Access Memory (RAM), and may also include a Non-Volatile Memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located far from the aforementioned processor.

[0277] The above-mentioned processor may be a general-purpose processor, including a central processing unit, a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit, a field programmable gate array, or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0278] The embodiments of the present disclosure also provide a computer storage medium. The computer-readable storage medium stores a computer program. The processor of the electronic device reads the computer program from the computer-readable storage medium, and the processor executes the computer program, so that the electronic device executes the appreciation score determination method in any of the above method embodiments.

[0279] In a specific implementation process, the computer storage medium may include: various storage media that can store program codes, such as a universal serial bus flash drive (USB), a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc.

[0280] Based on the same inventive concept as the above method embodiments, the embodiments of the present application provide a computer program product or a computer program. The computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the steps of any of the above appreciation score determination methods.

[0281] The program product may adopt any combination of one or more readable media. The readable media may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0282] Those skilled in the art should understand that the embodiments of the present disclosure can be provided as a method, a system, or a computer program product. Therefore, the present disclosure can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present disclosure can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0283] The present disclosure is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to the present disclosure. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in Figure 1 one or more of the processes and / or Figure 1 blocks or multiple blocks.

[0284] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufacture including instruction means that realizes the functions specified in Figure 1 one or more of the processes and / or Figure 1 blocks or multiple blocks.

[0285] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in Figure 1 one or more of the processes and / or Figure 1 blocks or multiple blocks.

[0286] Obviously, those skilled in the art can make various modifications and variations to the present disclosure without departing from the spirit and scope of the present disclosure. Thus, if these modifications and variations of the present disclosure fall within the scope of the claims of the present disclosure and their equivalent technologies, the present disclosure is also intended to include these modifications and variations.

Claims

1. A method for determining appreciation scores, characterized in that, Including: Based on the behavior information of the target user in the target application, obtain behavior feature information in multiple dimensions; wherein, each dimension represents a usage dimension of the target user for the target application; Respectively obtain the dimension evaluation values corresponding to the multiple dimensions according to the behavior feature information of the multiple dimensions; Based on the obtained multiple dimension evaluation values, determine the appreciation score of the target user on the target application, and the appreciation score represents the appreciation ability; The target user includes multiple users, and the method further includes: Divide each of the multiple dimensions into multiple levels according to a preset level rule, and construct multiple level grids, and each level grid is composed of the levels of the corresponding dimension; Obtain the levels of the multiple dimensions corresponding to each of the multiple users according to the multiple dimension evaluation values of each of the multiple users; Determine the level grid where the user is located according to the levels of the multiple dimensions corresponding to any user; the level grid where the user is located represents the appreciation level of the user on the target application; Obtain the user conversion probability of any first level grid in the multiple level grids transitioning to any second level grid within a set historical time period; For any first level grid, use the second level grid corresponding to the maximum user conversion probability between the first level grid and any second level grid as the transition level grid of the first level grid; For any first user in the first level grid, calculate the similarity between any second user in the transition level grid and the first user, and recommend the behavior information of the second user whose similarity reaches the similarity threshold to the first user.

2. The method according to claim 1, wherein The obtaining the levels of the multiple dimensions corresponding to each of the multiple users according to the multiple dimension evaluation values of each of the multiple users includes: Perform normalization processing on the multiple dimension evaluation values of each of the multiple users respectively to obtain the multiple dimension normalization values of each of the multiple users; Obtain the levels of the multiple dimensions corresponding to each of the multiple users according to the multiple dimension normalization values of each of the multiple users.

3. The method according to claim 1, wherein The obtaining the user conversion probability of any first level grid in the multiple level grids transitioning to any second level grid within a set historical time period includes: Determine the user proportion of the users transitioning from any first level grid to any second level grid within the set historical time period; Use the user proportion as the user conversion probability of the first level grid transitioning to the second level grid.

4. The method according to claim 1, wherein The user conversion probability includes the user conversion probabilities of different user portraits; The obtaining the user conversion probability of any first level grid in the multiple level grids transitioning to any second level grid within a set historical time period includes: Determine the user proportion of different user portraits of any first level grid transitioning to any of its second level grids within the set historical time period; Use the user proportion of different user portraits as the user conversion probability of different user portraits of the first level grid transitioning to the second level grid.

5. The method according to claim 4, characterized in that, For any of the first-level cells, taking the second-level cell corresponding to the maximum user conversion probability between the first-level cell and any second-level cell as the transition-level cell of the first-level cell includes: For any first-level cell among the multiple-level cells, taking the second-level cell corresponding to the maximum user conversion probability of the corresponding user portrait between the first-level cell and any second-level cell as the transition-level cell of the corresponding user portrait within the first-level cell.

6. The method according to claim 5, characterized in that, For any first user within the first-level cell, calculating the similarity between any second user within the transition-level cell and the first user, and recommending the behavior information of the second user whose similarity reaches the similarity threshold to the first user includes: For the first user of any user portrait within the first-level cell, calculating the similarity between the first user and any second user within the corresponding transition-level cell, and recommending the behavior information of the second user whose similarity reaches the similarity threshold to the first user.

7. The method according to claim 1, wherein For any first user within the first-level cell, calculating the similarity between any second user within the transition-level cell and the first user includes: For any first user within the first-level cell, based on the behavior information of the first user in the target application and the behavior information of any second user in the transition-level cell in the target application, calculating the similarity between any second user within the transition-level cell and the first user.

8. The method according to any one of claims 1 to 7, characterized in that, The multiple dimensions include abundance, breadth, and depth; obtaining the behavior feature information of multiple dimensions based on the behavior information of the target user in the target application includes: Based on the behavior information of the target user in the target application, obtaining the behavior feature information of abundance, the behavior feature information of breadth, and the behavior feature information of depth.

9. The method according to claim 8, wherein The behavior information of the target user in the target application at least includes: login information, media resource operation information; Obtaining the behavior feature information of abundance, the behavior feature information of breadth, and the behavior feature information of depth based on the behavior information of the target user in the target application includes: Based on the login information, determining the login duration information of the target user in the target application, and taking the login duration information as the behavior feature information of abundance; Based on the media resource operation information, determining the type information of the media resources played by the target user in the target application and the character type information in the played media resources; Taking the type information of the played media resources as the behavior feature information of breadth, and taking the character type information in the played media resources as the behavior feature information of depth.

10. The method according to claim 9, wherein Respectively obtaining the dimension evaluation values corresponding to the multiple dimensions according to the behavior feature information of the multiple dimensions includes: According to the login duration information of the target user in the target application, obtaining the abundance evaluation value of abundance; Obtain the breadth evaluation value of the breadth according to the type information of the media resources played by the target user in the target application; Obtain the depth evaluation value of the depth according to the character type information in the media resources played by the target user in the target application.

11. The method according to claim 10, characterized in that, The login duration information includes the number of login hours and the number of login days; The obtaining the multiplicity evaluation value of the multiplicity according to the login duration information of the target user in the target application includes: Obtain the multiplicity evaluation value of the multiplicity according to the number of login hours and the number of login days of the target user in the target application within the first set time period.

12. The method according to claim 10, wherein The type information of the media resources includes language categories and style categories; The obtaining the breadth evaluation value of the breadth according to the type information of the media resources played by the target user in the target application includes: Determine the first breadth evaluation value according to the language categories and style categories of the media resources played by the target user in the target application within the second set time period; Determine the second breadth evaluation value according to the language categories and style categories of the media resources liked by the target user; Based on the first breadth evaluation value, the second breadth evaluation value, the number of media resources liked by the target user, and the playback duration of the media resources within the second set time period, determine the breadth evaluation value.

13. The method according to claim 12, wherein The determining the first breadth evaluation value according to the language categories and style categories of the media resources played by the target user in the target application within the second set time period includes: Determine the first breadth evaluation value according to the number of language categories, the maximum language category ratio, the number of style categories, and the maximum style category ratio of the media resources played by the target user in the target application within the second set time period.

14. The method according to claim 12, wherein The determining the second breadth evaluation value according to the language categories and style categories of the media resources liked by the target user includes: Determine the second breadth evaluation value according to the number of language categories of the media resources liked by the target user, the number of acceptable language categories of the media resources, the number of style categories of the liked media resources, and the number of acceptable style categories of the media resources.

15. The method according to claim 14, wherein, The method further includes: If the ratio of a language category in the media resource like list of the target user reaches the first preset ratio and the number of media resources of the one language category reaches the first preset quantity, determine the one language category as the language category of the media resources liked by the target user; If the ratio of a language category in the media resource like list of the target user does not reach the first preset ratio and the number of media resources of the one language category does not reach the first preset quantity, determine the one language category as the language category of the media resources acceptable to the target user; If the ratio of a style category in the media resource like list of the target user reaches the second preset ratio and the number of media resources of the one style category reaches the second preset quantity, determine the one style category as the style category of the media resources liked by the target user; If the proportion of a style category in the media resource like list of the target user does not reach the second preset ratio, and / or the number of media resources of the style category does not reach the second preset quantity, then determine that the style category is the style category of the media resources acceptable to the target user.

16. The method according to claim 10, wherein The obtaining of the depth evaluation value of the depth according to the character type information in the media resources played by the target user in the target application includes: Determine a proportion evaluation value according to the respective quantity proportions of multiple character types in the media resources played by the target user in the target application; Determine a play evaluation value according to the respective play proportion of the multiple character types played by the target user in the target application within a third set time period; Determine a like degree evaluation value according to the respective number of likes of the multiple character types by the target user in the target application; Based on the proportion evaluation value, the play evaluation value and the like degree evaluation value, determine the depth evaluation value of the depth.

17. The method according to claim 16, characterized in that, The multiple character types at least include head characters, waist characters and tail characters; the method further includes: According to the respective play times and the number of likes corresponding to multiple characters, determine the respective performance evaluation values of the multiple characters; Arrange the respective performance evaluation values of the multiple characters in descending order; Respectively use the top N characters as head characters, the characters ranked from the (N + 1)-th to the M-th as waist characters, and the characters ranked after the M-th as tail characters; where N and M are positive integers, and M > N + 1.

18. The method according to claim 16, wherein Respectively determine the respective performance evaluation values of the multiple characters according to the respective play information and the number of likes corresponding to the multiple characters, including: For multiple characters, respectively perform the following operations: According to the play times and the number of likes corresponding to a character within a fourth set time period, the maximum play times and the minimum play times among the play times corresponding to multiple characters, and the maximum number of likes and the minimum number of likes among the number of likes corresponding to multiple characters, determine the performance evaluation value of the character.

19. An appreciation score determination device, characterized in that, Include: A partitioning module, configured to obtain behavior feature information of multiple dimensions based on the behavior information of the target user in the target application; where each dimension represents a usage dimension of the target user for the target application; An evaluation module, configured to respectively obtain the respective dimension evaluation values corresponding to the multiple dimensions according to the behavior feature information of the multiple dimensions; A determination module, configured to determine the appreciation score of the target user on the target application based on the obtained multiple dimension evaluation values, where the appreciation score represents the appreciation ability; The target user includes multiple users, and the device further includes a grid construction module, configured to: Divide each of the multiple dimensions into multiple levels according to a preset level rule, and construct multiple level grids, where each level grid is composed of the levels of the corresponding dimension; According to the multiple dimension evaluation values of each user among the multiple users, obtain the levels of the multiple dimensions corresponding to each user; Determine the rank grid where the user is located according to the ranks of multiple dimensions corresponding to any user; the rank grid where the user is located represents the appreciation rank of the user on the target application; The device further includes a grid transition module, configured to: Obtain the user conversion probability of any first rank grid among the multiple rank grids transitioning to any second rank grid within a set historical time period; For any first rank grid, use the second rank grid corresponding to the maximum user conversion probability between the first rank grid and any second rank grid as the transition rank grid of the first rank grid; For any first user within the first rank grid, calculate the similarity between any second user within the transition rank grid and the first user, and recommend the behavior information of the second user whose similarity reaches the similarity threshold to the first user.

20. The device according to claim 19, characterized in that, When obtaining the ranks of multiple dimensions corresponding to each of the multiple users according to the evaluation values of multiple dimensions of each user among the multiple users, the grid construction module is further configured to: Perform normalization processing on the evaluation values of multiple dimensions of each user among the multiple users respectively to obtain the normalized values of multiple dimensions of each user among the multiple users; Obtain the ranks of multiple dimensions corresponding to each of the multiple users according to the normalized values of multiple dimensions of each user among the multiple users.

21. The device according to claim 19, characterized in that, When obtaining the user conversion probability of any first rank grid among the multiple rank grids transitioning to any second rank grid within a set historical time period, the grid transition module is further configured to: Determine the user proportion of the users transitioning from any first rank grid to any second rank grid within the set historical time period; Use the user proportion as the user conversion probability of the first rank grid transitioning to the second rank grid.

22. The device according to claim 19, wherein The user conversion probability includes the user conversion probabilities of different user portraits; When obtaining the user conversion probability of any first rank grid among the multiple rank grids transitioning to any second rank grid within a set historical time period, the grid transition module is further configured to: Determine the user proportion of different user portraits of any first rank grid transitioning to any of its second rank grids within the set historical time period; Use the user proportion of different user portraits as the user conversion probability of different user portraits of the first rank grid transitioning to the second rank grid.

23. The device according to claim 22, characterized in that, When, for any first rank grid, using the second rank grid corresponding to the maximum user conversion probability between the first rank grid and any second rank grid as the transition rank grid of the first rank grid, the grid transition module is further configured to: For any first rank grid among the multiple rank grids, use the second rank grid corresponding to the maximum user conversion probability of the corresponding user portrait between the first rank grid and any second rank grid as the transition rank grid of the corresponding user portrait within the first rank grid.

24. The device according to claim 23, characterized in that, When calculating the similarity between any second user in the transition level grid and the first user for any first user in the first level grid and recommending the behavior information of the second user whose similarity reaches the similarity threshold to the first user, the grid transition module is further configured to: For any first user with a user portrait in the first level grid, calculate the similarity between the first user and any second user in the corresponding transition level grid, and recommend the behavior information of the second user whose similarity reaches the similarity threshold to the first user.

25. The device according to claim 19, characterized in that, When calculating the similarity between any second user in the transition level grid and the first user for any first user in the first level grid, the grid transition module is further configured to: For any first user in the first level grid, calculate the similarity between any second user in the transition level grid and the first user based on the behavior information of the first user in the target application and the behavior information of any second user in the transition level grid in the target application.

26. The device according to any one of claims 19 to 25, characterized in that The multiple dimensions include abundance, breadth, and depth; the partitioning module is further configured to: Partition the behavior feature information of the target user in the target application into behavior feature information of abundance, behavior feature information of breadth, and behavior feature information of depth.

27. The device according to claim 26, characterized in that, The behavior feature information of the target user in the target application at least includes: login information, type information of the media resources played, and character type information in the media resources played. When partitioning the behavior feature information of the target user in the target application into behavior feature information of abundance, behavior feature information of breadth, and behavior feature information of depth, the partitioning module is further configured to: Use the login information of the target user in the target application as the behavior feature information of abundance; use the type information of the media resources played by the target user in the target application as the behavior feature information of breadth, and use the character type information in the media resources played by the target user in the target application as the behavior feature information of depth.

28. The device according to claim 27, characterized in that, The evaluation module further includes: An abundance evaluation sub-module, configured to obtain the abundance evaluation value of abundance according to the login duration information of the target user in the target application; A breadth evaluation sub-module, configured to obtain the breadth evaluation value of breadth according to the type information of the media resources played by the target user in the target application; A depth evaluation sub-module, configured to obtain the depth evaluation value of depth according to the character type information in the media resources played by the target user in the target application.

29. The device according to claim 28, wherein, The abundance evaluation sub-module is further configured to: Obtain the abundance evaluation value of abundance according to the login hours and login days of the target user in the target application within a first set time period.

30. The device according to claim 28, characterized in that, The type information of the media resources includes language categories and style categories; The breadth evaluation sub-module is further configured to: Determine a first breadth evaluation value according to the language categories and style categories of the media resources played by the target user in the target application within a second set time period. Determine a second breadth evaluation value according to the language category and style category of the media resources liked by the target user; Based on the first breadth evaluation value, the second breadth evaluation value, the number of media resources liked by the target user, and the playback duration of the media resources within the second set time period, determine the breadth evaluation value.

31. The device according to claim 30, characterized in that, When determining the first breadth evaluation value according to the language category and style category of the media resources played by the target user in the target application within the second set time period, the breadth evaluation sub-module is further configured to: Determine the first breadth evaluation value according to the number of language categories, the maximum language category proportion, the number of style categories, and the maximum style category proportion of the media resources played by the target user in the target application within the second set time period.

32. The device according to claim 30, characterized in that, When determining the second breadth evaluation value according to the language category and style category of the media resources liked by the target user, the breadth evaluation sub-module is further configured to: Determine the second breadth evaluation value according to the number of language categories of the media resources liked by the target user, the number of acceptable language categories of the media resources, the number of style categories of the liked media resources, and the number of acceptable style categories of the media resources.

33. The device according to claim 32, wherein, The device further includes a preference category acquisition module, configured to: If the proportion of a language category in the media resource like list of the target user reaches a first preset ratio, and the number of media resources of the one language category reaches a first preset number, determine the one language category as the language category of the media resources liked by the target user; If the proportion of a language category in the media resource like list of the target user does not reach the first preset ratio, and the number of media resources of the one language category does not reach the first preset number, determine the one language category as the language category of the media resources acceptable to the target user; If the proportion of a style category in the media resource like list of the target user reaches a second preset ratio, and the number of media resources of the one style category reaches a second preset number, determine the one style category as the style category of the media resources liked by the target user; If the proportion of a style category in the media resource like list of the target user does not reach the second preset ratio, and / or, the number of media resources of the one style category does not reach the second preset number, determine the one style category as the style category of the media resources acceptable to the target user.

34. The device according to claim 28, characterized in that, The depth evaluation sub-module is further configured to: Determine a proportion evaluation value according to the respective quantity proportions of multiple character types in the media resources played by the target user in the target application; Determine a playback evaluation value according to the respective playback times proportions of the multiple character types played by the target user in the target application within a third set time period; Determine a like degree evaluation value according to the respective like numbers of the multiple character types by the target user in the target application; Based on the proportion evaluation value, the playback evaluation value, and the like degree evaluation value, determine the depth evaluation value of the depth.

35. The device according to claim 34, wherein The multiple character types at least include a head character, a waist character, and a tail character; the device further includes a character type acquisition module, configured to: Determine the performance evaluation values of the multiple characters respectively according to the playback times and the number of likes corresponding to each of the multiple characters; Arrange the performance evaluation values of the multiple characters in descending order; Respectively use the top N characters as head characters, the characters ranked from the (N + 1)-th to the M-th as waist characters, and the characters ranked after the M-th as tail characters; where N and M are positive integers, and M > N + 1.

36. The device according to claim 34, wherein When determining the performance evaluation values of the multiple characters respectively according to the playback information and the number of likes corresponding to each of the multiple characters, the character type acquisition module is further configured to: For the multiple characters, respectively perform the following operations: According to the playback times and the number of likes corresponding to one character within a fourth set time period, the maximum and minimum playback times among the playback times corresponding to the multiple characters, and the maximum and minimum numbers of likes among the numbers of likes corresponding to the multiple characters, determine the performance evaluation value of the one character.

37. An electronic device, characterized in that, It includes a processor and a memory, wherein the memory stores program code, and when the program code is executed by the processor, the processor is caused to execute the steps of any one of claims 1 to 18.

38. A computer-readable storage medium, characterized in that, It includes program code, and when the program code runs on an electronic device, the program code is used to cause the electronic device to execute the steps of any one of claims 1 to 18.

Images