Preference calculation method, device, electronic device and storage medium

Abstract

The present disclosure relates to the field of computer technology, and in particular to a preference calculation method, device, electronic device and storage medium, which respectively determine the resource preference scores between the target object and the corresponding resource data based on the operation behavior information corresponding to each resource data associated with the target object, wherein the resource preference score represents the score corresponding to the preference degree of the target object for the resource data; and respectively determine the dimension preference scores of the target object under each attribute dimension based on each resource preference score and each attribute dimension corresponding to the dimension weight. In this way, the accuracy of preference calculation can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a preference calculation method, device, electronic device and storage medium. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the disclosure that are recited in the claims. No description herein is admitted to be prior art by inclusion in this section.

[0003] At present, in order to recommend preferred resource data to users, such as songs and videos, we can calculate the user's preferences under different attribute dimensions, and select the resource data under the attribute dimension with the highest preference from the resource database for recommendation.

[0004] In the related art, when calculating the preference under different attribute dimensions, the user's preference under different attribute dimensions is usually calculated based on the user's red heart operation on resource data, or based on the user's subjective rating feedback.

[0005] However, this method in the related art only focuses on users who have performed the red heart operation on resource data. For users who have not performed the red heart operation, it is impossible to determine the preference under different attribute dimensions. Moreover, the method of calculating the preference based on the user's subjective rating feedback is not very accurate. Summary of the invention

[0006] In the related art, only the preference of users who have performed the red heart operation under different attribute dimensions can be determined. Therefore, the accuracy of this method of calculating the preference is not high. In view of the above technical problems, an improved method is needed to calculate the preference of all users, thereby improving the accuracy of calculating the preference.

[0007] On the one hand, an embodiment of the present disclosure provides a preference calculation method, including:

[0008] Based on the operation behavior information corresponding to each resource data associated with the target object, respectively determine the resource preference score between the target object and the corresponding resource data, wherein the resource preference score represents the score corresponding to the preference degree of the target object for the resource data;

[0009] Based on the resource preference scores and the dimension weights corresponding to the attribute dimensions, the dimension preference scores of the target object under the attribute dimensions are determined respectively.

[0010] On the one hand, an embodiment of the present disclosure provides a preference calculation device, including:

[0011] A first processing module is used to determine the resource preference score between the target object and the corresponding resource data based on the operation behavior information corresponding to each resource data associated with the target object, wherein the resource preference score represents the score corresponding to the preference degree of the target object for the resource data;

[0012] The second processing module is used to determine the dimension preference score of the target object in each attribute dimension based on each resource preference score and the dimension weight corresponding to each attribute dimension.

[0013] On the one hand, an embodiment of the present disclosure provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any of the above methods when executing the computer program.

[0014] On the one hand, an embodiment of the present disclosure provides a computer-readable storage medium having computer program instructions stored thereon, and the computer program instructions implement the steps of any of the above methods when executed by a processor.

[0015] On the one hand, an embodiment of the present disclosure provides a computer program product, wherein the computer program product includes a computer program stored on a computer-readable storage medium, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the steps of any of the above methods are implemented.

[0016] The preference calculation method, device, electronic device and storage medium provided by the embodiments of the present disclosure respectively determine the resource preference scores between the target object and the corresponding resource data based on the operational behavior information corresponding to each resource data associated with the target object, and respectively determine the dimension preference scores of the target object under each attribute dimension based on each resource preference score and the dimension weight corresponding to each attribute dimension. In this way, there is no need for subjective scoring feedback, but rather the resource preference scores between the target object and the corresponding resource data are respectively determined based on the comprehensive operational behavior information corresponding to each resource data. Determining the resource preference scores through objective operational behavior information can improve the accuracy of determining the resource preference scores; and, determining the dimension preference scores of the target object under each attribute dimension through the resource preference scores and the dimension weights corresponding to each attribute dimension can reduce the influence of the popularity of resource data on the target object preference, thereby improving the accuracy of determining the dimension preference scores. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and other objects, features and advantages of the exemplary embodiments of the present disclosure will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present disclosure are shown in an exemplary and non-limiting manner, in which:

[0018] Figure 1 A schematic diagram of an application scenario of the preference calculation method provided in an embodiment of the present disclosure;

[0019] Figure 2 A schematic diagram of a flow chart of a preference score calculation method in an embodiment of the present disclosure;

[0020] Figure 3 A schematic diagram of a process for identifying opinion leader objects in an embodiment of the present disclosure;

[0021] Figure 4 This is an example diagram of the calculation process of the dimension preference score in the embodiment of the present disclosure;

[0022] Figure 5 This is an example diagram of the calculation process of the resource identification capability score in the embodiment of the present disclosure;

[0023] Figure 6 A schematic diagram of the structure of a preference calculation device provided in an embodiment of the present disclosure;

[0024] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0025] The principles and spirit of the present disclosure will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present disclosure, and are not intended to limit the scope of the present disclosure in any way. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0026] Those skilled in the art will appreciate that the embodiments of the present disclosure may be implemented as a system, device, apparatus, method or computer program product. Therefore, the present disclosure may be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0027] It should be understood herein that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction rather than having any limiting meaning.

[0028] For ease of understanding, the terms involved in the embodiments of the present disclosure are explained below:

[0029] Inverse document frequency (IDF): The main idea of ​​IDF is that among all documents, the fewer documents contain the term, the larger the IDF value, indicating that the term has good category discrimination ability.

[0030] The principle and spirit of the present disclosure are explained in detail below with reference to several representative embodiments of the present disclosure. SUMMARY OF THE INVENTION

[0032] In the related art, when calculating the preference under different attribute dimensions, it is usually based on the user's red heart operation, or determined through the user's subjective feedback score. Therefore, this method of determining the preference in the related art is not very accurate.

[0033] In order to solve the above problems, the present disclosure provides a preference calculation method, which specifically includes the following steps: based on the operation behavior information corresponding to each resource data associated with the target object, respectively determine the resource preference score between the target object and the corresponding resource data; based on each resource preference score and the dimension weight corresponding to each attribute dimension, respectively determine the dimension preference score of the target object under each attribute dimension, which can improve the accuracy of determining the resource preference score, thereby improving the accuracy of determining the dimension preference score; and when determining the preference score of each dimension, based on the dimension weight corresponding to each attribute dimension, determine the user's preference under different attribute dimensions, which can reduce the influence of the popularity of the resource data on the calculation of the preference, and further improve the accuracy of the calculation.

[0034] After introducing the basic principles of the present disclosure, various non-limiting embodiments of the present disclosure are described in detail below.

[0035] Application Scenario Overview

[0036] refer to Figure 1 , which is a schematic diagram of an application scenario of the preference calculation method provided in an embodiment of the present disclosure. The application scenario includes a user terminal 101 and a server 102. The user terminal 101 and the server 102 may be connected via a wired or wireless communication network. The user terminal 101 includes but is not limited to electronic devices such as desktop computers, mobile phones, mobile computers, tablet computers, media players, smart wearable devices, smart TVs, vehicle-mounted devices, personal digital assistants (PDAs), etc. The server 102 may 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, and big data and artificial intelligence platforms.

[0037] The target object can send a resource data request to the server 102 through the user terminal 101, so that after receiving the resource data request for the target object sent by the user terminal 101, the server 102 determines the resource preference scores between the target object and the corresponding resource data based on the operation behavior information corresponding to each resource data associated with the target object, and then, based on each resource preference score and the dimension weight corresponding to each attribute dimension, determines the dimension preference score of the target object under each attribute dimension, and based on the determined dimension preference scores, determines the corresponding resource data from the resource library, and sends the determined resource data to the user terminal 101, so that the user terminal 101 plays the received resource data.

[0038] Exemplary Methods

[0039] Combine the following Figure 1 The speech detection method according to the exemplary embodiment of the present disclosure is described by using the application scenario. It should be noted that the above application scenario is only shown to facilitate understanding of the spirit and principle of the present disclosure, and the embodiments of the present disclosure are not limited in this respect. On the contrary, the embodiments of the present disclosure can be applied to any applicable scenario.

[0040] See also Figure 2 FIG. 1 is a flow chart of a method for calculating preference scores in an embodiment of the present disclosure, which specifically includes:

[0041] S20: Based on the operation behavior information corresponding to each resource data associated with the target object, respectively determine the resource preference score between the target object and the corresponding resource data.

[0042] The resource preference score represents the score corresponding to the target object's preference for resource data; the operation behavior information represents the target object's operation behavior on resource data, such as collection behavior, single loop, etc.

[0043] In an embodiment of the present disclosure, in response to a preference score calculation instruction for a target object triggered by relevant staff, each resource data associated with the target object is obtained, and the operation behavior information corresponding to each resource data is obtained. Then, based on the operation behavior information corresponding to each resource data, the resource preference score between the target object and the corresponding operation resource data is determined respectively.

[0044] The following takes the operation behavior information including a plurality of preference behavior information, a plurality of interactive behavior information and a plurality of playback behavior information as an example, and S20 in the embodiment of the present disclosure is described in detail, specifically including:

[0045] S201: Based on each piece of preference behavior information, each piece of interaction behavior information and each piece of playback behavior information, in combination with a preset score conversion strategy, a preference behavior score, an interaction score and a playback score are obtained.

[0046] In the embodiments of the present disclosure, based on each preference behavior information and in combination with a corresponding score conversion strategy, the preference behavior score between the target object and the resource data is determined; based on each interaction behavior information and in combination with a corresponding score conversion strategy, the interaction score between the target object and the resource data is determined; based on each playback behavior information and in combination with a corresponding score conversion strategy, the playback score between the target object and the resource data is determined.

[0047] Optionally, when determining the preferred behavior score, interaction score, and playback score, specifically include:

[0048] S2011: Using a preset first score conversion method and each preference behavior information to obtain a preference behavior score.

[0049] In the disclosed embodiment, a first score conversion method is determined, and based on each piece of preference behavior information and in combination with the determined first score conversion method, a preference behavior score between the target object and the resource data is determined.

[0050] Among them, preference behavior information: preference behavior information represents the preference information fed back by the target object through the interactive behavior with the playing resource data during the playing of the resource data, such as favorites, hearts, etc., and the addition and deletion of the blacklist. In order to ensure the stability of the data, the historical cumulative behavior is calculated as the long-term feature, and the behavior in the past 30 days is calculated as the recent feature.

[0051] Optionally, when determining the preference score between the target object and the resource data, specifically including:

[0052] S2011-1: Determine the preference scores to be processed corresponding to each piece of preference behavior information respectively.

[0053] In the disclosed embodiment, for any preference behavior information in each preference behavior information, based on the preference behavior information and the corresponding relationship between the preference behavior information and the preference score to be processed, the preference score to be processed corresponding to the preference behavior information is determined.

[0054] In specific implementation, a correspondence table between preference behavior information and preference scores to be processed is pre-set in the resource database, so that the preference scores to be processed corresponding to each preference behavior information can be determined based on the correspondence table between preference behavior information and preference scores to be processed. See Table 1, which is a correspondence table between preference behavior information and preference scores to be processed in the embodiment of the present disclosure.

[0055] Table 1.

[0056] Preference behavior information Preference score to be processed Historical Collection 1 History not collected 0 Historical Heart 1 History without red hearts 0 Recent Hearts 1 No recent hearts 0 Recent Collections 1 No recent collection 0 Recently deleted favorites 1 No collections deleted recently 0 like 1 dislike 0

[0057] For example, based on Table 1, when the preference behavior information is historical collection, the corresponding pending preference score is 1; when the preference behavior information is historical non-collection, the corresponding pending preference score is 0.

[0058] S2011-2: Determine a preference behavior score based on each preference score to be processed and the corresponding first weight.

[0059] In the disclosed embodiment, after determining each preference score to be processed, the preference behavior score between the target object and the resource data is determined based on each preference score to be processed and a first weight corresponding to each preference score to be processed.

[0060] Among them, the preference behavior score can be expressed as:

[0061] Preference behavior score = (preference score corresponding to historical collection behavior + preference score corresponding to historical red heart behavior) * first weight a + preference score corresponding to recent red heart behavior * first weight b + preference score corresponding to recent collection behavior * first weight c - preference score corresponding to recent deletion of collection behavior * first weight d - preference score corresponding to dislike behavior * first weight e.

[0062] For example, assuming that the resource data is song A, the first weight a is 0.4, the first weight b is 0.2, the first weight c is 0.2, the first weight d is 0.1, and the first weight e is 0.1. The preference behavior information of song A are historical collections, historical hearts, recent hearts, recent uncollected, recent undeleted collections, and likes. Based on Table 1, it can be concluded that the preference behavior score between the target object and song A = (1+1) *0.4+1*0.2+0*0.2-0*0.1+1*0.1=1.1.

[0063] S2012: Using a preset second score conversion method and each interactive behavior information to obtain an interactive score.

[0064] In the disclosed embodiment, a preset second score conversion method is determined, and based on each interactive behavior information, the interaction score between the target object and the resource data is determined in combination with the determined second score conversion method.

[0065] Among them, interactive behavior information: interaction refers to the user's comments, sharing, downloading and other behaviors on the content in the process of consuming resource data.

[0066] Optionally, when determining the interaction score, specifically include:

[0067] S2012-1: Determine the amount of interactive behavior information corresponding to each interactive type.

[0068] In the disclosed embodiment, each interaction type is determined, and the amount of interaction behavior information under each interaction type is counted to obtain the amount of information corresponding to each interaction type.

[0069] For example, assuming that each interaction type is comment, share, and download, count the number of interaction behavior information whose interaction type is comment in each interaction behavior information, and obtain the number of information corresponding to comment, and count the number of interaction behavior information whose interaction type is share in each interaction behavior information, and obtain the number of information corresponding to sharing, and count the number of interaction behavior information whose interaction type is download in each interaction behavior information, and obtain the number of information corresponding to download.

[0070] S2012-2: Determine the interaction score based on the quantity of each information and the corresponding second weight.

[0071] In the disclosed embodiment, the interaction score is determined based on the quantity of each message and the second weight corresponding to each interaction type.

[0072] Among them, when the interaction type includes comment, share and download, the interaction score = number of comment information * second weight a + number of shared information * second weight b + number of downloaded information * second weight c.

[0073] S2013: Using a preset third score conversion method and each playback behavior information to obtain a playback score.

[0074] In the disclosed embodiment, a preset third score conversion method is determined, and based on each playback behavior information, the playback score between the target object and the resource data is determined in combination with the determined third score conversion method.

[0075] Among them, the playback behavior information in the embodiment of the present disclosure includes 7 types of user playback behaviors, which can be active playback information or passive playback information. In the embodiment of the present disclosure, the playback score between the target object and the resource data can be determined based on each active playback information and each passive playback information. The process of determining the playback score in the embodiment of the present disclosure is described below, specifically including:

[0076] Based on each active play information and each passive play information in each play behavior information, combined with a preset third score conversion method, a play score is determined.

[0077] In the disclosed embodiment, since the playback behavior information is divided into active playback information and passive playback information, each playback behavior information includes each active playback information and each passive playback information. Based on each active playback information and each passive playback information, combined with a preset third score conversion method, the playback score between the target object and the resource data is determined.

[0078] Among them, active playback information represents the playback count information collected after the target object plays resource data through active channels. The active playback information is any one of the following: active effective playback count, loop effective playback count, search effective playback count, and active complete playback count; passive playback information represents the playback count information collected after the target object plays resource data through passive channels. The passive playback information is any one of the following: passive effective playback count and passive complete playback count. Active channels include: favorite song playlists, artist pages, album pages, searches, etc. Passive channels include: daily recommendations, playlist squares, private FM, etc.

[0079] Optionally, in an embodiment of the present disclosure, when the active playback information includes active valid playback times, loop valid playback times, search valid playback times, and active complete playback times, and each passive playback information includes passive valid playback times and passive complete playback times, a possible implementation method for determining the playback score is provided, specifically including:

[0080] S2013-1: Obtain the legal number of plays by determining the cumulative result between the active valid number of plays and the passive valid number of plays.

[0081] In the disclosed embodiment, the active valid playback times and the passive valid playback times are added to obtain a cumulative result, and the cumulative result is used as the legal playback times. Therefore, in the disclosed embodiment, the legal playback times = active valid playback times + passive valid playback times.

[0082] Among them, the number of active effective playbacks represents the number of times the target object plays resource data through active channels, and the playback duration exceeds a preset duration threshold; the number of passive effective playbacks represents the number of times the target object plays resource data through passive channels, and the playback duration exceeds a preset duration threshold. The preset duration threshold can be set to 15 seconds, which is not limited in the embodiments of the present disclosure.

[0083] S2013-2: Obtain the number of complete playbacks by determining the cumulative result between the number of active complete playbacks and the number of passive complete playbacks.

[0084] In the embodiment of the present disclosure, the active complete play times and the passive complete play times are added to obtain a cumulative result, and the cumulative result is used as the complete play times. The complete play times in the embodiment of the present disclosure can be expressed as: complete play times = active complete play times + passive complete play times.

[0085] Among them, the number of active complete playbacks represents the number of times the target object plays resource data through active channels, and the playback end time is consistent with the end time of the resource data content; the number of passive complete playbacks represents the number of times the target object plays resource data through passive channels, and the playback end time is consistent with the end time of the resource data content.

[0086] S2013-3: Determine the total playback duration corresponding to each resource data.

[0087] In the disclosed embodiment, when determining the total playback duration corresponding to each resource data, the total playback duration associated with the corresponding resource data is obtained from the resource database.

[0088] It should be noted that the total playback time is recorded in the following manner: when the resource data is played, the duration of the resource data played this time is recorded, and the recorded duration is added to the total playback time of the resource data obtained before the playback, so as to obtain the total playback time of the resource data after the current playback. Therefore, the total playback time of the resource data is updated in real time after each playback is completed.

[0089] S2013-4: Determine the cumulative result of the playback durations corresponding to the target resource data whose playback durations exceed the preset playback duration threshold in each resource data.

[0090] In the embodiment of the present disclosure, based on the playback duration corresponding to each resource data, resource data whose playback duration exceeds a duration threshold are determined from each resource data, and the determined resource data are used as target resource data, and the sum of the playback duration corresponding to each target resource data is calculated to obtain a duration accumulation result.

[0091] S2013-5: Determine the average duration according to the duration accumulation result and the amount of resource data of the target resource data.

[0092] In the embodiment of the present disclosure, the number of resource data of the target resource data in each resource data is determined, and the ratio between the accumulated result of the duration and the number of resource data is calculated to obtain the average duration.

[0093] For example, assuming that the resource data is songs, and there are 10 target songs determined from the songs, then the number of resource data corresponding to the target songs is determined to be 10, and the cumulative duration is determined to be 3000s. Then the ratio between the cumulative duration and the number of resource data is calculated as 3000 / 10=300s, and the average duration is 300s.

[0094] S2013-6: Determine the playback score based on the active effective playback times, loop effective playback times, search effective playback times, complete playback times, legal playback times, total playback time, average playback time and the corresponding third weight.

[0095] In the disclosed embodiment, the playback score is determined based on the number of active effective playbacks, the number of loop effective playbacks, the number of search effective playbacks, the number of complete playbacks, the number of legal playbacks, the total playback time and the average playback time, combined with the corresponding third weight.

[0096] The playback score in the embodiment of the present disclosure can be expressed as:

[0097]

[0098] Among them, x is the first parameter, for example, x can be 1, and for example, x is 2. The first parameter is used to ensure that the value of 0 will not appear during the natural logarithm conversion. For example, assuming that the number of active and effective playback times is 0 and x is 1, then ln(active and effective playback times + x) is ln1.

[0099] It should be noted that the number of effective loop playbacks represents the number of times the target object loops and plays resource data, and the playback duration exceeds the preset duration threshold; the number of effective search playbacks represents the number of times the target object actively searches for resource data and plays it, and the playback duration exceeds the preset duration threshold.

[0100] In addition, it should be noted that in the embodiment of the present disclosure, since the active effective playback times, the loop effective playback times, the search effective playback times, the complete playback times, and the legal playback times have a significant impact on the playback score when the values ​​are small, but as the values ​​increase, the impact gradually decays, and the logarithmic curve distribution can just meet the actual needs. Therefore, the natural logarithm conversion method is used to calculate the playback score.

[0101] S202: Determine a resource preference score between the target object and the resource data based on the preference behavior score, the interaction score, and the playback score.

[0102] In the disclosed embodiment, after the preferred behavior score, interaction score and playback score are determined, the resource preference score between the target object and the resource data can be calculated based on the preferred behavior score, interaction score and playback score.

[0103] Optionally, when calculating the resource preference score, specifically include:

[0104] S2021: Accumulate the preference behavior score, interaction score, and playback score to obtain a cumulative result.

[0105] In the disclosed embodiment, the sum of the preference behavior score, the interaction score and the playback score is calculated to obtain a cumulative result.

[0106] S2022: Using the accumulated result as the resource preference score between the target object and the resource data.

[0107] In the disclosed embodiment, the accumulated result obtained by calculation is used as the resource preference score between the target object and the resource data, and the resource preference score = preference behavior score + interaction score + playback score.

[0108] Furthermore, in the embodiment of the present disclosure, the product of the preference behavior score, the interaction score and the playback score can also be calculated to obtain a cumulative result, and the obtained cumulative result is used as the resource preference score between the target object and the resource data.

[0109] Of course, the natural index of the preference behavior score, the interaction score and the play score can also be calculated to obtain the resource preference score between the target object and the resource data. The method of determining the resource preference score is not limited in the embodiment of the present disclosure.

[0110] S21: Based on the preference scores of each resource and the dimension weights corresponding to each attribute dimension, the dimension preference scores of the target object under each attribute dimension are determined respectively.

[0111] In the disclosed embodiment, based on the resource preference scores corresponding to each resource data and the dimension weights corresponding to each attribute dimension, the dimension preference scores of the target object under each attribute dimension are determined respectively. The dimension preference score is the score corresponding to the preference degree for the song under the resource attribute dimension.

[0112] Optionally, when determining the dimension preference score, specifically include:

[0113] S211: Determine the cumulative score of each resource preference score in the same attribute dimension.

[0114] In the embodiment of the present disclosure, for any attribute dimension, the following operations are performed:

[0115] Determine the resource data belonging to the attribute dimension, and obtain the resource preference score corresponding to each determined resource data, add up the determined resource preference scores, and obtain the cumulative score of each resource preference score of the same attribute dimension. Based on this operation, the cumulative score of the resource preference score of each attribute dimension can be obtained.

[0116] For example, assuming that the resource preference score of song A is x1, which belongs to the attribute dimension "light", the corresponding resource preference score of song B is x2, which belongs to the attribute dimension "light", the resource preference score of song C is x3, which belongs to the attribute dimension "light", and the resource preference score of song D is x4, which belongs to the attribute dimension "ancient style". Therefore, song A, song B and song C all belong to the attribute dimension "light", then calculate the cumulative score between x1, x2 and x3, so the cumulative score is = x1+x2+x3.

[0117] S212: Based on the dimension weights and accumulated scores corresponding to each attribute dimension, respectively determine the dimension preference scores of the target object under each attribute dimension.

[0118] The weight of each dimension is determined based on the total number of target objects and the number of target objects under the corresponding attribute dimension.

[0119] In the embodiment of the present disclosure, for any attribute dimension, the following operations are performed respectively:

[0120] Based on the accumulated scores and the dimension weights corresponding to the attribute dimensions, the dimension preference scores of the target object under the attribute dimensions are determined.

[0121] In specific implementation, the dimension preference score of the target object under the attribute dimension can be determined by calculating the product between the dimension weight and the accumulated score. The dimension weight in the embodiment of the present disclosure is determined by performing a natural logarithm operation on the total number of target objects and the number of target objects under the corresponding attribute dimension.

[0122] Among them, the dimension weight can be expressed as:

[0123]

[0124] In the disclosed embodiment, the resource preference score is determined by combining the interactive behavior information, the playback behavior information and the preference behavior information, so as to improve the coverage of the resource preference score at the user level and the recognition accuracy of the resource preference score.

[0125] Based on the above embodiments, in the embodiments of the present disclosure, after obtaining the preference scores of each dimension, the opinion leader objects of the target object in each attribute dimension can be determined based on the preference scores of each dimension. Figure 3 FIG. 1 is a flow chart of identifying opinion leader objects in an embodiment of the present disclosure, which specifically includes:

[0126] S30: Based on the preference scores of each dimension and the level information of each resource data, respectively determine the resource identification ability scores of the target object in each attribute dimension.

[0127] Among them, the resource identification ability score represents the score corresponding to the target object's ability to identify potential resource data.

[0128] In the embodiment of the present disclosure, for any attribute dimension in each attribute dimension, the following operations are performed respectively: based on the dimension preference score under the attribute dimension and the level information of each resource data, the resource identification ability score of the target object under the attribute dimension is determined. Thus, the resource identification ability score of the target object under each attribute dimension can be obtained.

[0129] Optionally, in the embodiment of the present disclosure, a possible implementation method is provided for determining the resource identification capability score, specifically including:

[0130] S301: Determine the first resource data collected by the target object from various resource data.

[0131] In the embodiment of the present disclosure, when calculating the resource identification ability score of the target object, it is necessary to perform the calculation based on the resource data collected by the target object. Therefore, the first resource data collected by the target object is determined from various resource data.

[0132] In specific implementation, when the target object collects resource data, a label "Collection" is set for the collected resource data. Therefore, when determining the first resource data collected by the target object, the resource data corresponding to the label "Collection" is selected from various resource data as the first resource data.

[0133] In the embodiment of the present disclosure, when the target object performs a collection operation on a certain resource data, the resource data is added to the collection set. When it is necessary to determine the first resource data collected by the target object from various resource data, each resource data in the collection set can be directly used as the first resource data collected by the target object.

[0134] S302: Determine a level score of first resource data, and determine a first number of resource data that have undergone level upgrade in the first resource data.

[0135] In the disclosed embodiment, a level score corresponding to the first resource data is determined, and a first number of resource data of which the level is upgraded in the first resource data is determined.

[0136] Optionally, a possible implementation method for performing S302 is provided, which specifically includes:

[0137] S302-1: Determine any second resource data of the first resource data under any attribute dimension.

[0138] In the disclosed embodiment, second resource data of the first resource data under any attribute dimension is determined.

[0139] For example, assuming that the attribute dimension is the "rock" dimension under the genre dimension, the second resource data of the first resource data under the "rock" attribute dimension is determined.

[0140] S302-2: Determine a first play score and a first level value of the second resource data according to the number of people or the number of times the second resource data is played when it is collected.

[0141] In the disclosed embodiment, the first level value of the second resource data when it is collected is determined, and the first playback score of the second resource data is determined according to the number of people or the number of times the second resource data is played when it is collected.

[0142] Optionally, determining the first playback score of the second resource data may be implemented in the following three ways:

[0143] The first method is to determine the number of people who played the second resource data when it was collected, and determine the first playback score based on the number of people who played the second resource data when it was collected and the scaling processing parameters.

[0144] The scaling parameter has a value range of 0.1 to 0.5, and is used to scale the number of times of playback.

[0145] The second method is to determine the number of times the second resource data is played when it is collected, and determine the first play score based on the number of times the second resource data is played when it is collected and the scaling processing parameter.

[0146] The third method: determine the number of people who play the second resource data when it is collected, and the number of times the second resource data is played when it is collected, and add the number of people who play the second resource data when it is collected and the number of times the second resource data is played when it is collected to obtain the addition result, and determine the first playback score based on the addition result and the scaling processing parameters.

[0147] S302-3: Determine a second playback score and a second level value of the second resource data according to the number of people or the number of times the second resource data is currently played in the resource library.

[0148] In the disclosed embodiment, the second level value of the second resource data currently in the resource library is determined, and the second playback score of the second resource data is determined according to the number of people or the number of times the second resource data is currently played in the resource library.

[0149] Optionally, determining the second playback score of the second resource data may be implemented in the following three ways:

[0150] The first method is to determine the number of people who are currently playing the second resource data in the resource library, and determine the second playback score based on the number of people who are currently playing the second resource data in the resource library and the scaling processing parameters.

[0151] The scaling parameter has a value range of 0.1 to 0.5, and is used to scale the number of times of playback.

[0152] It should be noted that, in the embodiment of the present disclosure, the scaling processing parameter used when determining the second playback score may be the same as the scaling processing parameter used when determining the first playback score.

[0153] The second method: determining the number of times the second resource data is currently played in the resource library, and determining the second play score based on the number of times the second resource data is currently played in the resource library and the scaling processing parameters.

[0154] The third method: determine the number of times the second resource data is currently played in the resource library, and the number of people who play the second resource data in the current resource library, and add the number of times the second resource data is currently played in the resource library and the number of people who play the second resource data in the current resource library to obtain the addition result, and determine the second playback score based on the addition result and the scaling processing parameters.

[0155] S302-4: Determine the first day from the current time according to the second resource data collection time.

[0156] In the embodiment of the present disclosure, the first day from the current time when the second resource data is collected is determined.

[0157] For example, assuming that the user collected the second resource data on February 3, and the current time is February 7, the first day is determined to be 4. The current time represents the time for identifying the opinion leader object.

[0158] S302-5: Determine the number of days between the current time and the first time that the level value of the second resource data reaches the first level value before the collection time.

[0159] In the embodiment of the present disclosure, the first time when the level value of the second resource data reaches the first level value before the collection time is determined, which is the second day from the current time.

[0160] For example, assuming that the current time is November 22, the first level value is B, the collection time of the second resource data is November 20, and the time when the level value of the second resource data reaches B is November 19, then the second day is determined to be 3.

[0161] S302-6: Determine the level score of the second resource data under the attribute dimension according to the first playback score, the second playback score, the first day number and the second day number.

[0162] In the disclosed embodiment, the ratio between the first day number and the second day number is calculated, and the product between the first play score, the second play score and the ratio is calculated to determine the grade score.

[0163] Optionally, in the embodiment of the present disclosure, the level score may also be obtained in the following manner:

[0164]

[0165] Wherein, p represents the scaling processing parameter.

[0166] In specific implementation, since the first resource data may include multiple resource data, the grade score corresponding to each resource data is determined respectively, and the determined grade scores are added together to obtain the grade score corresponding to the first resource data.

[0167] S302-7: Determine a first quantity of second resource data that has undergone level upgrade under the attribute dimension according to the first level value and the second level value.

[0168] S303: Determine the resource identification capability score of the target object in each attribute dimension according to the preference score of each dimension, the grade score and the first quantity, wherein the grade information includes the grade score and the first quantity.

[0169] In the disclosed embodiment, the following operations are performed for each attribute dimension: a natural logarithm operation is performed on the first quantity under an attribute dimension to obtain the processed first quantity, and the product of the dimension preference score, the grade score and the processed first quantity under an attribute dimension is calculated to obtain the resource identification capability score of the target object under the attribute dimension.

[0170] Among them, resource identification ability score = dimension preference score × level score × ln (first quantity).

[0171] Optionally, a possible implementation method for determining the resource identification capability score is provided, which specifically includes:

[0172] S303-1: Determine the precision score and the recall score.

[0173] Optionally, when determining the precision score and recall score, specifically include:

[0174] S303-1-1: Determine the second number of resource data that have undergone level upgrade in the first resource data.

[0175] In the embodiment of the present disclosure, it is determined whether each resource data in the first resource data has been upgraded, and resource data that has been upgraded is screened out from each resource data in the first resource data, and a second number of resource data that has been upgraded is counted. The resource data having been upgraded indicates that the level of the resource data at the current time is higher than the level when it was collected.

[0176] For example, assuming that the level of song A is C when it is collected, and the level of song A is B at the current time, and level B is higher than level C, it is determined that the level of song A is upgraded.

[0177] S303-1-2: Determine the third quantity of the second resource data in any attribute dimension in the first resource data.

[0178] In the embodiment of the present disclosure, statistics are performed on the second resource data under any attribute dimension in the first resource data to obtain a third quantity.

[0179] For example, the first resource data includes songs A, B, C, D, E, F and G. The resource attribute of A is rock, the resource attribute of B is ancient style, the resource attribute of C is ancient style, the resource attribute of D is small fresh, the resource attribute of E is rock, the resource attribute of F is small fresh, and the resource attribute of G is small fresh. Then, in the first resource data, the second resource data under the "rock" attribute dimension is determined to be song A and song E, and the third quantity is 2.

[0180] S303-1-3: Determine a fourth number of resource data that have undergone level upgrades among all resource data belonging to the attribute dimension in the resource library.

[0181] In the disclosed embodiment, resource data belonging to the attribute dimension is found among all resource data in the resource library, resource data that has undergone level upgrade is determined from the found resource data, and a fourth number of resource data that has undergone level upgrade is determined.

[0182] For example, suppose that the resource data belonging to the "ancient style" attribute dimension found in the resource library are songs A, B and C, among which A was level one when collected and its current level is level two, song A was level one when collected and its current level is level three, and C was level two when collected and its current level is level two, then A and B are determined to be resource data that have undergone level upgrades, and the fourth quantity is determined to be 2.

[0183] S303-1-4: Calculate the precision score of the target object in the attribute dimension according to the second quantity and the third quantity.

[0184] In the disclosed embodiment, after the second number and the third number are obtained, a precision score of the target object in the attribute dimension is obtained based on the second number and the third number.

[0185] Optionally, in an embodiment of the present disclosure, a possible implementation method is provided for determining the precision score: calculating the ratio between the third number and the second number, and calculating the product between the determined ratio and the percentage coefficient to obtain the precision score of the target object in the attribute dimension.

[0186] Among them, the precision score can be expressed as:

[0187]

[0188] The percentage coefficient may be, for example, 100%, which is not limited in the embodiments of the present disclosure.

[0189] S303-1-5: Calculate the recall score of the target object in the attribute dimension according to the second quantity and the fourth quantity.

[0190] In the embodiment of the present disclosure, after the second number and the fourth number are obtained, a recall score of the target object in the attribute dimension is obtained based on the second number and the fourth number.

[0191] Optionally, in an embodiment of the present disclosure, a possible implementation method is provided for determining the recall rate score: calculating the ratio between the fourth number and the second number, and calculating the product between the determined ratio and the percentage coefficient to obtain the recall rate score of the target object under the attribute dimension.

[0192]

[0193] The percentage coefficient may be, for example, 100%, which is not limited in the embodiments of the present disclosure.

[0194] S303-2: Determine the resource identification capability score of the target object in each attribute dimension according to the preference score, grade score, first quantity, precision score and recall score of each dimension.

[0195] In the disclosed embodiment, the following operations are performed for each attribute dimension: a natural logarithm operation is performed on the first quantity under an attribute dimension to obtain the processed first quantity, and the product of the dimension preference score, the grade score, the recall score, the precision score and the processed first quantity under an attribute dimension is calculated to obtain the resource identification capability score of the target object under the attribute dimension.

[0196] Among them, the resource identification ability score can be expressed as:

[0197]

[0198] Optionally, in the embodiment of the present disclosure, the dimension preference scores may be standardized, and the resource identification capability scores may be determined based on the standardized dimension preference scores, specifically including:

[0199] The preference scores of each dimension are standardized, and the resource identification ability scores of the target object in each attribute dimension are determined according to the standardized preference scores of each dimension, the grade scores and the first quantity.

[0200] In the embodiments of the present disclosure, the following operations are performed for each attribute dimension: the dimension preference score under an attribute dimension is standardized to obtain the standardized dimension preference score, and the natural logarithm operation is performed on the first quantity under the attribute dimension to obtain the processed first quantity, and the product of the standardized dimension preference score, the grade score and the processed first quantity under the attribute dimension is calculated to obtain the resource identification capability score of the target object under the attribute dimension.

[0201] Therefore, the resource identification ability score can be expressed as:

[0202] Resource identification ability score = standardized dimension preference score × level score × ln (first quantity)

[0203] When calculating the standardized dimension preference score, the standardized dimension preference score is determined based on the dimension preference score of one attribute dimension, the minimum value among the dimension preference scores of each attribute dimension, and the maximum value among the preference scores of each attribute dimension.

[0204] The dimension preference score can be expressed as:

[0205]

[0206] Among them, min (dimension preference score) is the minimum value of the dimension preference scores corresponding to each attribute dimension, and max (dimension preference score) is the maximum value of the dimension preference scores corresponding to each attribute dimension.

[0207] S31: Based on the resource identification ability scores of each target object in each attribute dimension, opinion leaders in each attribute dimension are determined from each target object.

[0208] In the disclosed embodiment, the following operations are performed for each attribute dimension respectively: based on the resource identification capability score of each target object in any attribute dimension, an opinion leader object in the attribute dimension is determined from each target object.

[0209] In specific implementation, the target object with the highest resource identification ability score among all target objects can be used as the opinion leader object under this attribute dimension, indicating that he has a certain appreciation and influence under this attribute dimension.

[0210] In the disclosed embodiment, based on the dimension preference score, the first day number and the score level, the opinion leader objects under different attribute dimensions can be accurately identified, and through the fusion calculation of multiple indicators, the random factors that cause the resource data to suddenly become hot resource data can be relatively well avoided; and after the opinion leader objects are determined, the resource data that may become popular can be mined in advance through the behaviors and preferences of the opinion leader objects under each attribute dimension, thereby realizing the automation of resource data operation; by cultivating the target objects' cognition of the opinion leader objects and establishing a reward mechanism for the opinion leader objects, the user experience and retention within the site can be improved, thereby realizing a self-closed loop of user operation.

[0211] Based on the above embodiments, see Figure 4 As shown, it is an example diagram of the calculation process of the dimension preference score in the embodiment of the present disclosure:

[0212] First, obtain the user's operation behavior information, which includes preference behavior information, interactive behavior information, and playback behavior information. Preference behavior information includes hearts, favorites, and blacklisting; playback behavior information includes the number of plays and the duration of play; interactive behavior information includes sharing, downloading, and commenting. The long-term in the long-term and short-term represents recent hearts, recent favorites, and recent blacklisting, while the short-term represents historical hearts, historical favorites, and historical blacklisting; the positive in the positive and negative represents adding hearts, adding favorites, and deleting blacklists, while the negative represents deleting hearts, deleting favorites, and adding blacklists.

[0213] Then, based on the preference behavior information, the preference behavior feature is obtained, based on the playback behavior information, the playback feature is obtained, and based on the interaction behavior information, the interaction feature is obtained. A natural logarithm operation is performed on the preference behavior feature to obtain a preference behavior score, a natural logarithm operation is performed on the playback feature to obtain a playback score, and a natural logarithm operation is performed on the interaction feature to obtain an interaction score. The preference behavior score, the playback score, and the interaction score are added together to obtain a resource preference score, and the resource preference scores of each resource data are added together to obtain the total resource preference score.

[0214] Finally, calculate the product of the artist weight and the sum of the resource preference scores to obtain the artist preference score, calculate the product of the instrument weight and the sum of the resource preference scores to obtain the instrument preference score, calculate the product of the style weight and the sum of the resource preference scores to obtain the style preference score, and calculate the product of the era weight and the sum of the resource preference scores to obtain the era preference score.

[0215] Based on the above embodiment, the calculation process of the resource identification capability score in the embodiment of the present disclosure is described in detail using a specific example. Figure 5 As shown in FIG. 1 , it is an example diagram of the calculation process of the resource identification capability score in an embodiment of the present disclosure, which specifically includes:

[0216] First, determine the songs collected by the user and determine the song level corresponding to each song. Then, determine the difference between the current song level and the level when the song is collected and the difference in the number of plays. For each attribute dimension, based on the dimension preference score under any attribute dimension, calculate the precision, recall rate, first quantity and level score under the attribute dimension. Finally, for each attribute dimension, based on the dimension preference score, precision, recall rate, first quantity and level score under any attribute dimension, calculate the resource identification ability score of the user under any attribute dimension. Based on the above method, the resource identification ability score of the user under each attribute dimension can be obtained.

[0217] Exemplary Devices

[0218] After introducing the method according to the exemplary embodiment of the present disclosure, the preference calculation device according to the exemplary embodiment of the present disclosure will be introduced next.

[0219] like Figure 6 , which is a schematic diagram of the structure of a preference calculation device provided by an embodiment of the present disclosure. In one embodiment, the preference calculation device 60 includes: a first processing module 601 , a second processing module 602 , a third processing module 603 and an identification module 604 .

[0220] The first processing module 601 is used to determine the resource preference score between the target object and the corresponding resource data based on the operation behavior information corresponding to each resource data associated with the target object, wherein the resource preference score represents the score corresponding to the preference degree of the target object for the resource data;

[0221] The second processing module 602 is used to determine the dimension preference score of the target object in each attribute dimension based on each resource preference score and the dimension weight corresponding to each attribute dimension.

[0222] Optionally, the operation behavior information includes each preference behavior information, each interaction behavior information and each playback behavior information;

[0223] Then the first processing module 601 is further used for:

[0224] For any resource data associated with the target object, perform the following operations:

[0225] Based on the information of each preference behavior, the information of each interaction behavior and the information of each playback behavior, in combination with a preset score conversion strategy, a preference behavior score, an interaction score and a playback score are obtained;

[0226] A resource preference score between the target object and the resource data is determined based on the preference behavior score, the interaction score, and the playback score.

[0227] Optionally, when obtaining the preference behavior score, the interaction score and the playback score based on the preference behavior information, the interaction behavior information and the playback behavior information in combination with a preset score conversion strategy, the first processing module 601 is further used to:

[0228] Using a preset first score conversion method and each of the preference behavior information to obtain a preference behavior score; and,

[0229] Using a preset second score conversion method and each interactive behavior information to obtain an interactive score; and,

[0230] The preset third score conversion method and the playback behavior information are used to obtain the playback score.

[0231] Optionally, when the preset first score conversion method and the information of each preference behavior are used to obtain the preference behavior score, the first processing module 601 is further used to:

[0232] Respectively determine the to-be-processed preference scores corresponding to each of the preference behavior information;

[0233] Based on each preference score to be processed and the corresponding first weight, a preference behavior score is determined.

[0234] Optionally, when the interaction score is obtained by using the preset second score conversion method and the interaction behavior information, the first processing module 601 is further used to:

[0235] Determine the information quantity of the interactive behavior information corresponding to each interactive type respectively;

[0236] An interaction score is determined based on the quantity of each piece of information and the corresponding second weight.

[0237] Optionally, when the preset third score conversion method and the playback behavior information are used to obtain the playback score, the first processing module 601 is further used to:

[0238] Based on each active play information and each passive play information in the play behavior information, combined with a preset third score conversion method, a play score is determined.

[0239] Optionally, each active playback information includes active valid playback times, loop valid playback times, search valid playback times, and active complete playback times, and each passive playback information includes passive valid playback times and passive complete playback times. When determining the playback score based on each active playback information and each passive playback information and in combination with a preset third score conversion method, the first processing module 601 is further used to:

[0240] Obtaining the legal playback number by determining the cumulative result between the active and effective playback number and the passive and effective playback number;

[0241] Obtaining the number of complete plays by determining the cumulative result between the number of active complete plays and the number of passive complete plays;

[0242] Determine the total playback duration corresponding to each of the resource data;

[0243] Determine the cumulative result of the playback durations of the target resource data whose playback durations exceed a preset playback duration threshold value in the resource data;

[0244] Determine the duration average value according to the duration accumulation result and the amount of resource data of the target resource data;

[0245] A playback score is determined based on the active effective playback times, the loop effective playback times, the search effective playback times, the complete playback times, the legal playback times, the total playback time, the average time and the corresponding third weight.

[0246] Optionally, when determining the resource preference score between the target object and the resource data based on the preference behavior score, the interaction score and the playback score, the first processing module 601 is further configured to:

[0247] Accumulate the preference behavior score, the interaction score, and the playback score to obtain an accumulation result;

[0248] The accumulated result is used as the resource preference score between the target object and the resource data.

[0249] Optionally, the second processing module 602 is further used for:

[0250] Determine a cumulative score of the resource preference scores of the same attribute dimension;

[0251] Based on the dimension weights corresponding to each attribute dimension and the accumulated scores, the dimension preference scores of the target object under each attribute dimension are determined respectively, wherein each dimension weight is determined based on the total number of target objects and the number of target objects under the corresponding attribute dimension.

[0252] Optionally, the device further comprises:

[0253] The third determination module 603 is used to determine the resource identification ability score of the target object in each attribute dimension based on the preference score of each dimension and the level information of each resource data, wherein the resource identification ability score represents the score corresponding to the ability of the target object to identify potential resource data;

[0254] The identification module 604 is used to determine the opinion leader objects in each attribute dimension from each target object based on the resource identification capability score of each target object in each attribute dimension.

[0255] Optionally, the third determining module 603 is further configured to:

[0256] Determine the first resource data collected by the target object from the resource data;

[0257] Determine a level score of the first resource data, and determine a first number of resource data of which the level is upgraded in the first resource data;

[0258] The resource identification capability score of the target object in each attribute dimension is determined respectively according to the preference score of each dimension, the grade score and the first quantity, wherein the grade information includes the grade score and the first quantity.

[0259] Optionally, when determining the level score of the first resource data and determining the first number of resource data with upgraded levels in the first resource data, the third determining module 603 is further configured to:

[0260] Determine any second resource data of the first resource data under any attribute dimension;

[0261] Determine a first play score and a first level value of the second resource data according to the number of people or the number of times the second resource data is played when it is collected;

[0262] Determine a second play score and a second level value of the second resource data according to the number of people or the number of times the second resource data is currently played in the resource library;

[0263] Determine the first day from the current time according to the collection time of the second resource data;

[0264] Determine the number of days between the current time and the first time that the level value of the second resource data reaches the first level value before the collection time;

[0265] Determine a level score of the second resource data under the attribute dimension according to the first playback score, the second playback score, the first day number, and the second day number;

[0266] A first quantity of second resource data that has undergone level upgrade under the attribute dimension is determined according to the first level value and the second level value.

[0267] Optionally, when determining the resource identification capability score of the target object in each attribute dimension according to the preference score of each dimension, the grade score and the first quantity, the third processing module 603 is further used to:

[0268] Determine precision and recall scores;

[0269] According to the preference scores, grade scores, first quantity, precision scores and recall scores of each dimension, the resource identification capability scores of the target object in each attribute dimension are determined respectively.

[0270] Optionally, when determining the precision score and the recall score, the third processing module 603 is further used to:

[0271] Determining a second amount of resource data of which level has been upgraded in the first resource data;

[0272] Determine a third quantity of second resource data under any attribute dimension in the first resource data;

[0273] Determining a fourth number of resource data that have been upgraded in level among all resource data belonging to the attribute dimension in the resource library;

[0274] Calculate the precision score of the target object in the attribute dimension according to the second number and the third number;

[0275] A recall score of the target object in the attribute dimension is calculated based on the second number and the fourth number.

[0276] Optionally, when determining the resource identification capability score of the target object in each attribute dimension according to the preference score of each dimension, the grade score and the first quantity, the third processing module 603 is further used to:

[0277] The preference scores of each dimension are standardized, and the resource identification capability scores of the target object in each attribute dimension are determined respectively according to the standardized preference scores of each dimension, the grade scores and the first quantity.

[0278] The preference calculation device provided in the embodiment of the present disclosure adopts the same inventive concept as the above-mentioned preference calculation method and can achieve the same beneficial effects, which will not be described in detail here.

[0279] Based on the same inventive concept as the above preference calculation method, the present disclosure also provides an electronic device, which can be Figure 1 The user terminal in. Figure 7 As shown, the electronic device 70 may include a processor 701 and a memory 702 .

[0280] The processor 701 may be a general-purpose processor, such as a central processing unit (CPU), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, and may implement or execute the methods, steps and logic block diagrams disclosed in the embodiments of the present disclosure. A general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present disclosure may be directly embodied as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0281] The memory 702 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules. The memory may include at least one type of storage medium, such as a flash memory, a hard disk, a multimedia card, a card-type memory, a random access memory (Random Access Memory, RAM), a static random access memory (Static Random Access Memory, SRAM), a programmable read-only memory (Programmable Read Only Memory, PROM), a read-only memory (Read Only Memory, ROM), an electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, EEPROM), a magnetic memory, a disk, an optical disk, etc. The memory is any other medium that can be used to carry or store a desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The memory 702 in the disclosed embodiment can also be a circuit or any other device that can realize a storage function, for storing program instructions and / or data.

[0282] Exemplary Program Products

[0283] An embodiment of the present disclosure provides a computer-readable storage medium for storing computer program instructions used by the above-mentioned electronic device, which includes a program for executing the preference calculation method in any exemplary embodiment of the present disclosure.

[0284] The above-mentioned computer storage media can be any available media or data storage devices that can be accessed by a computer, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MO), etc.), optical storage (such as CD, DVD, BD, HVD, etc.), and semiconductor storage (such as ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)), etc.

[0285] In some possible implementations, various aspects of the present disclosure may also be implemented as a computer program product, which includes program code. When the computer program product is run on a server device, the computer program product is used to enable the server device to execute the steps of the preference calculation method according to various exemplary implementations of the present disclosure described in the above "Exemplary Method" section of this specification.

[0286] The computer program product may use any combination of one or more readable media. The readable medium 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, device or device, or any combination of the above. More specific examples of readable storage media (a non-exhaustive list) 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 disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0287] According to the computer program product for instant messaging application of the embodiment of the present disclosure, it can adopt portable compact disk read-only memory (CD-ROM) and include program code, and can be run on server device. However, the program product of the present disclosure is not limited to this. In this document, computer readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, device or device.

[0288] A readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, wherein readable program code is carried. Such propagated data signals may take a variety of forms, including, but not limited to, electromagnetic signals, optical signals, or any suitable combination of the foregoing. A readable signal medium may also be any readable medium other than a readable storage medium, which may transmit, propagate, or transfer a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0289] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0290] Program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0291] It should be noted that although several units or subunits of the device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more units described above can be embodied in one unit. Conversely, the features and functions of one unit described above can be further divided into multiple units to be embodied.

[0292] In addition, although the operations of the disclosed method are described in a specific order in the drawings, this does not require or imply that the operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0293] Although the spirit and principle of the present disclosure have been described with reference to several specific embodiments, it should be understood that the present disclosure is not limited to the disclosed specific embodiments, and the division of various aspects does not mean that the features in these aspects cannot be combined to benefit, and such division is only for the convenience of expression. The present disclosure is intended to cover various modifications and equivalent arrangements included in the spirit and scope of the attached claims.

Claims

1. A method for calculating preference score, characterized in that: include: Based on the operation behavior information corresponding to each resource data associated with the target object, respectively determine the resource preference score between the target object and the corresponding resource data, wherein the resource preference score represents the score corresponding to the preference degree of the target object for the resource data; Based on the resource preference scores and the dimension weights corresponding to the attribute dimensions, respectively determine the dimension preference scores of the target object under each attribute dimension; The operation behavior information includes each preference behavior information, each interactive behavior information and each playback behavior information; Then, based on the operation behavior information corresponding to each resource data associated with the target object, respectively determining the resource preference score between the target object and the corresponding resource data includes: For any resource data associated with the target object, perform the following operations: Using a preset first score conversion method and each of the preference behavior information to obtain a preference behavior score; and, Using a preset second score conversion method and each interactive behavior information to obtain an interactive score; and, Determine the playback score based on each active playback information and each passive playback information in the playback behavior information and in combination with a preset third score conversion method; The active play information includes active valid play times, loop valid play times, search valid play times, and active complete play times, and the passive play information includes passive valid play times and passive complete play times. Based on the active play information and passive play information, combined with a preset third score conversion method, the play score is determined, including: Obtaining the legal playback number by determining the cumulative result between the active and effective playback number and the passive and effective playback number; Obtaining the number of complete plays by determining the cumulative result between the number of active complete plays and the number of passive complete plays; Determine the total playback duration corresponding to each of the resource data; Determine the cumulative result of the playback durations of the target resource data whose playback durations exceed a preset playback duration threshold value in the resource data; Determine the duration average value according to the duration accumulation result and the amount of resource data of the target resource data; Determine a playback score based on the active effective playback times, the loop effective playback times, the search effective playback times, the complete playback times, the legal playback times, the total playback time, the average of the time and the corresponding third weight; A resource preference score between the target object and the resource data is determined based on the preference behavior score, the interaction score, and the playback score.

2. The method according to claim 1, characterized in that The preferred behavior scores are obtained by using the preset first score conversion method and the preferred behavior information, including: Respectively determine the to-be-processed preference scores corresponding to each of the preference behavior information; Based on each preference score to be processed and the corresponding first weight, a preference behavior score is determined.

3. The method according to claim 1, characterized in that The preset second score conversion method and each interactive behavior information are used to obtain an interactive score, including: Determine the information quantity of the interactive behavior information corresponding to each interactive type respectively; An interaction score is determined based on the quantity of each piece of information and the corresponding second weight.

4. The method according to claim 1, characterized in that Determining a resource preference score between the target object and the resource data based on the preference behavior score, the interaction score, and the playback score includes: Accumulate the preference behavior score, the interaction score, and the playback score to obtain an accumulation result; The accumulated result is used as the resource preference score between the target object and the resource data.

5. The method according to any one of claims 1 to 4, characterized in that: Based on the resource preference scores and the dimension weights corresponding to the attribute dimensions, the dimension preference scores of the target object in each attribute dimension are determined respectively, including: Determine a cumulative score of the resource preference scores of the same attribute dimension; Based on the dimension weights corresponding to each attribute dimension and the accumulated scores, the dimension preference scores of the target object under each attribute dimension are determined respectively, wherein each dimension weight is determined based on the total number of target objects and the number of target objects under the corresponding attribute dimension.

6. The method according to claim 1, characterized in that The method further comprises: Based on the preference scores of each dimension and the level information of each resource data, respectively determine the resource identification ability scores of the target object in each attribute dimension, wherein the resource identification ability scores represent the scores corresponding to the ability of the target object to identify potential resource data; Based on the resource identification capability scores of the target objects in each attribute dimension, opinion leader objects in each attribute dimension are determined from the target objects.

7. The method according to claim 6, characterized in that Based on the preference scores of each dimension and the level information of each resource data, respectively determining the resource identification capability scores of the target object in each attribute dimension, including: Determine the first resource data collected by the target object from the resource data; Determine a level score of the first resource data, and determine a first number of resource data of which the level is upgraded in the first resource data; The resource identification capability score of the target object in each attribute dimension is determined respectively according to the preference score of each dimension, the grade score and the first quantity, wherein the grade information includes the grade score and the first quantity.

8. The method according to claim 7, characterized in that Determining the level score of the first resource data and determining a first number of resource data of which the level is upgraded in the first resource data includes: Determine any second resource data of the first resource data under any attribute dimension; Determine a first play score and a first level value of the second resource data according to the number of people or the number of times the second resource data is played when it is collected; Determine a second play score and a second level value of the second resource data according to the number of people or the number of times the second resource data is currently played in the resource library; Determine the first day from the current time according to the collection time of the second resource data; Determine the number of days between the current time and the first time that the level value of the second resource data reaches the first level value before the collection time; Determine a level score of the second resource data under the attribute dimension according to the first playback score, the second playback score, the first day number, and the second day number; A first quantity of second resource data that has undergone level upgrade under the attribute dimension is determined according to the first level value and the second level value.

9. The method according to claim 7, characterized in that Determining the resource identification capability score of the target object in each attribute dimension according to the preference score of each dimension, the grade score and the first quantity, further comprising: Determine precision and recall scores; According to the preference scores, grade scores, first quantity, precision scores and recall scores of each dimension, the resource identification capability scores of the target object in each attribute dimension are determined respectively.

10. The method according to claim 7, characterized in that Determining the resource identification capability score of the target object in each attribute dimension according to the preference score of each dimension, the grade score and the first quantity, respectively, includes: The preference scores of each dimension are standardized, and the resource identification capability scores of the target object in each attribute dimension are determined respectively according to the standardized preference scores of each dimension, the grade scores and the first quantity.

11. A preference score calculation device, characterized in that: include: A first processing module is used to determine the resource preference score between the target object and the corresponding resource data based on the operation behavior information corresponding to each resource data associated with the target object, wherein the resource preference score represents the score corresponding to the preference degree of the target object for the resource data; A second processing module is used to determine the dimension preference score of the target object under each attribute dimension based on each resource preference score and the dimension weight corresponding to each attribute dimension; The operation behavior information includes each preference behavior information, each interaction behavior information and each playback behavior information; Then the first processing module is further used for: For any resource data associated with the target object, perform the following operations: Using a preset first score conversion method and each of the preference behavior information to obtain a preference behavior score; and, Using a preset second score conversion method and each interactive behavior information to obtain an interactive score; and, Determine the playback score based on each active playback information and each passive playback information in the playback behavior information and in combination with a preset third score conversion method; The active play information includes active valid play times, loop valid play times, search valid play times, and active complete play times, and the passive play information includes passive valid play times and passive complete play times. When determining the play score based on the active play information and the passive play information and in combination with the preset third score conversion method, the first processing module is further used to: Obtaining the legal playback number by determining the cumulative result between the active and effective playback number and the passive and effective playback number; Obtaining the number of complete plays by determining the cumulative result between the number of active complete plays and the number of passive complete plays; Determine the total playback duration corresponding to each of the resource data; Determine the cumulative result of the playback durations of the target resource data whose playback durations exceed a preset playback duration threshold value in the resource data; Determine the duration average value according to the duration accumulation result and the amount of resource data of the target resource data; Determine a playback score based on the active effective playback times, the loop effective playback times, the search effective playback times, the complete playback times, the legal playback times, the total playback time, the average of the time and the corresponding third weight; A resource preference score between the target object and the resource data is determined based on the preference behavior score, the interaction score, and the playback score.

12. The device according to claim 11, characterized in that When the preferred behavior scores are obtained by using the preset first score conversion method and the preferred behavior information, the first processing module is further used to: Respectively determine the to-be-processed preference scores corresponding to each of the preference behavior information; Based on each preference score to be processed and the corresponding first weight, a preference behavior score is determined.

13. The device according to claim 11, characterized in that When the interaction score is obtained by using the preset second score conversion method and the interaction behavior information, the first processing module is further used to: Determine the information quantity of the interactive behavior information corresponding to each interactive type respectively; An interaction score is determined based on the quantity of each piece of information and the corresponding second weight.

14. The device according to claim 11, characterized in that When determining the resource preference score between the target object and the resource data based on the preference behavior score, the interaction score and the playback score, the first processing module is further used to: Accumulate the preference behavior score, the interaction score, and the playback score to obtain an accumulation result; The accumulated result is used as the resource preference score between the target object and the resource data.

15. The device according to any one of claims 11 to 14, characterized in that: The second processing module is also used for: Determine a cumulative score of the resource preference scores of the same attribute dimension; Based on the dimension weights corresponding to each attribute dimension and the accumulated scores, the dimension preference scores of the target object under each attribute dimension are determined respectively, wherein each dimension weight is determined based on the total number of target objects and the number of target objects under the corresponding attribute dimension.

16. The device according to claim 11, characterized in that The device also includes: A third determination module is used to determine the resource identification ability score of the target object in each attribute dimension based on the preference score of each dimension and the level information of each resource data, wherein the resource identification ability score represents the score corresponding to the ability of the target object to identify potential resource data; The identification module is used to determine the opinion leader objects in each attribute dimension from each target object based on the resource identification capability score of each target object in each attribute dimension.

17. The device according to claim 16, characterized in that The third determination module is further used for: Determine the first resource data collected by the target object from the resource data; Determine a level score of the first resource data, and determine a first number of resource data of which the level is upgraded in the first resource data; The resource identification capability score of the target object in each attribute dimension is determined respectively according to the preference score of each dimension, the grade score and the first quantity, wherein the grade information includes the grade score and the first quantity.

18. The device according to claim 17, characterized in that When determining the level score of the first resource data and determining the first number of resource data of which the level is upgraded in the first resource data, the third determining module is further used to: Determine any second resource data of the first resource data under any attribute dimension; Determine a first play score and a first level value of the second resource data according to the number of people or the number of times the second resource data is played when it is collected; Determine a second play score and a second level value of the second resource data according to the number of people or the number of times the second resource data is currently played in the resource library; Determine the first day from the current time according to the collection time of the second resource data; Determine the number of days between the current time and the first time that the level value of the second resource data reaches the first level value before the collection time; Determine a level score of the second resource data under the attribute dimension according to the first playback score, the second playback score, the first day number, and the second day number; A first quantity of second resource data that has undergone level upgrade under the attribute dimension is determined according to the first level value and the second level value.

19. The device according to claim 17, characterized in that When determining the resource identification capability score of the target object in each attribute dimension according to the preference score of each dimension, the grade score and the first quantity, the third determination module is further used to: Determine precision and recall scores; According to the preference scores, grade scores, first quantity, precision scores and recall scores of each dimension, the resource identification capability scores of the target object in each attribute dimension are determined respectively.

20. The device according to claim 17, characterized in that When determining the resource identification capability score of the target object in each attribute dimension according to the preference score of each dimension, the grade score and the first quantity, the third determination module is further used to: The preference scores of each dimension are standardized, and the resource identification capability scores of the target object in each attribute dimension are determined respectively according to the standardized preference scores of each dimension, the grade score and the first quantity.

21. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 10 are implemented.

22. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

Images