Virtual item push method, device, electronic device and readable medium

Abstract

The embodiments of the present disclosure disclose a method, device, electronic device and readable medium for pushing virtual items. A specific implementation of the method includes: filtering a user rating information set; clustering the filtered user rating information set to obtain the filtered user rating information set after clustering as a user rating information group set; determining a user rating information group; generating a user similarity set; determining the similarity of each user that meets the neighbor user filtering condition as a neighbor user similarity set; determining a neighbor user rating information set based on the neighbor user similarity set; determining a target user rating information set; filtering out each target user rating information as a user rating information set to be recommended; determining a virtual item information set to be recommended; pushing the virtual item information set to be recommended to a client to present virtual items to the user corresponding to the client. This implementation can reduce the waste of CPU processor resources.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technology, and in particular to a method, device, electronic device, and readable medium for pushing virtual items. Background Art

[0002] With the development of Internet technology, the Internet virtual goods industry has also seen a booming development. At the same time, there has also been an overload of virtual goods information (for example, stocks, financial products, etc.), a wide variety of products, and difficulty for users to choose. Therefore, how to recommend virtual goods suitable for different users has become an urgent problem to be solved. At present, when pushing virtual goods to users, the usual method is to determine the user similarity between the target user (the user to be recommended) and all users (all users stored on the server accessed by the target user) to determine the user set similar to the target user. Then, based on the preferences of the similar user set for virtual goods, relevant virtual goods are recommended.

[0003] However, when the above method is used to push virtual items to users, the following technical problems often occur:

[0004] Since there are many users or virtual items when recommending virtual items, determining the user similarity between the target user and all users causes the CPU processor to occupy more resources and take longer to process the user similarity, thus wasting CPU processor resources.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure concept and therefore it may contain information that does not form the prior art that is already known in this country to a person of ordinary skill in the art. Summary of the invention

[0006] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0007] Some embodiments of the present disclosure propose a virtual item push method, apparatus, electronic device, and computer-readable medium to solve one or more of the technical problems mentioned in the above background technology section.

[0008] In a first aspect, some embodiments of the present disclosure provide a method for pushing a virtual item, the method comprising: in response to receiving a virtual item recommendation request sent by a client, obtaining a user rating information set; filtering the user rating information set to obtain the filtered user rating information set as a filtered user rating information set; clustering the filtered user rating information set to obtain the clustered filtered user rating information set as a user rating information group set; in response to determining that the user corresponding to the client is a registered user, determining a user rating information group corresponding to the user in the user rating information group set; generating a user rating information group according to the user rating information group and the user rating information corresponding to the user; The method comprises the steps of: generating a user similarity set; determining a neighbor user similarity set as a neighbor user similarity set, determining a neighbor user rating information set according to the neighbor user similarity set; determining a target user rating information set according to the user rating information corresponding to the user and the neighbor user rating information set; filtering out each target user rating information that meets the rating condition to be recommended from the target user rating information set as the user rating information set to be recommended; determining a virtual item information set to be recommended according to the user rating information set to be recommended; and pushing the virtual item information set to be recommended to the client to present virtual items to the user corresponding to the client.

[0009] In a second aspect, some embodiments of the present disclosure provide a virtual item push device, the device comprising: an acquisition unit, configured to acquire a user rating information set in response to receiving a virtual item recommendation request sent by a client; a first screening unit, configured to perform screening processing on the user rating information set, and obtain the filtered user rating information set as a screened user rating information set; a clustering processing unit, configured to perform clustering processing on the screened user rating information set, and obtain the clustered screened user rating information set as a user rating information group set; a first determination unit, configured to determine a user rating information group corresponding to the user in the user rating information group set in response to determining that the user corresponding to the client is a registered user; a generation unit, configured to generate a user similarity set based on the user rating information group and the user rating information corresponding to the user; The second determination unit is configured to determine the similarities of each user in the above-mentioned user similarity set that meets the neighbor user screening condition as the neighbor user similarity set; the third determination unit is configured to determine the neighbor user rating information set based on the above-mentioned neighbor user similarity set; the fourth determination unit is configured to determine the target user rating information set based on the user rating information corresponding to the above-mentioned user and the above-mentioned neighbor user rating information set; the second screening unit is configured to screen out the target user rating information that meets the rating condition to be recommended from the above-mentioned target user rating information set as the user rating information set to be recommended; the fifth determination unit is configured to determine the virtual item information set to be recommended based on the above-mentioned user rating information set to be recommended; the pushing unit is configured to push the above-mentioned virtual item information set to be recommended to the above-mentioned client, so as to present the virtual item to the user corresponding to the above-mentioned client.

[0010] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner of the above-mentioned first aspect.

[0011] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation manner of the above-mentioned first aspect is implemented.

[0012] The above-mentioned embodiments of the present disclosure have the following beneficial effects: through the virtual item push method of some embodiments of the present disclosure, the waste of CPU processor resources can be reduced. Specifically, the reason for the waste of CPU processor resources is that when making virtual item recommendations, the number of users or the number of virtual items is large, and the user similarity between the target user and all users is determined, resulting in a large number of CPU processor resources occupied by the CPU processor for processing user similarity, and a long processing time, thereby resulting in a waste of CPU processor resources. Based on this, the virtual item push method of some embodiments of the present disclosure, first, in response to receiving a virtual item recommendation request sent by a client, obtains a user rating information set. Thus, a user rating information set representing the evaluation of all virtual items by all users can be obtained. Then, the above user rating information set is filtered and processed to obtain the filtered user rating information set as the filtered user rating information set. Thus, the main component virtual item evaluation information representing the user's interest can be obtained. Thus, redundant features can be removed, the data scale can be reduced, and the memory space for storing data can be reduced. Afterwards, the above filtered user rating information set is clustered to obtain the filtered user rating information set after clustering as the user rating information group set. Thus, the user rating information group set can be obtained. Thus, users with similar preferences can be divided into the same cluster. Subsequently, in response to determining that the user corresponding to the above-mentioned client is a registered user, the user rating information group corresponding to the above-mentioned user in the above-mentioned user rating information group set is determined. Thus, the user rating information group with similar preferences to the target user can be determined. Subsequently, a user similarity set is generated based on the above-mentioned user rating information group and the user rating information corresponding to the above-mentioned user. Thus, the similarity of each user with similar preferences to the target user can be determined. Secondly, the similarity of each user in the above-mentioned user similarity set that meets the neighbor user screening condition is determined as the neighbor user similarity set. Thus, each neighbor user similarity set with a high similarity to the target user can be obtained. Then, based on the above-mentioned neighbor user similarity set, a neighbor user rating information set is determined. Thus, a neighbor user rating information set corresponding to each user with a high similarity can be obtained. Afterwards, the target user rating information set is determined based on the user rating information corresponding to the above-mentioned user and the above-mentioned neighbor user rating information set. Thus, based on the rating information of similar users on virtual items, the rating information of the target user on each virtual item can be determined. Subsequently, each target user rating information satisfying the to-be-recommended rating condition is selected from the above target user rating information set as the to-be-recommended user rating information set. Thus, a to-be-recommended user rating information set that represents the target user's high preference can be obtained. Subsequently, a to-be-recommended virtual item information set is determined based on the above to-be-recommended user rating information set. Thus, a to-be-recommended virtual item set can be obtained, thereby determining each virtual item that the user has a high preference for.Finally, the above virtual item information set to be recommended is pushed to the above client to present virtual items to the user corresponding to the client. In this way, each virtual item that the target user prefers can be sent to the target user for selection. Because all users are clustered, users with similar preferences can be divided into the same cluster, and by determining the similarity between the target user and users with similar preferences to the target user, users with higher similarity to the target user can be further screened out, which can reduce the CPU's processing pressure on user similarity, that is, it can reduce the waste of CPU processor resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0014] Figure 1 is a flow chart of some embodiments of the virtual item push method according to the present disclosure;

[0015] Figure 2 is a schematic diagram of the structure of some embodiments of the virtual item pushing device according to the present disclosure;

[0016] Figure 3 It is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0017] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0018] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0019] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0020] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0021] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0022] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0023] Figure 1 The process 100 of some embodiments of the virtual item push method according to the present disclosure is shown. The virtual item push method includes the following steps:

[0024] Step 101: in response to receiving a virtual item recommendation request sent by a client, a user rating information set is obtained.

[0025] In some embodiments, in response to receiving a virtual item recommendation request sent by a client, the execution subject (e.g., a computing device) of the virtual item push method may obtain a user rating information set from a user rating information database through a wired connection or a wireless connection. The user rating information database may be a database storing user rating information. The user rating information in the user rating information set may include: user information and a rating information set. The virtual item recommendation request may be a request for recommending a virtual item. The user information may be an identifier that uniquely represents the user. For example, it may be a user ID number. It may also be a user name. The rating information set may be a set of rating information of each virtual item by the user corresponding to the user information. Here, the user corresponding to the user information and the user corresponding to the client are the same user. The virtual item may be a non-physical item that is used for exchange and has value. For example, the virtual item may be an item (financial product) designed and issued by a value conversion institution (financial institution). It should be noted that the wireless connection method may include, but is not limited to, 3G / 4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods that are currently known or will be developed in the future.

[0026] Optionally, after obtaining the user rating information set, the method further includes:

[0027] In the first step, the user rating information set is subjected to data cleaning processing to obtain the user rating information set after data cleaning processing as the cleaned user rating information set. Here, the data cleaning processing may include but is not limited to: missing value processing and outlier processing.

[0028] The second step is to perform data conversion processing on the above-mentioned cleaned user rating information set to obtain the cleaned user rating information set after data conversion processing as the converted user rating information set. Here, the data conversion processing includes but is not limited to: data discretization processing and data standardization processing. Data standardization processing can include min-max standardization and z-score standardization.

[0029] The third step is to determine the converted user rating information set as the user rating information set to update the user rating information set.

[0030] Step 102, filtering the user rating information set to obtain a filtered user rating information set as a filtered user rating information set.

[0031] In some embodiments, the execution entity may filter the user rating information set to obtain the filtered user rating information set as the filtered user rating information set.

[0032] In some optional implementations of some embodiments, the execution subject may filter the user rating information set through the following steps to obtain the filtered user rating information set as the filtered user rating information set:

[0033] The first step is to construct a user rating matrix based on the above user rating information set. In practice, for each user rating information in the above user rating information set, the above execution entity can construct a row vector with the user information included in the user rating information as the first element of the row vector and the rating information set included in the user rating information as the remaining elements. Then, the constructed row vectors can be spliced ​​in order from top to bottom to form a user rating matrix.

[0034] In the second step, the column vector corresponding to the user information in the above user rating matrix is ​​determined as the user information column vector.

[0035] The third step is to determine each column vector in the above user rating matrix except the user information column vector as a rating information column vector set.

[0036] In the fourth step, for each scoring information column vector in the scoring information column vector set, the mean of each scoring information included in the scoring information column vector is determined as the scoring mean information.

[0037] The fifth step is to determine the obtained individual score mean information as a score mean information set.

[0038] Step 6: Update the user rating matrix based on the above-mentioned mean rating information set to obtain an updated user rating matrix.

[0039] The seventh step is to determine the covariance matrix corresponding to the updated user rating matrix according to the updated user rating matrix. In practice, the execution subject can generate the covariance matrix corresponding to the updated user rating matrix by a preset covariance matrix generation function. The preset covariance matrix generation function can be a pre-packaged method that can generate a covariance matrix.

[0040] The eighth step is to determine the eigenvalue set and eigenvector set corresponding to the above covariance matrix. The eigenvalues ​​in the above eigenvalue set correspond to the eigenvectors in the above eigenvector set one by one. In practice, the above execution subject can use a preset matrix characteristic function to determine the eigenvalue set and eigenvector set corresponding to the above covariance matrix. The above preset matrix characteristic function can be a pre-packaged method that can generate eigenvalues ​​and eigenvectors of the covariance matrix.

[0041] The ninth step is to sort the eigenvalues ​​included in the eigenvalue set to obtain an eigenvalue sequence. In practice, the execution subject may sort the eigenvalues ​​included in the eigenvalue set according to the size of the eigenvalue to obtain an eigenvalue sequence.

[0042] The tenth step is to construct an eigenvector matrix according to the above eigenvector set and the above eigenvalue sequence. In practice, first, the above execution subject can sort the eigenvectors corresponding to the eigenvalues ​​in the above eigenvector set according to the order of the eigenvalues ​​in the eigenvalue sequence. Then, the sorted eigenvectors can be combined to generate an eigenvector matrix. Here, the combination processing method can be splicing. The above splicing method can be splicing according to the format of the matrix.

[0043] Step 11, based on the above eigenvector matrix, generate a key eigenvector matrix. In practice, first, the above executor can determine the matrix composed of each eigenvector in the eigenvector matrix that meets the key eigenvector screening condition as the key eigenvector matrix. Among them, the above key eigenvector screening condition can be that the column number of the eigenvector in the eigenvector matrix is ​​less than or equal to the preset column number threshold. The above preset column number threshold can be a pre-set column number threshold. For example, the above preset column number threshold can be 6.

[0044] Step 12: Determine the user key rating information matrix based on the key feature vector matrix and the user rating matrix. In practice, first, the execution subject may determine the matrix composed of the unit feature vectors corresponding to the key feature vectors in the key feature vector matrix as the unit feature vector matrix. Then, the product of the unit feature vector matrix and the transposed matrix of the user rating matrix may be determined as the user key rating information matrix.

[0045] In the thirteenth step, each user key rating information included in the above user key rating information matrix is ​​determined as a filtered user rating information set. Thus, by filtering the user rating information set, redundant features can be removed and the data size can be reduced, thereby reducing the memory space used to store data. At the same time, the principal component information representing the user's interest can be retained, reducing the negative impact caused by matrix sparsity, improving the accuracy of recommendations, and reducing the complexity of subsequent calculations.

[0046] In some optional implementations of some embodiments, according to the above-mentioned score mean information set, the above-mentioned execution subject may update the above-mentioned user score matrix through the following steps to obtain an updated user score matrix:

[0047] In the first step, for each scoring information column vector in the above scoring information column vector set, perform the following steps:

[0048] In the first sub-step, the score mean information corresponding to the score information column vector in the score mean information set is determined as the target score mean information.

[0049] The second sub-step is to perform a difference processing on each score information included in the score information column vector and the target score mean information, and obtain each score information after the difference processing as a difference score information column vector. The difference processing can be understood as a subtraction processing between each score information and the target score mean information. Here, the score information can be a minuend. The target score mean information can be a subtrahend.

[0050] In the second step, the obtained difference score information column vectors are determined as a difference score information column vector set.

[0051] The third step is to construct a difference score matrix based on the difference score information column vector set. In practice, the execution subject can use the difference score information column vector as the column of the matrix, and splice the difference score information columns included in the difference score information column set by column to obtain the spliced ​​difference score information column vector set as the difference score matrix.

[0052] The fourth step is to determine the constructed difference rating matrix as the updated user rating matrix.

[0053] Step 103: cluster the filtered user rating information set to obtain the filtered user rating information set after clustering as the user rating information group set.

[0054] In some embodiments, the execution subject may perform clustering on the above-mentioned screening user rating information set, and obtain the screening user rating information set after clustering as the user rating information group set. In practice, the execution subject may perform clustering on the above-mentioned screening user rating information set using a clustering algorithm, and obtain the screening user rating information set after clustering as the user rating information group set. The clustering algorithm may include but is not limited to: K-means clustering algorithm and DBSCAN clustering algorithm.

[0055] Step 104: in response to determining that the user corresponding to the client is a registered user, determining a user rating information group corresponding to the user in the user rating information group set.

[0056] In some embodiments, in response to determining that the user corresponding to the client is a registered user, the execution subject may determine the user rating information group corresponding to the user in the user rating information group set. In practice, in response to determining that the user corresponding to the client is a registered user, the execution subject may determine the user rating information group to which the user belongs in the user rating information group set as the user rating information group corresponding to the user. Here, it can be determined whether the user corresponding to the client is a registered user by user identification information. User identification information can characterize whether the user is a registered user or an unregistered user. For example, the user identification information may be a user ID. When the first character of the user identification information is 0, it is an unregistered user, and when the first character of the user identification information is 1, it is a registered user. The registered user may be a user who has an access operation record to the virtual item display platform (e.g., a virtual item browsing website) corresponding to the execution subject end. The unregistered user may be a user who has no access operation record to the virtual item display platform corresponding to the execution subject end. Here, the access operation record may include: browsing, liking, and rating virtual items.

[0057] Step 105: Generate a user similarity set based on the user rating information group and the user rating information corresponding to the user.

[0058] In some embodiments, the execution entity may generate a user similarity set based on the user rating information group and the user rating information corresponding to the user.

[0059] In some optional implementations of some embodiments, based on the user rating information group and the user rating information corresponding to the user, the execution subject may generate a user similarity set by following the steps below:

[0060] In the first step, for each user rating information in the user rating information group, the similarity between the user rating information corresponding to the user and the user rating information is determined as the user similarity. In practice, the execution entity may determine the cosine similarity between the user rating information corresponding to the user and the user rating information as the user similarity.

[0061] In the second step, the determined user similarities are determined as a user similarity set.

[0062] Step 106: determine the similarities of each user in the user similarity set that meets the neighbor user screening condition as the neighbor user similarity set.

[0063] In some embodiments, the execution entity may determine the similarities of each user in the user similarity set that meets the neighbor user screening condition as the neighbor user similarity set. The neighbor user screening condition may be that the user similarity is greater than or equal to a preset similarity threshold. The similarity threshold may be a pre-set threshold. For example, the similarity threshold may be 0.6. In practice, the execution entity may determine the similarities of each user in the user similarity set that meets the neighbor user screening condition as the neighbor user similarity set.

[0064] Step 107: Determine a set of neighbor user rating information based on the set of neighbor user similarities.

[0065] In some embodiments, based on the neighbor user similarity set, the execution subject may determine a neighbor user rating information set. In practice, the execution subject may determine each user rating information in the user rating information group corresponding to the neighbor user similarity set as the neighbor user rating information set.

[0066] Step 108: Determine a target user rating information set based on the user rating information corresponding to the user and the neighboring user rating information set.

[0067] In some embodiments, the execution subject may determine the target user rating information set based on the user rating information corresponding to the user and the neighbor user rating information set. The neighbor user rating information in the neighbor user rating information set may include: neighbor user information and virtual item rating information set. The neighbor user rating information may be the same as the user information. The virtual item rating information set may be a subset of the rating information set after screening.

[0068] In some optional implementations of some embodiments, according to the user rating information corresponding to the above-mentioned user and the above-mentioned neighboring user rating information set, the above-mentioned execution subject may determine the target user rating information set through the following steps:

[0069] In the first step, the sum of the similarities of each neighboring user included in the neighboring user similarity set is determined as the total similarity information.

[0070] In the second step, each virtual item rating information set included in the above-mentioned neighbor user rating information set is subjected to item information classification processing to obtain an item rating information group set.

[0071] Step 3: For each item rating information group in the above item rating information group set, perform the following steps:

[0072] In the first sub-step, for each item rating information in the item rating information group, the product of the item rating information and the similarity of the neighboring users corresponding to the item rating information is determined as the rating similarity information.

[0073] In the second sub-step, the sum of the determined individual score similarity information is determined as the total score similarity information.

[0074] The third sub-step is to determine the ratio of the total score similarity information to the total similarity information as the target user score information.

[0075] The fourth step is to determine the determined rating information of each target user as a target user rating information set.

[0076] In some optional implementations of some embodiments, the execution subject may perform item information classification processing on each virtual item rating information set included in the neighboring user rating information set to obtain an item rating information group set by the following steps:

[0077] In the first step, the virtual item scoring information included in the above virtual item scoring information sets are classified according to the item names of the virtual items corresponding to the virtual item scoring information, and the virtual item scoring information corresponding to the same virtual item is obtained as an item scoring information group.

[0078] In the second step, the obtained item rating information groups are determined as an item rating information group set.

[0079] Step 109 , selecting each target user rating information satisfying the to-be-recommended rating condition from the target user rating information set as the to-be-recommended user rating information set.

[0080] In some embodiments, the execution subject may filter out the target user rating information that satisfies the rating condition to be recommended from the target user rating information set as the user rating information set to be recommended. The rating condition to be recommended may be that the rating ranking of the target user rating information in the target user rating information set is less than a preset ranking threshold. The preset ranking threshold may be a preset ranking threshold. For example, the preset ranking threshold may be 10. In practice, first, the execution subject may sort the target user rating information included in the target user rating information set in descending order according to the size of the rating information. Then, the target user rating information that is less than the preset ranking threshold may be determined as the user rating information set to be recommended. For example, the user rating information set to be recommended may include the top 10 user rating information to be recommended. That is, the 10 user rating information to be recommended with the largest rating information corresponding to the user rating information to be recommended.

[0081] Step 110, determining a set of virtual item information to be recommended based on the set of user rating information to be recommended.

[0082] In some embodiments, the execution subject may determine the virtual item information set to be recommended based on the user rating information set to be recommended. In practice, the execution subject may determine the information of each virtual item corresponding to the user rating information set to be recommended as the virtual item information set to be recommended. The virtual item information to be recommended in the virtual item information set to be recommended may be the information of the virtual item to be recommended. For example, the virtual item information to be recommended may include, but is not limited to: the name of the virtual item and the validity period of the virtual item.

[0083] Step 111 : Push the virtual item information set to be recommended to the client to present the virtual items to the user corresponding to the client.

[0084] In some embodiments, the execution entity may push the virtual item information set to be recommended to the client to present the virtual item to the user corresponding to the client. In practice, the execution entity may push the virtual item information set to be recommended to the client to present the virtual item to the user corresponding to the client. The user corresponding to the client may be a user who sends a virtual item recommendation request through the client.

[0085] Furthermore, in the process of adopting technical solutions to solve the technical problems mentioned in the background technology, it is found that users have a high demand for personalized recommendation of virtual items. The conventional technical solution to meet the personalized recommendation needs of users is that for registered users, the user's item preferences can be determined based on the user's historical behavior information. For new users, recommendations are made based on items with higher popularity. However, this method is not conducive to personalized recommendations for new users, and is not conducive to improving the user's click-through rate of recommended virtual items and the conversion rate of items, that is, the utilization rate of relevant information of virtual items is low, resulting in a waste of push resources (such as push traffic). Therefore, considering the need to meet the personalized recommendation needs of users for virtual items, especially the personalized recommendation needs of new users, to improve the user's click-through rate and conversion rate of virtual items, and to reduce the waste of push resources, it is decided to adopt the basic attributes of the target user as the basis, and recommend items liked by users with similar basic attributes to the target user, so as to meet the personalized recommendation needs of users for virtual items, especially the personalized recommendation needs of new users, so as to improve the user's click-through rate of recommended virtual items and the conversion rate of items, that is, to improve the utilization rate of relevant information of virtual items and reduce the waste of push resources (such as push traffic).

[0086] Optionally, the above method further comprises:

[0087] The first step, in response to determining that the user corresponding to the above-mentioned client is an unregistered user, obtains a user basic attribute information group set corresponding to the above-mentioned user rating information group set and user attribute information corresponding to the above-mentioned user. Among them, the above-mentioned user attribute information can be information used to describe user characteristics. The above-mentioned user attribute information can include but is not limited to: age, gender, occupation and education level. The above-mentioned user basic attribute information group set can be a set of user attribute information of each user included in the above-mentioned user rating information group set. In practice, the above-mentioned execution entity can obtain the user basic attribute information group set corresponding to the above-mentioned user rating information group set and the user attribute information corresponding to the above-mentioned user from a basic attribute information database. The above-mentioned basic attribute information database can be a database that stores basic attribute information of users.

[0088] In the second step, for each user basic attribute information group in the above user basic attribute information group set, perform the following steps:

[0089] The first step is to cluster the above user basic attribute information groups to obtain a clustered user basic attribute information group set.

[0090] The second step is to perform the following steps for each clustering user basic attribute information group in the above clustering user basic attribute information group set:

[0091] The first sub-step is to determine the group attribute label information corresponding to the above-mentioned clustered user basic attribute information group according to the above-mentioned clustered user basic attribute information group. In practice, the above-mentioned execution subject can determine the mode of each attribute information included in the clustered user basic attribute information group as the group attribute label information corresponding to the above-mentioned clustered user basic attribute information group. For example, the mode of the gender attribute included in the above-mentioned clustered user basic attribute information group can be female. Then, the gender attribute label information corresponding to the clustered user basic attribute information group can be female.

[0092] The second sub-step is to determine the tag attribute similarity based on the user attribute information and the group attribute tag information. In practice, the execution subject may determine the cosine similarity of the user attribute information and the group attribute tag information as the tag attribute similarity.

[0093] In the third step, the maximum tag attribute similarity among the determined tag attribute similarities is determined as the maximum tag attribute similarity.

[0094] The fourth step is to determine the user basic attribute information group corresponding to the maximum tag attribute similarity as the neighboring user attribute information group.

[0095] The third step is to determine each determined neighbor user attribute information group as a neighbor user attribute information group set.

[0096] The fourth step is to determine the user rating information set corresponding to the neighbor user attribute information set according to the neighbor user attribute information set. In practice, the execution subject may determine each user included in the neighbor user attribute information set as the target neighbor user set. Then, each user rating information corresponding to the target neighbor user set may be determined as the user rating information set corresponding to the neighbor user attribute information set.

[0097] Step 5: Determine the virtual item rating mean information set based on the user rating information set corresponding to the above-mentioned neighboring user attribute information set. In practice, first, the above-mentioned execution subject may determine the mean of each rating information corresponding to the same virtual item in the user rating information set as the virtual item rating mean information. Then, the determined virtual item rating mean information may be determined as the virtual item rating mean information set.

[0098] Step 6: sort the virtual item score average information included in the virtual item score average information set to obtain a virtual item score average information sequence. Here, the sorting method can be descending order.

[0099] The seventh step is to determine the sequence of virtual items to be recommended based on the above-mentioned virtual item score mean information sequence and the preset screening conditions for virtual items to be recommended. In practice, the above-mentioned execution entity can determine the set of virtual item score mean information to be recommended based on the individual virtual item score mean information that meets the preset screening conditions for virtual items to be recommended in the virtual item score mean information sequence. Then, the individual virtual items to be recommended corresponding to the above-mentioned set of virtual item score mean information can be determined as the sequence of virtual items to be recommended. Among them, the above-mentioned preset screening conditions for virtual items to be recommended can be that the position number of the virtual item score mean information in the above-mentioned virtual item score mean information sequence is less than or equal to the preset sequence number threshold. The above-mentioned preset sequence number threshold can be a pre-set sequence number threshold. For example, the above-mentioned preset sequence number threshold can be 10.

[0100] In the eighth step, the sequence of virtual items to be recommended is pushed to the client to present virtual items to the user corresponding to the client.

[0101] The above-mentioned first to eighth steps and their related contents serve as an inventive point of an embodiment of the present disclosure, which solves the technical problem that "when recommending items to users, recommendations are made based on items with higher popularity for unregistered users, which cannot better meet the personalized recommendation needs of unregistered users, is not conducive to improving the user's click rate on the recommended virtual items and the conversion rate of items, that is, the utilization rate of relevant information of virtual items is low, thereby causing a waste of push resources (such as push traffic)." The factors that lead to the following when recommending items to users, for new users, recommendations are made based on items with higher popularity, which cannot meet the personalized recommendation needs of unregistered users well, which is not conducive to improving the click rate of users on the recommended virtual items and the conversion rate of items, that is, the utilization rate of the relevant information of virtual items is low, thereby causing the waste of push resources (such as push traffic): When recommending items to users, for unregistered users, recommendations are made based on items with higher popularity, which cannot meet the personalized recommendation needs of unregistered users well, which is not conducive to improving the click rate of users on the recommended virtual items and the conversion rate of items, that is, the utilization rate of the relevant information of virtual items is low, thereby causing the waste of push resources (such as push traffic). If the above factors are solved, it is possible to meet the personalized recommendation needs of users for virtual items, especially the personalized recommendation needs of unregistered users, improve the click rate and conversion rate of users for virtual items, and reduce the waste of push resources. In order to achieve this effect, the present disclosure clusters all users' historical behavior information to obtain large clusters, and then distinguishes users in each large cluster into small clusters based on basic attribute labels, and determines the basic attribute labels of the small clusters at the same time. By determining the similarity between the basic attribute labels of the target unregistered user and each small cluster in each large cluster, a set of similar users with high similarity to the target unregistered user in each large cluster is determined. The recommended items for the target unregistered user are determined by the item rating preferences of the similar user set. This can better meet the personalized recommendations of unregistered users, improve the click-through rate and conversion rate of users for virtual items, and reduce the waste of push resources.

[0102] The above-mentioned embodiments of the present disclosure have the following beneficial effects: through the virtual item push method of some embodiments of the present disclosure, the waste of CPU processor resources can be reduced. Specifically, the reason for the waste of CPU processor resources is that when making virtual item recommendations, the number of users or the number of virtual items is large, and the user similarity between the target user and all users is determined, resulting in a large number of CPU processor resources occupied by the CPU processor for processing user similarity, and a long processing time, thereby resulting in a waste of CPU processor resources. Based on this, the virtual item push method of some embodiments of the present disclosure, first, in response to receiving a virtual item recommendation request sent by a client, obtains a user rating information set. Thus, a user rating information set representing the evaluation of all virtual items by all users can be obtained. Then, the above user rating information set is filtered and processed to obtain the filtered user rating information set as the filtered user rating information set. Thus, the main component virtual item evaluation information representing the user's interest can be obtained. Thus, redundant features can be removed, the data scale can be reduced, and the memory space for storing data can be reduced. Afterwards, the above filtered user rating information set is clustered to obtain the filtered user rating information set after clustering as the user rating information group set. Thus, the user rating information group set can be obtained. Thus, users with similar preferences can be divided into the same cluster. Subsequently, in response to determining that the user corresponding to the above-mentioned client is a registered user, the user rating information group corresponding to the above-mentioned user in the above-mentioned user rating information group set is determined. Thus, the user rating information group with similar preferences to the target user can be determined. Subsequently, a user similarity set is generated based on the above-mentioned user rating information group and the user rating information corresponding to the above-mentioned user. Thus, the similarity of each user with similar preferences to the target user can be determined. Secondly, the similarity of each user in the above-mentioned user similarity set that meets the neighbor user screening condition is determined as the neighbor user similarity set. Thus, each neighbor user similarity set with a high similarity to the target user can be obtained. Then, based on the above-mentioned neighbor user similarity set, a neighbor user rating information set is determined. Thus, a neighbor user rating information set corresponding to each user with a high similarity can be obtained. Afterwards, the target user rating information set is determined based on the user rating information corresponding to the above-mentioned user and the above-mentioned neighbor user rating information set. Thus, based on the rating information of similar users on virtual items, the rating information of the target user on each virtual item can be determined. Subsequently, each target user rating information satisfying the to-be-recommended rating condition is selected from the above target user rating information set as the to-be-recommended user rating information set. Thus, a to-be-recommended user rating information set that represents the target user's high preference can be obtained. Subsequently, a to-be-recommended virtual item information set is determined based on the above to-be-recommended user rating information set. Thus, a to-be-recommended virtual item set can be obtained, thereby determining each virtual item that the user has a high preference for.Finally, the above virtual item information set to be recommended is pushed to the above client to present virtual items to the user corresponding to the above client. In this way, each virtual item that the target user prefers can be sent to the target user for selection by the target user. Because all users are clustered, each user with similar preferences can be divided into the same cluster, and by determining the similarity between the target user and each user with similar preferences to the target user, each user with a higher similarity to the target user can be further screened out, which can reduce the CPU's processing pressure on user similarity, that is, it can reduce the waste of CPU processor resources.

[0103] Further references Figure 2 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a virtual item push device. These device embodiments are similar to Figure 1 Corresponding to the method embodiments shown, the device can be specifically applied to various electronic devices.

[0104] like Figure 2As shown, the virtual item pushing device 200 of some embodiments includes: an acquisition unit 201, a first screening unit 202, a clustering processing unit 203, a first determination unit 204, a generation unit 205, a second determination unit 206, a third determination unit 207, a fourth determination unit 208, a second screening unit 209, a fifth determination unit 210 and a pushing unit 211. The acquisition unit 201 is configured to acquire a user rating information set in response to receiving a virtual item recommendation request sent by a client; the first screening unit 202 is configured to screen the user rating information set to obtain the screened user rating information set as a screened user rating information set; the clustering processing unit 203 is configured to cluster the screened user rating information set to obtain the screened user rating information set as a user rating information group set; the first determination unit 204 is configured to determine the user rating information group corresponding to the user in the user rating information group set in response to determining that the user corresponding to the client is a registered user; the generation unit 205 is configured to generate a user similarity set based on the user rating information group and the user rating information corresponding to the user; the second determination unit 206 is configured to cluster the above-mentioned user rating information group into a user rating information group. The similarities of each user in the user similarity set that meets the neighbor user screening condition are determined as the neighbor user similarity set; the third determination unit 207 is configured to determine the neighbor user rating information set based on the above neighbor user similarity set; the fourth determination unit 208 is configured to determine the target user rating information set based on the user rating information corresponding to the above user and the above neighbor user rating information set; the second screening unit 209 is configured to filter out each target user rating information that meets the to-be-recommended rating condition from the above target user rating information set as the to-be-recommended user rating information set; the fifth determination unit 210 is configured to determine the to-be-recommended virtual item information set based on the above to-be-recommended user rating information set; the push unit 211 is configured to push the above to-be-recommended virtual item information set to the above client to present the virtual item to the user corresponding to the above client.

[0105] It is understood that the units described in the device 200 are similar to those described in the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the device 200 and the units included therein, and will not be described in detail here.

[0106] Reference below Figure 3 , which shows a schematic structural diagram of an electronic device (such as a computing device) 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0107] like Figure 3 As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0108] Typically, the following devices may be connected to the I / O interface 305: input devices 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 308 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 309. The communication devices 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0109] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from the network through the communication device 309, or installed from the storage device 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.

[0110] It should be noted that the computer-readable medium recorded in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-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 computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer 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. In some embodiments of the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In some embodiments of the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0111] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (Hyper Text Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0112] The above-mentioned computer-readable medium may be included in the above-mentioned electronic device; or it may exist independently without being assembled into the electronic device. The above-mentioned computer-readable medium carries one or more programs. When the above-mentioned one or more programs are executed by the electronic device, the electronic device: obtains a user rating information set in response to receiving a virtual item recommendation request sent by a client; performs filtering processing on the above-mentioned user rating information set to obtain a filtered user rating information set as a filtered user rating information set; performs clustering processing on the above-mentioned filtered user rating information set to obtain a clustered filtered user rating information set as a user rating information group set; in response to determining that the user corresponding to the client is a registered user, determines the user rating information group corresponding to the user in the user rating information group set; and selects a user rating information group from the user rating information group set based on the user rating information group and the user corresponding to the user. The system comprises the following steps: according to the first step, the first step is to use the user rating information of the target user to generate a user similarity set; the second step is to determine the neighboring user rating information set according to the neighboring user similarity set; the third step is to determine the neighboring user rating information set according to the neighboring user similarity set; the fourth step is to determine the neighboring user rating information set according to the neighboring user rating information set; the fourth step is to determine the target user rating information set according to the neighboring user rating information set; the fifth step is to determine the virtual item information set to be recommended according to the user rating information set to be recommended; the fifth step is to push the virtual item information set to the client to present the virtual item to the user corresponding to the client.

[0113] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0114] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0115] The units described in some embodiments of the present disclosure may be implemented by software or by hardware. The described units may also be set in a processor, for example, may be described as: a processor includes an acquisition unit, a first screening unit, a clustering processing unit, a first determination unit, a generation unit, a second determination unit, a third determination unit, a fourth determination unit, a second screening unit, a fifth determination unit and a push unit. Among them, the names of these units do not constitute a limitation on the units themselves under certain circumstances. For example, the push unit may also be described as "a unit that pushes the set of virtual item information to be recommended to the client to present virtual items to the user corresponding to the client."

[0116] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0117] The above descriptions are only some preferred embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) and the technical solutions formed.

Claims

1. A method for pushing virtual items, comprising: In response to receiving a virtual item recommendation request sent by a client, obtaining a user rating information set; Screening the user rating information set to obtain a screened user rating information set as a screened user rating information set; Performing clustering processing on the screening user rating information set to obtain the screening user rating information set after clustering processing as the user rating information group set; In response to determining that the user corresponding to the client is a registered user, determining a user rating information group corresponding to the user in the user rating information group set; Generate a user similarity set according to the user rating information group and the user rating information corresponding to the user; Determine the similarities of each user in the user similarity set that meets the neighbor user screening condition as the neighbor user similarity set; Determine a neighbor user rating information set based on the neighbor user similarity set; Determine a target user rating information set according to the user rating information corresponding to the user and the neighboring user rating information set; Filtering out each target user rating information satisfying the to-be-recommended rating condition from the target user rating information set as the to-be-recommended user rating information set; Determining a set of virtual item information to be recommended based on the set of user rating information to be recommended; Pushing the virtual item information set to be recommended to the client to present virtual items to the user corresponding to the client; In response to determining that the user corresponding to the client is an unregistered user, obtaining a user basic attribute information group set corresponding to the user rating information group set and user attribute information corresponding to the user; For each user basic attribute information group in the user basic attribute information group set, perform the following steps: Performing clustering processing on the user basic attribute information groups to obtain a clustered user basic attribute information group set; For each clustering user basic attribute information group in the clustering user basic attribute information group set, the following steps are performed: According to the clustered user basic attribute information group, determining group attribute label information corresponding to the clustered user basic attribute information group; Determining tag attribute similarity according to the user attribute information and the group attribute tag information; Determine the maximum tag attribute similarity among the determined tag attribute similarities as the maximum tag attribute similarity; Determine the user basic attribute information group corresponding to the maximum tag attribute similarity as the neighbor user attribute information group; Determining each determined neighbor user attribute information group as a neighbor user attribute information group set; Determine, based on the neighbor user attribute information group set, a user rating information set corresponding to the neighbor user attribute information group set; Determining a virtual item rating mean information set according to a user rating information set corresponding to the neighboring user attribute information group set; Sorting each piece of virtual item score average information included in the virtual item score average information set to obtain a virtual item score average information sequence; Determining a sequence of virtual items to be recommended according to the virtual item rating mean information sequence and preset screening conditions for virtual items to be recommended; The sequence of virtual items to be recommended is pushed to the client to present virtual items to the user corresponding to the client.

2. The method according to claim 1, wherein: The user rating information in the user rating information set includes: user information and a rating information set; and after obtaining the user rating information set, the method further includes: Performing data cleaning on the user rating information set to obtain a user rating information set after data cleaning as a cleaned user rating information set; Performing data conversion processing on the cleaned user rating information set to obtain a cleaned user rating information set after data conversion processing as a converted user rating information set; The converted user rating information set is determined as the user rating information set to update the user rating information set.

3. The method according to claim 1, wherein: The generating a user similarity set according to the user rating information group and the user rating information corresponding to the user includes: For each user rating information in the user rating information group, determining the similarity between the user rating information corresponding to the user and the user rating information as the user similarity; The determined user similarities are determined as a user similarity set.

4. The method according to claim 1, wherein: The neighbor user rating information in the neighbor user rating information set includes: neighbor user information and virtual item rating information set; and determining the target user rating information set according to the user rating information corresponding to the user and the neighbor user rating information set includes: Determine the sum of similarities of each neighboring user included in the neighboring user similarity set as total similarity information; Performing item information classification processing on each virtual item rating information set included in the neighboring user rating information set to obtain an item rating information group set; For each item rating information group in the item rating information group set, perform the following steps: For each item rating information in the item rating information group, determining the product of the item rating information and the similarity of the neighboring user corresponding to the item rating information as the rating similarity information; Determine the sum of each determined score similarity information as total score similarity information; Determine the ratio of the total score similarity information to the total similarity information as the target user score information; The determined individual target user rating information is determined as a target user rating information set.

5. The method according to claim 4, wherein: The performing item information classification processing on each virtual item rating information set included in the neighboring user rating information set to obtain an item rating information group set includes: According to the item name of the virtual item corresponding to each virtual item rating information, item information classification processing is performed on each virtual item rating information set, so as to obtain each virtual item rating information corresponding to the same virtual item as an item rating information group; The obtained item rating information groups are determined as an item rating information group set.

6. The method according to claim 1, wherein: The filtering process of the user rating information set to obtain the filtered user rating information set as the filtered user rating information set includes: Constructing a user rating matrix according to the user rating information set; Determine the column vector corresponding to the user information in the user rating matrix as the user information column vector; Determine each column vector in the user rating matrix except the user information column vector as a rating information column vector set; For each scoring information column vector in the scoring information column vector set, determining the mean of each scoring information included in the scoring information column vector as scoring mean information; Determine the obtained individual score mean information as a score mean information set; According to the score mean information set, the user score matrix is ​​updated to obtain an updated user score matrix; Determine, according to the updated user rating matrix, a covariance matrix corresponding to the updated user rating matrix; Determine an eigenvalue set and an eigenvector set corresponding to the covariance matrix, wherein the eigenvalues ​​in the eigenvalue set correspond to the eigenvectors in the eigenvector set; Sorting each eigenvalue included in the eigenvalue set to obtain an eigenvalue sequence; Constructing an eigenvector matrix according to the eigenvector set and the eigenvalue sequence; Generate a key eigenvector matrix according to the eigenvector matrix; Determining a user key scoring information matrix according to the key feature vector matrix and the user scoring matrix; Each user key rating information included in the user key rating information matrix is ​​determined as a screening user rating information set.

7. The method according to claim 6, wherein: The updating of the user rating matrix according to the rating mean information set to obtain an updated user rating matrix includes: For each scoring information column vector in the scoring information column vector set, the following steps are performed: Determining the score mean information corresponding to the score information column vector in the score mean information set as the target score mean information; Performing a difference process on each score information included in the score information column vector and the target score mean information, and obtaining each score information after the difference process as a difference score information column vector; Determine each obtained difference score information column vector as a difference score information column vector set; Constructing a difference score matrix according to the difference score information column vector set; The constructed difference rating matrix is ​​determined as the updated user rating matrix.

8. A virtual item pushing device, comprising: A first acquisition unit is configured to acquire a user rating information set in response to receiving a virtual item recommendation request sent by a client; A first screening unit is configured to screen the user rating information set to obtain a screened user rating information set as a screened user rating information set; A cluster processing unit, configured to perform cluster processing on the screening user rating information set to obtain the screening user rating information set after cluster processing as the user rating information group set; A first determining unit is configured to, in response to determining that the user corresponding to the client is a registered user, determine a user rating information group corresponding to the user in the user rating information group set; A generating unit, configured to generate a user similarity set according to the user rating information group and the user rating information corresponding to the user; A second determining unit is configured to determine the similarities of each user in the user similarity set that meets the neighbor user screening condition as a neighbor user similarity set; A third determining unit is configured to determine a neighbor user rating information set according to the neighbor user similarity set; A fourth determining unit is configured to determine a target user rating information set according to the user rating information corresponding to the user and the neighboring user rating information set; A second screening unit is configured to screen out each target user rating information satisfying the to-be-recommended rating condition from the target user rating information set as the to-be-recommended user rating information set; A fifth determining unit is configured to determine a set of virtual item information to be recommended according to the set of user rating information to be recommended; A first pushing unit is configured to push the virtual item information set to be recommended to the client to present virtual items to a user corresponding to the client; A second acquisition unit is configured to, in response to determining that the user corresponding to the client is an unregistered user, acquire a user basic attribute information group set corresponding to the user rating information group set and user attribute information corresponding to the user; The execution unit is configured to execute the following steps for each user basic attribute information group in the user basic attribute information group set: Performing clustering processing on the user basic attribute information groups to obtain a clustered user basic attribute information group set; For each clustering user basic attribute information group in the clustering user basic attribute information group set, the following steps are performed: According to the clustered user basic attribute information group, determining group attribute label information corresponding to the clustered user basic attribute information group; Determining tag attribute similarity according to the user attribute information and the group attribute tag information; Determine the maximum tag attribute similarity among the determined tag attribute similarities as the maximum tag attribute similarity; Determine the user basic attribute information group corresponding to the maximum tag attribute similarity as the neighbor user attribute information group; a sixth determining unit, configured to determine the determined neighbor user attribute information groups as a neighbor user attribute information group set; A seventh determining unit is configured to determine, based on the neighboring user attribute information set, a user rating information set corresponding to the neighboring user attribute information set; An eighth determining unit is configured to determine a virtual item rating mean value information set according to a user rating information set corresponding to the neighboring user attribute information group set; a sorting unit configured to sort the individual virtual item score average information included in the virtual item score average information set to obtain a virtual item score average information sequence; a ninth determining unit, configured to determine a sequence of virtual items to be recommended according to the virtual item rating mean information sequence and preset screening conditions for virtual items to be recommended; The second pushing unit is configured to push the sequence of virtual items to be recommended to the client, so as to present virtual items to a user corresponding to the client.

9. An electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A computer readable medium having a computer program stored thereon, wherein: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images