Operational data processing method, device, computer equipment and storage medium

Abstract

The application relates to an operation data processing method and device, a computer device and a storage medium. The method comprises the following steps: obtaining object identification object operation data corresponding to candidate resource labels, and obtaining resource operation data obtained by performing operations on each target resource label; the candidate resource labels comprise target resource labels; determining the object operation frequency of the object identification based on the object operation data, and determining the label operation frequency of each target resource label being operated based on the resource operation data; determining the target operation frequency of each target resource label of the target resource of the object identification according to target operation data obtained by performing operations on each target resource label of the target resource of the object identification; and determining the attention degree of the object identification to the target resource based on the object operation frequency, the label operation frequencies and the target operation frequencies. The real-time processing efficiency of operation data can be improved by using the method.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to an operation data processing method and device, computer equipment and storage medium. BACKGROUND

[0002] With the development of Internet business, information recommendation technology appears, through which content interested by a user can be recommended to the user. In the information recommendation technology, the interest degree of a user to a certain information or a certain item is often predicted.

[0003] In a traditional way, the interest degree of a user to a certain information or a certain item is usually calculated by using the attribute features of the user itself, the related features of the historical operation data of the user to the information or the item, and the features of the information or the item itself. However, in the case of a large quantity of users and items, the above processing method cannot realize real-time processing, resulting in low processing efficiency. SUMMARY

[0004] Therefore, it is necessary to provide an operation data processing method, device, computer equipment and storage medium capable of improving real-time processing efficiency in view of the above technical problems.

[0005] An operation data processing method, the method comprising:

[0006] an obtaining module configured to obtain object operation data of an object identifier to candidate resource tags, and obtain resource operation data obtained by operation on each target resource tag; the candidate resource tags comprising the target resource tags;

[0007] a determining module configured to determine an object operation frequency of the object identifier based on the object operation data, and determine a tag operation frequency of each target resource tag being operated based on the resource operation data;

[0008] a determining module configured to determine a target operation frequency of each target resource tag of the target resource operated by the object identifier according to target operation data obtained by operation of the object identifier on each target resource tag of the target resource;

[0009] a determining module configured to determine an attention degree of the object identifier to the target resource based on the object operation frequency, the tag operation frequency and the target operation frequency.

[0010] An operation data processing device, the device comprising:

[0011] an obtaining module configured to obtain object operation data of an object identifier to candidate resource tags, and obtain resource operation data obtained by operation on each target resource tag; the candidate resource tags comprising the target resource tags;

[0012] a frequency determination module, configured to determine an object operation frequency of the object identifier based on the object operation data, and determine a tag operation frequency of each target resource tag being operated based on the resource operation data;

[0013] an operation module, configured to determine a target operation frequency of the object identifier on each target resource tag of the target resource according to target operation data obtained by operating the object identifier on each target resource tag of the target resource;

[0014] The operation data processing module is used to determine the attention degree of the object identifier to the target resource based on the object operation frequency, the tag operation frequency and the target operation frequency.

[0015] In one embodiment, the acquisition module is further configured to acquire object operation data corresponding to candidate resource tags of different versions of the object identifier, and to acquire resource operation data obtained by performing operations on each target resource tag of different versions; different versions correspond to different timestamps;

[0016] The frequency determination module is further configured to determine the object operation frequency corresponding to the object identifier based on different versions of object operation data; and determine the tag operation frequency of each target resource tag being operated based on different versions of resource operation data.

[0017] In one embodiment, the apparatus further comprises a weight determination module, the weight determination module being configured to determine an object weight of the object identifier according to the object operation frequency, and to determine a tag weight corresponding to each target resource tag according to each tag operation frequency;

[0018] The operation data processing module is further configured to determine the attention degree of the object identifier to the target resource based on the object weight, the weights of each tag, and the frequency of each target operation.

[0019] In one embodiment, the operation data processing module is also used to determine the operation evaluation value of the object identifier for each target resource tag based on the object weight, the weight of each tag and the frequency of each target operation; and determine the attention of the object identifier to the target resource based on the time decay factor and each operation evaluation value.

[0020] In one embodiment, the operation data processing module is also used to determine the candidate operation frequency corresponding to each target resource tag of each candidate resource for the object identifier, and the candidate resources include the target resource; determine the sub-evaluation value of each target resource tag for the corresponding candidate resource based on the object weight, the weight of each tag, and the candidate operation frequency; determine the operation evaluation value of the object identifier for each target resource tag based on each sub-evaluation value; filter out the target operation evaluation value that meets the evaluation conditions from each operation evaluation value, and determine the attention of the object identifier to the target resource based on the time attenuation factor and the target operation evaluation value.

[0021] In one embodiment, the apparatus further includes an evaluation module; the evaluation module is configured to determine, from the object operation data acquired within a preset time period, the number of valid operations of the object identifier for each target resource tag; and determine a stable evaluation value of the object identifier operating on each target resource tag based on each valid operation number and the preset time period;

[0022] The operation data processing module is further configured to determine the attention degree of the object identifier to the target resource based on the object operation frequency, the tag operation frequencies, the target operation frequencies and the stability evaluation values.

[0023] In one embodiment, the evaluation module is further used to divide the preset duration into multiple time periods through a preset time window, and determine the number of valid operations of the object identifier in the object operation data for each target resource tag in each time period; from the valid operation numbers corresponding to each time period, screen out the target operation number and the corresponding target time period that meet the valid operation conditions; and determine the stable evaluation value of the object identifier operating on each target resource tag based on the target operation number and the target time period.

[0024] In one embodiment, the evaluation module is further configured to determine, from object operation data of different promotion channels acquired within a preset time period, the number of valid operations of the object identifier on each target resource tag under each promotion channel; and for each promotion channel, determine, based on the corresponding number of valid operations and the preset time period, a stable evaluation value of the object identifier's operations on each target resource tag under the corresponding promotion channel.

[0025] The operation data processing module is further used to determine the attention of the object identifier to the target resource based on the object operation frequency, each tag operation frequency, each target operation frequency and each stability evaluation value corresponding to each promotion channel.

[0026] In one embodiment, the target resource is promotion information; the device also includes a promotion module; the promotion module is used to screen out target object identifiers whose attention levels meet push conditions based on the attention levels of each object identifier to the promotion information; and push associated information related to the promotion information to the target object identifiers.

[0027] In one embodiment, the target resource belongs to a member type object; the device also includes a distribution module; the distribution module is used to determine the membership activation probability of the object identifier for the member type object based on the attention of the object identifier to the member type object; according to the member activation probability and the consumption resource value corresponding to the member type object, the target virtual gift resource is selected from the candidate virtual gift resources and distributed to the object identifier.

[0028] In one embodiment, the issuance module is also used to determine the expected benefit value corresponding to each candidate virtual gift resource based on the member activation probability, the consumption resource value corresponding to the member type object, and the resource value contained in each candidate virtual gift resource; from the candidate virtual gift resources, select the target virtual gift resource that makes the expected benefit value meet the issuance conditions, and issue the target virtual gift resource according to the object identifier.

[0029] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0030] Obtaining object operation data of the object identifier for the candidate resource tag, and obtaining resource operation data obtained by the operation on each target resource tag; the candidate resource tags include the target resource tag;

[0031] Determining an object operation frequency of the object identifier based on the object operation data, and determining a tag operation frequency of each target resource tag being operated based on the resource operation data;

[0032] determining a target operation frequency of the object identifier on each target resource tag of the target resource according to target operation data obtained by operating the object identifier on each target resource tag of the target resource;

[0033] The attention degree of the object identifier to the target resource is determined based on the object operation frequency, the tag operation frequencies, and the target operation frequencies.

[0034] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0035] Obtaining object operation data of the object identifier for the candidate resource tag, and obtaining resource operation data obtained by the operation on each target resource tag; the candidate resource tags include the target resource tag;

[0036] Determining an object operation frequency of the object identifier based on the object operation data, and determining a tag operation frequency of each target resource tag being operated based on the resource operation data;

[0037] determining a target operation frequency of the object identifier on each target resource tag of the target resource according to target operation data obtained by operating the object identifier on each target resource tag of the target resource;

[0038] The attention degree of the object identifier to the target resource is determined based on the object operation frequency, the tag operation frequencies, and the target operation frequencies.

[0039] A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program performs the following steps:

[0040] Obtaining object operation data of the object identifier for the candidate resource tag, and obtaining resource operation data obtained by the operation on each target resource tag; the candidate resource tags include the target resource tag;

[0041] Determining an object operation frequency of the object identifier based on the object operation data, and determining a tag operation frequency of each target resource tag being operated based on the resource operation data;

[0042] determining a target operation frequency of the object identifier on each target resource tag of the target resource according to target operation data obtained by operating the object identifier on each target resource tag of the target resource;

[0043] The attention degree of the object identifier to the target resource is determined based on the object operation frequency, the tag operation frequencies, and the target operation frequencies.

[0044] The above-mentioned operation data processing method, apparatus, computer equipment, computer-readable storage medium and computer program product can accurately calculate the frequency of user operations based on the user's resource operation data on the candidate resource tags by obtaining the object operation data of the object identifier for the candidate resource tags and determining the object operation frequency of the object identifier based on the object operation data. The resource operation data obtained by the operation on each target resource tag is obtained, and the tag operation frequency of each target resource tag being operated is determined based on the resource operation data. The frequency of the target resource tag being operated can be accurately calculated based on the resource operation data generated on the target resource tag, effectively reflecting the frequency of the target resource tag being operated. The target operation frequency of the object identifier on each target resource tag of the target resource is determined based on the target operation data obtained by the object identifier operating on each target resource tag of the target resource, which can effectively reflect the frequency of the object identifier operating on each tag of the target resource. The object identifier's interest in the target resource is determined based on the object operation frequency, the frequency of each tag operation, and the frequency of each target operation. This accurately calculates the object identifier's interest in the target resource based on the frequency of user operations on candidate resource tags, the frequency of target resource tags being operated on, and the frequency of each target resource tag being operated on by the object identifier. Furthermore, by calculating interest based on statistical data such as user operation frequency and target resource tag operation frequency, the computational complexity is small and the processing speed is fast, effectively improving the real-time processing efficiency of users' interest in target resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A diagram of an application environment for operating a data processing method in one embodiment;

[0046] Figure 2 A schematic flow chart of an operation data processing method in one embodiment;

[0047] Figure 3 A flowchart illustrating steps for obtaining object operation data and resource operation data in one embodiment;

[0048] Figure 4 A flowchart of another embodiment of determining an operation evaluation value of an object identifier for each target resource tag based on the object weight, the weight of each tag, and the frequency of each target operation;

[0049] Figure 5 A schematic diagram of a process for determining a stable evaluation value of an object identifier operating on each target resource tag in one embodiment;

[0050] Figure 6A flowchart of determining the number of valid operations of an object identifier for each target resource tag in object operation data acquired within a preset time period in one embodiment;

[0051] Figure 7 A flowchart of determining the number of valid operations of an object identifier for each target resource tag in object operation data acquired within a preset time period in one embodiment;

[0052] Figure 8 A flowchart of an operating data processing method according to another embodiment;

[0053] Figure 9 is a structural block diagram of an operating data processing device in one embodiment;

[0054] Figure 10 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0056] The present application relates to the field of artificial intelligence (AI) technology, in which artificial intelligence is the theory, method, technology and application system for simulating, extending and expanding human intelligence using digital computers or machines controlled by digital computers, perceiving the environment, acquiring knowledge and using knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology in computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a similar way to human intelligence. Artificial intelligence is also the study of the design principles and implementation methods of various intelligent machines, so that the machines have the functions of perception, reasoning and decision-making. The solutions provided in the embodiments of the present application relate to an operation data processing method of artificial intelligence, which is specifically described by the following embodiments.

[0057] The operation data processing method provided in this application can be applied to Figure 1 The operating data processing system shown is as follows. Figure 1As shown, the operation data processing system includes a terminal 110 and a server 120. In one embodiment, each of the terminal 110 and the server 120 can independently execute the operation data processing method provided in the embodiments of the present application. The terminal 110 and the server 120 can also be used in conjunction to execute the operation data processing method provided in the embodiments of the present application. When the terminal 110 and the server 120 are used in conjunction to execute the operation data processing method provided in the embodiments of the present application, the terminal 110 obtains object operation data for the object identifier for candidate resource tags and obtains resource operation data resulting from operations performed on each target resource tag, wherein the candidate resource tags include the target resource tag. The terminal 110 obtains target operation data resulting from operations performed by the object identifier on each target resource tag of the target resource. The terminal 110 sends the object operation data, resource operation data, and target operation data to the server 120. The server 120 determines the object operation frequency of the object identifier based on the object operation data and determines the tag operation frequency of each target resource tag based on the resource operation data. The server 120 determines the target operation frequency of the object identifier for each target resource tag of the target resource based on the target operation data obtained from operations performed by the object identifier on each target resource tag of the target resource. The server 120 determines the attention degree of the object identifier to the target resource based on the object operation frequency, the tag operation frequency, and the target operation frequency, and returns the attention degree result of the target resource to the terminal 110 .

[0058] The server 120 may be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server providing cloud computing services. The terminal 110 may be a smartphone, tablet computer, laptop computer, desktop computer, smart speaker, smart watch, in-vehicle terminal, smart TV, etc., but is not limited thereto. The terminal 110 and the server 120 may be connected directly or indirectly via wired or wireless communication, which is not limited in this application.

[0059] In one embodiment, multiple servers may form a blockchain, with the servers acting as nodes on the blockchain.

[0060] In one embodiment, data related to the operation data processing method can be stored on the blockchain, such as object identification, candidate resource tags, object operation data, target resource tags, resource operation data, object operation frequency, tag operation frequency, target operation data, target operation frequency and attention, etc., but not limited to these.

[0061] In one embodiment, Figure 2 As shown, a method for processing operation data is provided, which is applied to Figure 1 Computer equipment in (the computer equipment can be Figure 1The method is described below with reference to a terminal or a server as an example, including the following steps:

[0062] In step S202, object identification object operation data for candidate resource tags is obtained, and resource operation data obtained by performing an operation on each target resource tag is obtained; the candidate resource tags include the target resource tags.

[0063] The candidate resource tags refer to categories or items corresponding to candidate resources, such as categories or items on an application or a webpage. The candidate resource tags can be, for example, subscriptions, selections, discoveries, TV series, variety shows, animations, movies, documentaries, live broadcasts, entertainment, games, and life categories in a video application. The candidate resource tags can also be, for example, popular, local, rankings, movies, social, funny, emotional, TV series, variety shows, sports, and campus categories in a sharing application, without limitation.

[0064] The candidate resources can be at least one of various types of information and various types of items. The various types of information include at least one of text, emoticons, pictures, audio, video, files, or links, but are not limited thereto. The various types of items can include physical items and virtual items. The physical items include various types of physical products, such as various electronic products such as mobile phones, computers, notebooks, and watches, and clothing and footwear products, without limitation.

[0065] The virtual items include, but are not limited to, insurance products, financial products, virtual gift resources, virtual scenes, virtual characters, and virtual props. The virtual scenes can be, for example, game scenes in a game device or virtual reality simulation scenes. The virtual characters can be, for example, various types of characters in a game. The virtual props can be, for example, various types of props in a game.

[0066] The object operation data refers to operation data obtained by performing a preset operation on all candidate resource tags by an object identification, or operation data obtained by performing a preset operation on a single candidate resource tag by an object identification. The object identification can be a user identification. For example, data obtained by performing a preset operation on five candidate resource tags by an object identification A. The preset operation includes, but is not limited to, at least one of a touch operation, a voice operation, an operation by an input device such as a mouse, or a gesture operation, such as any one of a click operation, a double-click operation, a long-press operation, a left-swipe operation, or a right-swipe operation, without limitation.

[0067] The target resource tags refer to categories or items corresponding to target resources. The target resources refer to resources for which a user's attention degree needs to be determined. The resource operation data refers to data obtained by performing an operation on a target resource tag, such as data obtained by performing an operation on the target resource tag by five object identifications.

[0068] Specifically, the computer device determines the object identifier for which attention needs to be calculated and obtains object operation data for the object identifier with respect to the candidate resource tags. Furthermore, the computer device obtains object operation data for each candidate resource tag. For example, if the object identifier performs a click operation on candidate resource tag 1, the object operation data is a click operation; if the object identifier does not perform a preset operation on candidate resource tag 2, the object operation data may be a no operation.

[0069] In one embodiment, if the object identifier performs a preset operation on the candidate resource tag, the object operation data record is 1; if the object identifier does not perform a preset operation on the candidate resource tag, the object operation data record is 0.

[0070] The computer device determines the target resource tag corresponding to the target resource for which attention needs to be calculated. The same target resource may correspond to multiple target resource tags. The candidate resource tags include the target resource tag. For example, 3 target resource tags are included in 10 candidate resource tags.

[0071] For each target resource tag, the computer device obtains resource operation data generated by a preset operation occurring on the target resource tag, and the resource operation data may include operations performed on the target resource tag by multiple object identifiers, where multiple refers to at least two.

[0072] Step S204 : determining the object operation frequency of the object identifier based on the object operation data, and determining the tag operation frequency of each target resource tag being operated based on the resource operation data.

[0073] The object operation frequency refers to the number of times an object identifier performs operations on a candidate resource tag within a preset time and range. The tag operation frequency refers to the number of times a preset operation is performed on a target resource tag within a preset time and range, such as the number of clicks or views within 15 days. It is understood that the preset operations performed on the target resource tag and the preset operations performed on the candidate tag by the user tag can be exactly the same or partially the same.

[0074] Specifically, after obtaining the object operation data of the object identifier for each candidate resource tag, the computer device calculates the object operation frequency corresponding to the object identifier based on the object operation data of each candidate resource tag.

[0075] For each target resource tag, the computer device obtains resource operation data corresponding to the target resource tag, and calculates the tag operation frequency of the target resource tag according to the resource operation data.

[0076] Step S206 : determining the target operation frequency of the object identifier on each target resource tag of the target resource according to the target operation data obtained by the object identifier operating on each target resource tag of the target resource.

[0077] Target operation data refers to the data obtained by an object identifier performing a specific operation on a target resource's target resource tag. For example, object identifier A clicks on target resource tags 1, 2, and 3 of target resource i, respectively, resulting in data. Each target resource tag corresponds to a piece of target operation data.

[0078] The specific operation includes but is not limited to at least one of a touch operation, a voice operation, an operation via an input device such as a mouse, or a gesture operation, and may be, for example, a click operation, a double-click operation, a long press operation, a left swipe operation, or a right swipe operation, without further limitation. It is understood that the preset operation and the specific operation may be identical or partially identical.

[0079] The target operation frequency refers to the number of times an object identifier operates on the target resource tag of a target resource within a preset time and range. For example, the number of times object identifier A clicks on target resource tag 3 of target resource i within 10 days.

[0080] Specifically, the computer device determines the target resource tag corresponding to the target resource and obtains target operation data obtained by performing a specific operation on each target resource tag of the target resource by the object identifier. If no specific operation is performed, the corresponding target operation data may be no operation.

[0081] For each target resource tag of the target resource, the computer device calculates the target operation frequency of the object identifier for the target resource tag based on the target operation data corresponding to the target resource tag. Using the same processing method, the target operation frequency corresponding to each target resource tag under the same target resource can be calculated separately.

[0082] In one embodiment, the computer device can represent the corresponding target operation data by a preset numerical value. If the object identifier performs a specific operation on the target resource tag 1 of the target resource, the target operation data of the target resource tag 1 of the target resource can be represented by the numerical value 1; if the object identifier does not perform a specific operation on the target resource tag 2 of the target resource, the target operation data corresponding to the target resource tag 2 of the target resource can be represented by the numerical value 0.

[0083] In one embodiment, determining a target operation frequency of the object identifier on each target resource tag of the target resource based on target operation data obtained by the object identifier operating on each target resource tag of the target resource includes:

[0084] Obtain target operation data of each target resource tag of the object identifier for different versions of the target resource, and determine the target operation frequency of the object identifier for each target resource tag of the target resource based on the target operation data of different versions corresponding to each target resource tag; different versions correspond to different timestamps.

[0085] Specifically, for each target resource tag of the target resource, the computer device obtains the target operation data obtained by operating each target resource tag by the object identifier at different timestamps. Each timestamp corresponds to a version, so different versions of target operation data can be obtained at different timestamps.

[0086] The target resource tag corresponds to multiple different versions of target operation data. The computer device sums the multiple different versions of target operation data to obtain the target operation frequency of the object identifier for the target resource tag. Using the same processing method, the target operation frequency of the object identifier for each target resource tag of the target resource can be obtained separately.

[0087] In one embodiment, the target operation frequency may be calculated according to the following formula:

[0088]

[0089] Among them, f(u,i,t) is the target operation frequency of object identifier u on target resource tag t with version v of target resource i.

[0090] Step S208 : determining the attention degree of the object identifier to the target resource based on the object operation frequency, the operation frequencies of each tag, and the operation frequencies of each target.

[0091] The attention degree refers to the degree of attention of the object identifier to the target resource, that is, the degree of interest of the user in the target resource.

[0092] Specifically, the computer device calculates the attention degree of the object identifier to the target resource based on the object operation frequency, the operation frequency of each tag, and the target operation frequency of the object identifier on each target resource tag of the target resource.

[0093] In one embodiment, the computer device may calculate the object identifier's attention to the target resource based on the object operation frequency, the tag operation frequency, the target operation frequency, and their corresponding weights.

[0094] In this embodiment, object operation data for candidate resource tags by the object identifier is obtained, and the object operation frequency of the object identifier is determined based on the object operation data. This allows accurate calculation of the frequency of user operations based on the resource operation data for the candidate resource tags. Resource operation data obtained from operations performed on each target resource tag is obtained, and the tag operation frequency of each target resource tag operated on is determined based on the resource operation data. This allows accurate calculation of the frequency of target resource tag operations based on the resource operation data generated on the target resource tags, effectively reflecting the frequency of target resource tag operations. The target operation frequency of the object identifier for each target resource tag of the target resource is determined based on the target operation data obtained from operations performed by the object identifier on each target resource tag of the target resource, effectively reflecting the frequency of operations performed by the object identifier on each target resource tag of the target resource. The degree of attention of the object identifier to the target resource is determined based on the object operation frequency, the frequency of each tag operation, and the frequency of each target operation. This allows accurate calculation of the degree of attention of the object identifier to the target resource based on the frequency of user operations on the candidate resource tags, the frequency of operations performed on the target resource tags, and the frequency of operations performed by the object identifier on each target resource tag of the target resource. In addition, the attention is calculated by counting frequency data such as the user's operation frequency and the frequency of target resource tags being operated. The calculation amount is small and the processing speed is fast, which can effectively improve the real-time processing efficiency of the user's interest in the target resource.

[0095] In traditional operation data processing methods, interest scores are usually calculated using the user's own attribute features, the relevant features of the user's historical operation data on the target resource, and the characteristics of the target resource itself. These features need to be extracted from the relevant raw data. When the number of users and items is large, traditional methods usually use batch processing. Batch processing often requires a batch of data to be collected before processing begins. Collecting a batch of user raw data, a batch of target resource related raw data, and extracting key features from each raw data takes a lot of time and the amount of processing calculation is also relatively large. It is difficult to achieve real-time data calculation, resulting in low processing efficiency. In this embodiment, by counting frequency data such as the frequency of user operations and the frequency of target resource tags being operated, there is no need for feature extraction, the amount of calculation is small, and each piece of data or each small batch of data can be processed immediately when it arrives. It can effectively achieve streaming calculation, thereby improving the efficiency of real-time data processing.

[0096] In one embodiment, Figure 3 As shown, obtaining object operation data of the object identifier for the candidate resource tag and obtaining resource operation data obtained by the operation on each target resource tag includes step S302:

[0097] Step S302 : obtaining object operation data of the object identifier for candidate resource tags of different versions, and obtaining resource operation data obtained by operating on each target resource tag of different versions; different versions correspond to different timestamps.

[0098] Specifically, the computer device can obtain object operation data for each candidate resource tag at different timestamps. Different versions correspond to different timestamps, that is, each timestamp corresponds to a version. Therefore, different versions of candidate resource tags exist, and different versions of object operation data can be obtained at different timestamps.

[0099] According to the same processing method, the computer device can obtain resource operation data obtained by the preset operation on each target resource tag at different timestamps, and can obtain different versions of resource operation data for each target resource tag.

[0100] In one embodiment, in the same version, the timestamp of obtaining the object operation data and the timestamp of obtaining the resource operation data are the same. For example, if the object operation data is obtained at 12:00 and the resource operation data is obtained at 12:00, the timestamps of the two are the same and the version is the same.

[0101] Determining the object operation frequency of the object identifier based on the object operation data, and determining the tag operation frequency of each target resource tag being operated based on the resource operation data, including steps S304 to S306:

[0102] Step S304: determining the object operation frequency corresponding to the object identifier based on the object operation data of different versions.

[0103] Specifically, the computer device may calculate the object operation frequency of the object identifier based on the object operation data of different versions. Further, the computer device may sum the object operation data of different versions to obtain the object operation frequency of the object identifier.

[0104] In one embodiment, the computer device may represent object operation data using a preset value. For example, if the object identifier performs a preset operation on the candidate resource tag, the object operation data is represented by a value of 1; if the object identifier does not perform a preset operation on the candidate resource tag, the object operation data is represented by a value of 0. The computer device represents different versions of object operation data using corresponding preset values, sums the preset values ​​corresponding to the different versions of object operation data, and obtains the object operation frequency for the object identifier.

[0105] For example, the object operation frequency can be calculated using the following formula:

[0106]

[0107] Among them, Cu The object operation frequency is an accumulation of object operation data of each candidate resource label t under different versions v. V is a version, V is the number of all versions, t is a candidate resource label, and T is the number of all candidate resource labels. Map(u, v).value is a value corresponding to the candidate resource label t of the object identifier u in version v, that is, a preset numerical value corresponding to the object operation data of the object identifier u on the candidate resource label t in version v.

[0108] In an embodiment, Map(u, v).value takes a value of 0 or 1. If the object identifier u performs a preset operation on the candidate resource label t of version v, Map(u, v).value is 1; if the object identifier u does not perform a preset operation on the candidate resource label t of version v, Map(u, v).value is 0.

[0109] In step S306, the label operation frequency of each target resource label being operated is determined based on resource operation data of different versions.

[0110] Specifically, for each target resource label, the computer device can calculate the label operation frequency of the target resource label being operated according to the resource operation data of different versions corresponding to the target resource label. In the same way, the label operation frequency of each target resource label being operated can be calculated.

[0111] Further, for each target resource label, the computer device can sum the resource operation data of different versions corresponding to the same target resource label to obtain the label operation frequency corresponding to the same target resource label.

[0112] In an embodiment, the computer device can represent the resource operation data by a preset numerical value, for example, if the target resource label is operated, the resource operation data is represented by a numerical value 1, and if the target resource label is not operated, the resource operation data is represented by a numerical value 0. The computer device represents the resource operation data of different versions by corresponding preset numerical values, sums the preset numerical values corresponding to the resource operation data of different versions, and obtains the label operation frequency of the target resource label.

[0113] For example, the label operation frequency can be calculated by the following formula:

[0114]

[0115] Wherein, C t The label operation frequency of the target resource label t is an accumulation of resource operation data of the candidate resource label t under different versions v. Map(t, v).value is a value corresponding to the target resource label t of version v, that is, a preset numerical value corresponding to the resource operation data of the target resource label t of version v.

[0116] In one embodiment, Map(t,v).value takes the value of 0 or 1. If the target resource tag t of version v is subjected to a preset operation, Map(t,v).value is 1; if the target resource tag t of version v is not subjected to a preset operation, Map(t,v).value is 0.

[0117] In this embodiment, object operation data of candidate resource tags with different versions of object identifiers are obtained, and the frequency of object operations corresponding to the object identifiers is determined based on the object operation data of different versions. The frequency of user operations on candidate resource tags obtained at different timestamps can be accurately calculated, effectively reflecting the frequency of user operations. Resource operation data obtained by operations on each target resource tag with different versions are obtained, and the frequency of tag operations on each target resource tag with different versions is determined based on the resource operation data of different versions. The frequency of target resource tags with different operations can be accurately calculated through the resource operation data of the target resource tags obtained at different timestamps, effectively reflecting the frequency of target resource tags with different operations.

[0118] In one embodiment, after determining the object operation frequency of the object identifier based on the object operation data and determining the tag operation frequency of each target resource tag being operated based on the resource operation data, the method further includes: determining the object weight of the object identifier according to the object operation frequency, and determining the tag weight corresponding to each target resource tag according to each tag operation frequency;

[0119] The attention degree of the object identifier to the target resource is determined based on the object operation frequency, the operation frequency of each tag and the operation frequency of each target, including: determining the attention degree of the object identifier to the target resource based on the object weight, the weight of each tag and the operation frequency of each target.

[0120] Specifically, the computer device may determine the object weight corresponding to the object identifier based on the object operation frequency. Further, the computer device may calculate the inverse of the object operation frequency and use the inverse of the object operation frequency as the object weight corresponding to the object identifier.

[0121] For each target resource tag, the tag weight corresponding to the target resource tag is calculated based on the tag operation frequency corresponding to the target resource tag. Furthermore, the computer device can calculate the reciprocal of each tag operation frequency and use each reciprocal as the tag weight of the corresponding target resource tag.

[0122] For example, the object weight and label weight can be calculated by the following formula:

[0123]

[0124] Among them, λu is an object weight, C u is an object identification corresponding object operation frequency, π t is a tag weight, C t is a tag operation frequency.

[0125] In this embodiment, the object weight of the object identification is determined according to the object operation frequency, which can accurately reflect the proportion of the operation frequency of the object identification on the candidate resource tag in all candidate resource tags. The tag weight corresponding to each target resource tag is determined according to the tag operation frequency, which can accurately reflect the proportion of the operation frequency of each target resource tag in the operation frequency of all target resource tags. Based on the object weight, the tag weight, and the target operation frequency, the attention degree of the object identification to the target resource can be accurately determined.

[0126] In one embodiment, based on the object weight, the tag weight, and the target operation frequency, the attention degree of the object identification to the target resource is determined, comprising:

[0127] According to the object weight, the tag weight, and the target operation frequency, the operation evaluation value of the object identification to each target resource tag is determined; and based on the time decay factor and the operation evaluation value, the attention degree of the object identification to the target resource is determined.

[0128] The time decay factor is an adjustment coefficient for adjusting different time lengths, for example, the time decay factors corresponding to the time lengths of 00:00-01:00, 01:00-02:00, 18:00-19:00, etc. The time decay factor can be a linear time decay factor, a Gaussian time decay factor, etc., but is not limited thereto.

[0129] Specifically, for each target resource tag, the computer device calculates the operation evaluation value of the object identification to the target resource tag of the target resource according to the tag weight corresponding to the target resource tag and the corresponding target operation frequency, and the object weight. According to the same processing mode, the operation evaluation value of the object identification to each target resource tag can be calculated.

[0130] The computer device obtains the time decay factor, and calculates the attention degree of the object identification to the target resource based on the time decay factor and the operation evaluation value corresponding to each target resource of the target resource.

[0131] Further, the computer device can calculate a time decay factor corresponding to the time length between the time when at least one of the object operation data, the resource operation data, and the target operation data is obtained and the current time according to the time and the current time. For example, the time decay factor is α l-1, where l is the duration, which can be measured in minutes, hours or days.

[0132] In one embodiment, the computer device may use the product of the object weight, the tag weight corresponding to the target resource tag, and the corresponding target operation frequency as the operation evaluation value of the object identifier for the target resource tag.

[0133] In one embodiment, the computer device may sum the operation evaluation values ​​corresponding to each target resource, and use the product of the sum of the operation evaluation values ​​and the time decay factor as the object identifier's attention to the target resource.

[0134] In this embodiment, the operation evaluation value of the object identifier for each target resource tag is determined based on the object weight, the weight of each tag, and the frequency of each target operation. This allows the object identifier to calculate the operation evaluation value for the target resource tag based on multiple evaluation dimensions, such as the proportion of user operations, the proportion of target resource tags being operated, and the frequency of the object identifier operating the target resource tag of the target resource. This takes into account multiple factors, making the calculated operation evaluation dimension more accurate and comprehensive. The time decay factor can effectively control the validity and importance of the acquired data over different time periods. The time decay factor can take into account the different impacts of the acquired data in different time periods, making the attention calculated based on the time decay factor and each operation evaluation value more accurate.

[0135] In one embodiment, Figure 4 As shown, according to the object weight, each tag weight and each target operation frequency, the operation evaluation value of the object identifier for each target resource tag is determined, including steps S402 to S406:

[0136] Step S402 : determining the candidate operation frequency corresponding to each target resource tag of each candidate resource by the object identifier, where the candidate resources include the target resource.

[0137] Specifically, the computer device identifies each candidate resource corresponding to a target resource tag. For example, if the target resource tag is "recommended," the computer device then "recommends" each candidate resource under that tag. For each candidate resource, the computer device obtains candidate operation data for each target resource tag associated with the object identifier of the candidate resource and determines a corresponding candidate operation frequency based on the candidate operation data.

[0138] For the acquisition of candidate operation data, please refer to the acquisition of target operation data in the above embodiments; for the calculation of target operation frequency, please refer to the calculation of object operation frequency or tag operation frequency in the above embodiments.

[0139] The target resource can be included in the plurality of candidate resources, and the candidate operation frequency corresponding to the candidate resource serving as the target resource is the target operation frequency corresponding to the target resource.

[0140] In step S404, a sub-evaluation value of each target resource label for a corresponding candidate resource is determined according to the object weight, each label weight, and the candidate operation frequency.

[0141] Specifically, for each candidate resource, the computer device calculates a sub-evaluation value corresponding to a target resource label of the object identifier for the candidate resource according to the object weight, the label weight corresponding to the target resource label, and the target operation frequency of the candidate resource for the target resource label. According to the same processing manner, the sub-evaluation value corresponding to each target resource label of each candidate resource can be calculated respectively.

[0142] In step S406, an operation evaluation value of the object identifier for each target resource label is determined based on the sub-evaluation values.

[0143] Specifically, the computer device can calculate an operation evaluation value of the object identifier for the same target resource label according to the sub-evaluation values corresponding to each candidate resource under the same target resource label. According to the same processing manner, the operation evaluation value corresponding to each target resource can be calculated respectively.

[0144] In one embodiment, the computer device can sum the sub-evaluation values corresponding to the same target resource label to obtain the operation evaluation value of the object identifier for the same target resource label.

[0145] For example, the operation evaluation value of the target resource label can be calculated according to the following formula:

[0146]

[0147] Wherein, g(u, t) is the operation evaluation value of the object identifier u for the tth target resource label, i represents the ith candidate resource, which can also be a target resource, I is the number of candidate resources, and f(u, i, t) is the candidate operation frequency of the object identifier u for the ith candidate resource for the tth candidate resource label with the version v.

[0148] It can be understood that when the ith candidate resource is a target resource, the tth candidate resource label of the ith candidate resource is the tth target resource label of the target resource.

[0149] Based on the time decay factor and the operation evaluation values, a degree of attention of the object identifier to the target resource is determined, including step S408:

[0150] Step S408 , selecting a target operation evaluation value that meets the evaluation conditions from each operation evaluation value, and determining the object identifier's attention to the target resource based on the time decay factor and the target operation evaluation value.

[0151] Here, satisfying the evaluation condition means that the operation evaluation value is greater than the evaluation threshold, or the difference between the operation evaluation value and the evaluation threshold is less than a preset difference.

[0152] Specifically, the computer device may obtain an evaluation condition, filter out a target operation evaluation value from each operation evaluation value based on the evaluation condition, and calculate the object identifier's attention to the target resource based on the filtered target operation evaluation value and the time decay factor.

[0153] In one embodiment, the computer device may sum the evaluation values ​​of each target operation, and use the product of the sum of the evaluation values ​​of each target operation and the time decay factor as the object identifier's attention to the target resource.

[0154] In this embodiment, the candidate operation frequency corresponding to each target resource tag of each candidate resource for the object identifier is determined. Based on the object weight, the weight of each tag, and the candidate operation frequency, the sub-evaluation value of each target resource tag for the corresponding candidate resource is determined, so that the sub-evaluation value can be accurately calculated by combining multiple factors. Based on each sub-evaluation value, the operation evaluation value of the object identifier for each target resource tag is determined, and the target operation evaluation value that meets the evaluation conditions is screened out from each operation evaluation value. This can screen out lower operation evaluation values ​​and reduce interference with the final result. Therefore, based on the time attenuation factor and the target operation evaluation value, the object identifier's attention to the target resource can be accurately calculated.

[0155] In one embodiment, Figure 5 As shown, the method further includes:

[0156] Step S502 : determining the valid operation times of the object identifier for each target resource tag in the object operation data acquired within a preset time period.

[0157] The number of valid operations refers to the number of valid operations performed by the object identifier on the target resource tag. A valid operation is one that satisfies the preset operation. For example, if the preset operation is a double-click operation and the object operation data contains both a single-click operation and a double-click operation, the single-click operation is considered invalid, while the double-click operation is considered valid.

[0158] Specifically, the computer device obtains object operation data for each candidate resource tag associated with the object identifier within a preset time period. The computer device may determine target resource tags present in the candidate resources and, for each target resource tag associated with the object operation data, determine the number of valid operations performed by the object identifier associated with the target resource tag from the object operation data associated with the target resource tag.

[0159] In other embodiments, the computer device may determine the number of valid operations of the object identifier for each target resource tag from resource operation data acquired within a preset time period.

[0160] In one embodiment, determining the number of valid operations of the object identifier for each target resource tag from the resource operation data acquired within a preset time period includes:

[0161] The preset time period is divided into a plurality of time periods through a preset time window, and the number of valid operations of the object identifier in the resource operation data for each target resource tag in each time period is determined.

[0162] Step S504 : determining a stable evaluation value of the object identifier operating on each target resource tag based on the number of valid operations and the preset duration.

[0163] The stability evaluation value refers to the stability of the object identifier in operating the target resource tag within a preset time period.

[0164] Specifically, the computer device calculates the stable evaluation value of the object identifier for each target resource tag according to the number of valid operations and the preset duration of each target resource.

[0165] Furthermore, the computer device uses the ratio of the number of valid operations corresponding to the target resource tag to the preset duration as the stability evaluation value of the target resource tag. According to the same processing method, the stability evaluation value corresponding to each target resource tag can be obtained.

[0166] Based on the object operation frequency, the tag operation frequency, and the target operation frequency, the attention degree of the object identifier to the target resource is determined, including step S506:

[0167] Step S506 : determining the attention degree of the object identifier to the target resource based on the object operation frequency, the operation frequencies of each tag, the operation frequencies of each target, and the stability evaluation values.

[0168] Specifically, the computer device can calculate the object weight based on the object operation frequency and the tag operation frequency corresponding to each target resource tag based on the tag operation frequency. Based on the object weight, the tag operation frequency, and the target operation frequency, the computer device calculates the object identifier's operation evaluation value for each target resource tag. Based on the object identifier's operation evaluation value for each target resource tag and the object identifier's stability evaluation value for each target resource tag, the computer device calculates the object identifier's attention to the target resource.

[0169] Furthermore, the computer device can calculate the product of the operation evaluation value of the same target resource tag and the stable evaluation value of the target resource tag to obtain the evaluation value product corresponding to each target resource tag, and use the sum of the evaluation value products as the object identifier's attention to the target resource.

[0170] In this embodiment, the number of valid operations performed by the object identifier on each target resource tag is determined from the object operation data acquired within a preset time period, thereby eliminating the adverse effects of invalid user operations on the results. This allows for the accurate calculation of the stable evaluation value of the object identifier's operations on each target resource tag based on each valid operation number and the preset time period. Based on the more accurate calculated stable evaluation value, the object identifier's attention to the target resource can be more accurately determined based on the object operation frequency, the operation frequency of each tag, the operation frequency of each target, and each stable evaluation value.

[0171] In one embodiment, Figure 6 As shown, in the object operation data obtained within the preset time period, determining the valid operation times of the object identifier for each target resource tag includes step S602:

[0172] Step S602 : dividing the preset duration into a plurality of time periods through a preset time window, and determining the number of valid operations of the object identifier in the object operation data for each target resource tag in each time period.

[0173] Specifically, the computer device obtains object operation data for each candidate resource tag of the object identifier within a preset time period. The computer device may obtain a preset time window, which is a preset time length, such as 10 minutes, 20 minutes, 1 hour, etc., but is not limited thereto.

[0174] The computer device divides the preset time into multiple time periods through a preset time window, and determines from each object operation data the number of valid operations performed by the object identifier on each target resource tag in each time period.

[0175] In one embodiment, the computer device may determine target resource tags present in the candidate resources, and for each target resource tag corresponding to the object operation data, the computer device may determine the number of valid operations performed by the object identifier on the target resource tag within each time period from the object operation data corresponding to the target resource tag. The number of valid operations performed by the object identifier on each target resource tag within each time period may be obtained using the same processing method.

[0176] Based on each valid operation number and the preset duration, a stable evaluation value of the object identifier operating on each target resource tag is determined, including steps S604 to S606:

[0177] Step S604 : Filter out the target operation times and corresponding target time periods that meet the valid operation conditions from the valid operation times corresponding to each time period.

[0178] Here, satisfying the valid operation condition means that the number of valid operations is greater than the valid threshold.

[0179] Specifically, the computer device obtains the valid operation condition, selects the target operation number that meets the valid operation condition from each valid operation number, and uses the time period corresponding to the target operation number that meets the valid operation condition as the target time period.

[0180] In one embodiment, the computer device compares the number of valid operations corresponding to each time period with the valid threshold, selects the number of valid operations greater than the valid threshold as the target number of operations, and uses the time period corresponding to the target number of operations as the target time period.

[0181] Step S606: Determine a stable evaluation value of the object identifier operating on each target resource tag according to the target operation times and the target time period.

[0182] Specifically, the computer device determines the target number of operations and the target time period corresponding to the same target resource tag. The ratio of the sum of the target number of operations corresponding to the same target resource tag to the target time period is used as the stability evaluation value of the object identifier's operations on the target resource tag. Using the same processing method, the stability evaluation value of the object identifier's operations on each target resource tag can be calculated separately.

[0183] In one embodiment, a computer device determines the target number of operations and the target time period corresponding to the same target resource tag, and uses the ratio between the sum of the target number of operations corresponding to the same target resource tag and the number of target time periods corresponding to the target resource tag as a stable evaluation value of the object identifier's operation on the target resource tag.

[0184] In one embodiment, after determining the number of valid operations of the object identifier for each target resource tag in each time period, the computer device filters out the target operation number that meets the valid operation conditions from the valid operation number corresponding to each time period, and determines the stable evaluation value of the object identifier's operation on each target resource tag based on the target operation number and preset duration corresponding to each target resource tag.

[0185] For example, the computer device can calculate the stability evaluation value P(u,t) by the following formula:

[0186]

[0187] Where k represents the kth time period; K is the number of time periods, that is, the number of time windows; x is the number of valid operations, β(x) is the threshold of the proportion of time windows with valid operations, and σ is the effective threshold.

[0188] In this embodiment, a preset time window is used to divide a preset duration into multiple time periods. The number of valid operations performed by the object identifier in the object operation data for each target resource tag in each time period is determined. From the valid operation counts corresponding to each time period, the target operation counts and corresponding target time periods that meet the valid operation conditions are screened out, thereby filtering out some data with a smaller number of valid operations and reducing the amount of data. Furthermore, the impact of the filtered valid operation counts on the final result is very small. Removing this portion of data can further improve processing efficiency, enabling the rapid calculation of a stable evaluation value for the object identifier's operation on each target resource tag based on the target operation counts and target time period.

[0189] In one embodiment, Figure 7 As shown, in the object operation data obtained within the preset time period, determining the valid operation times of the object identifier for each target resource tag includes step S702:

[0190] Step S702 : determining the number of valid operations of the object identifier for each target resource tag under each promotion channel from the object operation data of different promotion channels acquired within a preset time period.

[0191] Among them, promotion channels refer to channels for promoting information, including but not limited to applications, web pages, etc. Applications may further include parent applications and child applications running on the parent application. Parent applications refer to applications that can run independently. Child applications refer to applications that cannot run independently and require the help of parent applications to run. Both parent applications and child applications may include instant messaging applications, SNS (Social Network Sites) applications, short video applications, long video applications, game applications, music sharing applications, shopping and selling applications, UGC (User Generated Content) applications, and various types of intelligent recognition applications, but are not limited to these.

[0192] Specifically, the computer device obtains object operation data of the object identifier for each candidate resource tag from different promotion channels within a preset time period.

[0193] From each object operation data of different promotion channels, determine the object operation data corresponding to the object identifier and each target resource tag. From the object operation data corresponding to the object identifier under each promotion channel, determine the valid operation times of the object identifier for each target resource tag.

[0194] In other embodiments, the number of valid operations of the object identifier for each target resource tag under each promotion channel is determined from resource operation data of different promotion channels acquired within a preset time period.

[0195] Based on each valid operation number and the preset duration, a stable evaluation value of the object identifier operating on each target resource tag is determined, including step S704:

[0196] Step S704 : for each promotion channel, according to the corresponding valid operation times and preset durations under the corresponding promotion channel, determine the stable evaluation value of the object identifier operating each target resource tag under the corresponding promotion channel.

[0197] Specifically, the computer device calculates the stable evaluation value of the object identifier for each target resource tag under the promotion channel based on the number of valid operations on each target resource and the preset duration. Following the same processing method, the stable evaluation value of the object identifier's operations on each target resource tag under each promotion channel can be obtained.

[0198] Furthermore, the computer device uses the ratio of the number of valid operations corresponding to the target resource tag under the promotion channel to the preset duration as a stable evaluation value of the target resource tag under the promotion channel.

[0199] Based on the object operation frequency, the operation frequencies of each tag, the operation frequencies of each target, and the stability evaluation values, the attention degree of the object identifier to the target resource is determined, including step S706:

[0200] Step S706 : determining the attention degree of the object identifier to the target resource based on the object operation frequency, each tag operation frequency, each target operation frequency and each stability evaluation value corresponding to each promotion channel.

[0201] Specifically, the computer device can calculate the object weight corresponding to each promotion channel based on the object operation frequency corresponding to each promotion channel; and calculate the tag operation frequency corresponding to each target resource tag in each promotion channel based on the tag operation frequency in each promotion channel. The computer device calculates the operation evaluation value of the object identifier for each target resource tag in different promotion channels based on the object weight in different promotion channels, the tag operation frequency in different promotion channels, and the target operation frequency in different promotion channels. The computer device calculates the object identifier's attention to the target resource based on the object identifier's operation evaluation value for each target resource tag in different promotion channels and the object identifier's stable evaluation value for each target resource tag in different promotion channels.

[0202] Furthermore, the computer device can calculate the product of the operational evaluation value of the same target resource tag in the same promotion channel and the stable evaluation value of the target resource tag in the same promotion channel to obtain the product of the evaluation values ​​corresponding to each target resource tag in the same promotion channel. The sum of the products of the evaluation values ​​in each promotion channel is used as the object identifier's attention level for the target resource.

[0203] In this embodiment, the calculation of attention level combines data from various dimensions collected from different promotion channels over a certain period of time. From the object operation data of different promotion channels acquired within a preset time period, the number of valid operations performed by the object identifier on each target resource tag under each promotion channel is determined. For each promotion channel, the stable evaluation value of the object identifier's operations on each target resource tag under that channel is determined based on the corresponding number of valid operations and the preset time period. This allows for a more comprehensive consideration of factors affecting the accuracy of attention level from multiple aspects and dimensions, effectively improving the accuracy of attention level calculation.

[0204] In one embodiment, a method for processing operation data is provided, such as Figure 8 Shown, including:

[0205] Step S802: Collect data and report to the server:

[0206] The user operation and target operation data under different promotion channels are collected by software development kit (SDK) and reported to the backend server, and the backend server sends the data to Kafka. Kafka is an open source stream processing platform that can process all action stream data of users in the website.

[0207] The operation of the object identifier u on the target resource tag t on the target resource i at a certain moment is defined as target operation data A(u, i, t), and all operations of the object identifier u are defined as object operation data A u , and all operations on the target resource tag t are defined as resource operation data A t The object identifier is the user identifier.

[0208] Flink consumes Kafka data and extracts tags from item data and associates them with user data. Flink is a framework for unified stream processing and batch processing.

[0209] The object identifier u, the target operation data A(u, i, t), the object operation data A u , and the tag resource operation data A t are stored on Redis respectively. Redis is a key-value storage system.

[0210] Among them, the target operation data A(u, i, t) corresponding to the object identifier u is stored as Map(u, i, t, v), the key is (u, i, t), the current version number is v, and the timestamp of each write corresponds to a version number v; the value corresponding to (u, i, t, v) is 1, indicating that the operation has occurred. The object operation data A u is stored as Map(u, v), wherein the key is the object identifier u, the current version number is v, and the timestamp of each write corresponds to a version number v; the value corresponding to each (u, v) is 1, indicating that the operation has occurred. The resource operation data A t is stored as Map(t, v), wherein the key is the target resource tag t, the current version number is v, and the timestamp of each write corresponds to a version number v; the value corresponding to each (t, v) is 1, indicating that the operation has occurred.

[0211] Further, the expiration time w of Redis can be determined at the same time, and when the object identifier u, the target operation data A(u, i, t), the object operation data A u , and the resource operation data A t on Redis expire, the data corresponding to the version number less than the expiration time in Map(u, i, t, v), Map(u, v), and Map(t, v) is cleared.

[0212] Step S804, calculating the object operation frequency C of the object identifier u u :

[0213]

[0214] Step S806, calculating the label operation frequency C corresponding to each target resource label t t :

[0215]

[0216] Step S808, calculating the object weight λ corresponding to the object identifier u u , and the label weight π corresponding to each target resource label t respectively t :

[0217]

[0218] Step S810, calculating the target operation frequency f(u, i, t) of the object identifier u to each target resource label t of the target resource i:

[0219]

[0220] Step S812, calculating the sum of the operation scores of the object identifier u to each target resource label t, i.e., the operation evaluation value g(u, t):

[0221]

[0222] Step S814, calculating the timeliness decay factor α corresponding to the object identifier u k-1 , which is decayed by time:

[0223] α k-1 is a fixed decay coefficient, and k is the time length from now. α can be 0.5.

[0224] Step S816, calculating the stability score of the object identifier u to each target resource label t, i.e., the stability evaluation value p(u, s, t):

[0225] The stability evaluation value p(u, s, t) represents the window number ratio of the object identifier u performing effective operation on the target resource label t under the promotion channel s within the effective observation period. If the ratio is higher than a specified ratio, it is retained, otherwise it is filtered out.

[0226]

[0227] wherein k represents the kth time period; K represents the number of time periods, i.e., the number of time windows; x is the effective operation frequency, β(x) is the threshold of the time window number ratio of effective operation, and σ is the effective threshold.

[0228] Step S818: Perform strength screening on each operation evaluation value g(u,t). If g(u,t) is greater than a specified threshold, Then keep it, otherwise discard it:

[0229]

[0230] Step S820: Calculate the final interest score of the object identifier u for the target resource i, that is, the attention s(u,t):

[0231]

[0232] In one embodiment, the target resource is promotion information; the method further includes: based on the attention of each object identifier to the promotion information, screening out target object identifiers whose attention meets the push condition; and pushing associated information related to the promotion information to the target object identifiers.

[0233] Promotional information refers to information created by the promotion demander for the content, product, or service being promoted. Specifically, it may include an ad title, ad link, thumbnail, content summary, partial ad content, or all ad content. Push conditions may include the following: the attention level corresponding to the object identifier is greater than the attention level threshold, or the difference between the attention level corresponding to the object identifier and the attention level threshold is within a preset range.

[0234] Specifically, the computer device can determine the promotional information that needs to calculate the attention, and determine the target resource tag corresponding to the promotional information. The computer device obtains the object operation data of the object identifier for the candidate resource tag, and obtains the resource operation data obtained by the operation on each target resource tag; the candidate resource tag includes the target resource tag. The computer device determines the object operation frequency of the object identifier based on the object operation data, and determines the tag operation frequency of each target resource tag being operated based on the resource operation data. The computer device determines the target operation frequency of the object identifier for each target resource tag of the promotional information based on the target operation data obtained by the object identifier operating on each target resource tag of the promotional information. The computer device determines the attention of the object identifier to the promotional information based on the object operation frequency, the operation frequency of each tag and the frequency of each target operation. According to the same processing method, the attention of each object identifier to the promotional information can be calculated separately.

[0235] The computer device selects target object identifiers whose attention levels meet the push conditions based on the attention levels of each object identifier for the promotional information. The computer device obtains associated information related to the promotional information and pushes the associated information to the target object identifiers. For example, the computer device compares the attention levels corresponding to each object identifier with an attention threshold and selects object identifiers with an attention level greater than the attention threshold as target object identifiers; the computer device compares the attention levels corresponding to each object identifier with the attention threshold and selects object identifiers whose differences fall within a preset range as target object identifiers.

[0236] In this embodiment, based on the attention of each object identifier to the promotion information, the target object identifiers whose attention meets the push conditions are screened out, and the associated information related to the promotion information is pushed to the target object identifiers, which can effectively realize the personalized push of information and is conducive to the effective dissemination and promotion of information.

[0237] In one embodiment, the target resource belongs to a member type object; the method further includes: determining the membership activation probability of the object identifier for the member type object based on the object identifier's attention to the member type object; and selecting a target virtual gift resource from the candidate virtual gift resources and issuing it to the object identifier based on the member activation probability and the consumption resource value corresponding to the member type object.

[0238] The membership activation probability refers to the predicted probability of a user activating a membership. Virtual gift resources refer to virtual items distributed to the resource account corresponding to the object identifier, including at least one of account value, red envelopes, gift certificates, coupons, e-cards, avatar products, virtual recharge cards, and game equipment.

[0239] Specifically, the computer device obtains object operation data of the object identifier for candidate resource tags, and obtains resource operation data obtained by operations performed on each target resource tag; the candidate resource tags include the target resource tags. The computer device determines the object operation frequency of the object identifier based on the object operation data, and determines the tag operation frequency of each target resource tag being operated based on the resource operation data. The computer device determines the target operation frequency of the object identifier for each target resource tag of the member type object based on the target operation data obtained by the object identifier performing operations on each target resource tag of the member type object. The computer device determines the attention of the object identifier to the member type object based on the object operation frequency, the operation frequency of each tag, and the frequency of each target operation. The computer device calculates the probability of membership activation of the object identifier for the member type object based on the attention of the object identifier to the member type object.

[0240] In one embodiment, the computer device calculates the membership activation probability of the object identifier for each membership type based on the object identifier's attention to each membership type. For example, for two membership types, ordinary membership and premium membership, the computer device calculates the probability of the object identifier activating the ordinary membership based on the object identifier's attention to ordinary membership; and calculates the probability of the object identifier activating the premium membership based on the object identifier's attention to premium membership.

[0241] Specifically, the computer device obtains candidate virtual gift resources, selects a target virtual gift resource from the candidate virtual gift resources based on the membership activation probability of the member type object for the object identifier and the consumption resource value corresponding to the member type object, and distributes the target virtual gift resource to the resource account corresponding to the object identifier.

[0242] In one embodiment, a computer device obtains candidate virtual gift resources and determines the resource value contained in each candidate virtual gift resource. Based on the membership activation probability and the resource consumption value corresponding to the member type object, a target virtual gift resource is selected from the candidate virtual gift resources and distributed to the resource account corresponding to the object identifier.

[0243] In one embodiment, for each candidate virtual gift resource, the computer device may compare the resource value contained in the candidate virtual gift resource with the consumption resource value corresponding to the member type object, and based on the difference between the resource value contained in the candidate virtual gift resource and the consumption resource value, filter out the target virtual gift resource from each candidate virtual gift resource.

[0244] In this embodiment, based on the object identifier's attention to the member type object, the object identifier's membership activation probability for the member type object is accurately calculated. According to the member activation probability and the resource consumption value corresponding to the member type object, a suitable target virtual gift resource is selected from the candidate virtual gift resources and distributed to the object identifier. This can effectively realize personalized push of virtual gift resources, which is conducive to improving the membership activation probability.

[0245] In one embodiment, according to the membership activation probability and the resource consumption value corresponding to the membership type object, a target virtual gift resource is selected from the candidate virtual gift resources and distributed to the object identifier, including:

[0246] Based on the probability of member activation, the resource consumption value corresponding to the member type object, and the resource value contained in each candidate virtual gift resource, the expected benefit value corresponding to each candidate virtual gift resource is determined; from the candidate virtual gift resources, the target virtual gift resource whose expected benefit value meets the issuance conditions is selected, and the target virtual gift resource is issued according to the object identifier.

[0247] Among them, meeting the issuance conditions means that the expected return value reaches the expected threshold, or the expected return value is within a preset threshold range, or the difference between the expected return value and the expected threshold is within a specific range.

[0248] Specifically, the computer device calculates the expected benefit value corresponding to each candidate virtual gift resource based on the membership activation probability, the consumption resource value corresponding to the membership type object, and the resource value contained in each candidate virtual gift resource.

[0249] The computer device can obtain the resource value contained in each candidate virtual gift resource and the resource consumption value corresponding to the member type object. The computer device can calculate the expected benefit value corresponding to the candidate virtual gift resource based on the resource value contained in the candidate virtual gift resource and the resource consumption value corresponding to the member type object, as well as the member activation probability for the member type object. Using the same processing method, the expected benefit value corresponding to each candidate virtual gift resource can be calculated separately.

[0250] For example, the candidate virtual gift resource is a coupon, and the consumption resource value corresponding to the membership type object is the corresponding payment amount of the membership type object. The resource value contained in the candidate virtual gift resource is 8, and the membership type object is a regular member. The computer device can calculate the expected benefit value of object identifier A using the 8 yuan coupon to activate the regular membership based on the 8 yuan coupon, the probability of object identifier A becoming a regular member, and the payment amount of the regular member.

[0251] The computer device obtains a candidate virtual gift resource corresponding to the expected benefit value that meets the distribution conditions based on the expected benefit value corresponding to each candidate virtual gift resource. The computer device uses the obtained candidate virtual gift resource as the target virtual gift resource and distributes the target virtual gift resource to the resource account corresponding to the object identifier.

[0252] In this embodiment, the expected revenue value for each candidate virtual gift resource is accurately calculated based on the membership activation probability, the resource consumption value corresponding to the membership type object, and the resource value contained in each candidate virtual gift resource. This expected revenue value is then used as a condition for issuing the virtual gift resource. Virtual gift resources whose expected revenue value meets the issuance condition are selected and issued to the corresponding user, enabling personalized push of virtual gift resources.

[0253] In one embodiment, user description data corresponding to the object identifier is generated according to the degree of attention of the object identifier to each target resource.

[0254] In one embodiment, user description data corresponding to each object identifier is generated according to the degree of attention of different object identifiers to different target resources.

[0255] In one embodiment, an operation data processing method is provided, which is applied to a computer device and includes:

[0256] Obtain object operation data of the object identifier for candidate resource tags of different versions, and obtain resource operation data obtained by operations on each target resource tag of different versions; different versions correspond to different timestamps.

[0257] The object operation frequency corresponding to the object identifier is determined based on the object operation data of different versions, and the tag operation frequency of each target resource tag being operated is determined based on the resource operation data of different versions.

[0258] Target operation data obtained by operating each target resource tag of different versions of the target resource using the object identifier is obtained.

[0259] For each target resource tag of the target resource, a target operation frequency of the object identifier on each target resource tag of the target resource is determined based on target operation data of different versions.

[0260] The object weight of the object identifier is determined according to the object operation frequency, and the tag weight corresponding to each target resource tag is determined according to the tag operation frequency.

[0261] Determine the candidate operation frequency corresponding to each target resource tag of each candidate resource by the object identifier, where the candidate resources include the target resource.

[0262] Based on the object weight, the weight of each tag, and the candidate operation frequency, the sub-evaluation value of each target resource tag for the corresponding candidate resource is determined.

[0263] Based on each sub-evaluation value, an operation evaluation value of the object identifier on each target resource tag is determined.

[0264] The preset time length is divided into a plurality of time periods through a preset time window, and the number of valid operations of the object identifier in the object operation data for each target resource tag in each time period is determined.

[0265] From the valid operation times corresponding to each time period, the target operation times and the corresponding target time period that meet the valid operation conditions are screened out.

[0266] According to the target number of operations and the target time period, a stable evaluation value of the object identifier operating on each target resource tag is determined.

[0267] A target operation evaluation value that meets the evaluation conditions is selected from each operation evaluation value, and the attention degree of the object identifier to the target resource is determined based on the time decay factor, the operation evaluation value and the target operation evaluation value.

[0268] Based on the attention level of each object identifier to the promotion information, target object identifiers whose attention levels meet the push conditions are screened out; and associated information related to the promotion information is pushed to the target object identifiers.

[0269] In this embodiment, by determining the object operation frequency corresponding to the object identifier based on different versions of object operation data, the frequency of user operations can be accurately calculated through the user operations on candidate resource tags obtained at different timestamps, effectively reflecting the frequency of user operations.

[0270] The tag operation frequency of each target resource tag is determined based on resource operation data of different versions. The frequency of target resource tags being operated can be accurately calculated through the resource operation data of the target resource tags obtained at different timestamps, effectively reflecting the frequency of target resource tags being operated.

[0271] Determining the object weight of an object identifier based on the object operation frequency accurately reflects the proportion of the object identifier's operation frequency on candidate resource tags among all candidate resource tags. Determining the tag weight corresponding to each target resource tag based on the operation frequency of each tag accurately reflects the proportion of the operation frequency of each target resource tag among all target resource tags.

[0272] The target operation data obtained by the object identifier operating on each target resource tag of the target resource is used to determine the target operation frequency of the object identifier on each target resource tag of the target resource, which can effectively remind the frequency of the object identifier's operation on each tag of the target resource.

[0273] Data processing is performed by counting frequency data such as the user's operation frequency and the frequency of target resource tags being operated, with small calculation amount and fast processing speed.

[0274] Based on the object weight, the weight of each tag and the frequency of each target operation, the operation evaluation value of the object identifier for each target resource tag is determined. In this way, the operation evaluation value of the object identifier for the target resource tag can be calculated by combining multiple evaluation dimensions such as the proportion of user operations, the proportion of target resource tags operated, and the frequency of object identifier operations on the target resource tags of the target resources. By taking into account multi-dimensional factors, the calculated operation evaluation value is more accurate and comprehensive.

[0275] The preset time window divides the preset duration into multiple time periods. From the valid operation counts corresponding to each time period, the target number of operations and the corresponding target time period that meet the valid operation conditions are screened out. This allows for filtering out data with a smaller number of valid operations, reducing the data volume. Furthermore, the impact of the filtered valid operation counts on the final results is minimal. Removing this portion of data further improves processing efficiency, enabling the rapid calculation of a stable evaluation value for the object identifier's operations on each target resource tag based on the target number of operations and target time period.

[0276] The time decay factor can effectively control the validity and importance of the acquired data in different time periods. The time decay factor can take into account the different impacts of the acquired data in different time periods.

[0277] Based on the operation evaluation value and stability evaluation value obtained from multi-faceted and multi-dimensional data processing, and combined with the time attenuation factor, it is possible to more comprehensively consider various factors affecting the accuracy of attention, and effectively improve the accuracy of attention calculation.

[0278] Based on the attention of each object identifier to the promotion information, the target object identifiers whose attention meets the push conditions are screened out, and the associated information related to the promotion information is pushed to the target object identifiers, which can effectively realize the personalized push of information and is conducive to the effective dissemination and promotion of information.

[0279] It should be understood that the user information and related data involved in each embodiment are all information and data collected with the user's authorization or after full authorization by all parties. User information includes, but is not limited to, user device information and user personal information, such as object identifiers; related data includes, but is not limited to, data used for display and analysis, such as object operation data, resource operation data, and target operation data. Furthermore, users can choose not to authorize user information and related data, or to refuse to receive push notifications of related information.

[0280] It should be understood that although Figures 2-8 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. In addition, Figures 2-8 At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The order of execution of these steps or stages is not necessarily one by one, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0281] In one embodiment, as shown in Figure 9 An operation data processing apparatus 900 is provided, which can be a part of a computer device in the form of software modules or hardware modules, or a combination of both, and specifically comprises: an acquisition module 902, a frequency determination module 904, an operation module 906, and an operation data processing module 908, wherein:

[0282] The acquisition module 902 is configured to acquire object operation data of an object identifier for candidate resource labels, and acquire resource operation data obtained by performing operations on each target resource label; the candidate resource labels include the target resource labels.

[0283] The frequency determination module 904 is configured to determine an object operation frequency of the object identifier based on the object operation data, and determine a label operation frequency of each target resource label being operated based on the resource operation data.

[0284] The operation module 906 is configured to determine a target operation frequency of each target resource label of the target resource by the object identifier according to target operation data obtained by performing operations on each target resource label of the target resource by the object identifier.

[0285] The operation data processing module 908 is configured to determine a degree of attention of the object identifier to the target resource based on the object operation frequency, the label operation frequency, and the target operation frequency.

[0286] In this embodiment, by obtaining object operation data for candidate resource tags from an object identifier and determining the object operation frequency of the object identifier based on the object operation data, the frequency of user operations on the candidate resource tags can be accurately calculated based on the resource operation data of the user. By obtaining resource operation data obtained from operations on each target resource tag and determining the tag operation frequency of each target resource tag based on the resource operation data, the frequency of target resource tag operations can be accurately calculated based on the resource operation data generated on the target resource tag, effectively reflecting the frequency of target resource tag operations. The target operation frequency of the object identifier on each target resource tag of the target resource is determined based on the target operation data obtained from operations performed by the object identifier on each target resource tag of the target resource, effectively indicating the frequency of operations performed by the object identifier on each target resource tag of the target resource. The degree of attention of the object identifier on the target resource is determined based on the object operation frequency, the frequency of each tag operation, and the frequency of each target operation. The degree of attention of the object identifier on the target resource can be accurately calculated based on the frequency of user operations on the candidate resource tags, the frequency of operations performed on the target resource tags, and the frequency of operations performed by the object identifier on each target resource tag of the target resource. In addition, the attention is calculated by counting frequency data such as the user's operation frequency and the frequency of target resource tags being operated. The calculation amount is small and the processing speed is fast, which can effectively improve the real-time processing efficiency of the user's interest in the target resource.

[0287] In one embodiment, the acquisition module 902 is further configured to acquire object operation data corresponding to candidate resource tags of different versions of the object identifier, and to acquire resource operation data obtained by performing operations on each target resource tag of different versions; different versions correspond to different timestamps;

[0288] The frequency determination module 904 is further configured to determine the object operation frequency corresponding to the object identifier based on the object operation data of different versions; and determine the tag operation frequency of each target resource tag being operated based on the resource operation data of different versions.

[0289] In this embodiment, object operation data of candidate resource tags with different versions of object identifiers are obtained, and the frequency of object operations corresponding to the object identifiers is determined based on the object operation data of different versions. The frequency of user operations on candidate resource tags obtained at different timestamps can be accurately calculated, effectively reflecting the frequency of user operations. Resource operation data obtained by operations on each target resource tag with different versions are obtained, and the frequency of tag operations on each target resource tag with different versions is determined based on the resource operation data of different versions. The frequency of target resource tags with different operations can be accurately calculated through the resource operation data of the target resource tags obtained at different timestamps, effectively reflecting the frequency of target resource tags with different operations.

[0290] In one embodiment, the apparatus further comprises a weight determination module, the weight determination module being configured to determine an object weight of an object identifier according to an object operation frequency, and to determine a tag weight corresponding to each target resource tag according to each tag operation frequency;

[0291] The operation data processing module 908 is further configured to determine the attention degree of the object identifier to the target resource based on the object weight, the weights of each tag, and the frequency of each target operation.

[0292] In this embodiment, the object weight of the object identifier is determined based on the object operation frequency, which can accurately reflect the proportion of the frequency of the object identifier's operation on the candidate resource tag among all candidate resource tags. The tag weight corresponding to each target resource tag is determined based on the operation frequency of each tag, which can accurately reflect the proportion of the frequency of each target resource tag being operated among the frequencies of all target resource tags being operated. Based on the object weight, the weights of each tag and the frequency of each target operation, the object identifier's attention to the target resource can be accurately determined.

[0293] In one embodiment, the operation data processing module 908 is also used to determine the operation evaluation value of the object identifier for each target resource tag based on the object weight, the weight of each tag and the frequency of each target operation; and determine the attention of the object identifier to the target resource based on the time decay factor and each operation evaluation value.

[0294] In this embodiment, the operation evaluation value of the object identifier for each target resource tag is determined based on the object weight, the weight of each tag, and the frequency of each target operation. This allows the object identifier to calculate the operation evaluation value for the target resource tag based on multiple evaluation dimensions, such as the proportion of user operations, the proportion of target resource tags being operated, and the frequency of the object identifier operating the target resource tag of the target resource. This takes into account multiple factors, and the calculated operation evaluation dimension is more accurate and comprehensive. The time decay factor can effectively control the validity and importance of the acquired data over different time periods. The time decay factor can take into account the different impacts of the acquired data in different time periods, making the attention calculated based on the time decay factor and each operation evaluation value more accurate.

[0295] In one embodiment, the operation data processing module 908 is also used to determine the candidate operation frequency corresponding to each target resource tag of each candidate resource of the object identifier, where the candidate resources include the target resource; determine the sub-evaluation value of each target resource tag for the corresponding candidate resource based on the object weight, the weight of each tag, and the candidate operation frequency; determine the operation evaluation value of the object identifier for each target resource tag based on each sub-evaluation value; filter out the target operation evaluation value that meets the evaluation conditions from each operation evaluation value, and determine the attention of the object identifier to the target resource based on the time attenuation factor and the target operation evaluation value.

[0296] In this embodiment, the candidate operation frequency corresponding to each target resource tag of each candidate resource for the object identifier is determined. Based on the object weight, the weight of each tag, and the candidate operation frequency, the sub-evaluation value of each target resource tag for the corresponding candidate resource is determined, so that the sub-evaluation value can be accurately calculated by combining multiple factors. Based on each sub-evaluation value, the operation evaluation value of the object identifier for each target resource tag is determined, and the target operation evaluation value that meets the evaluation conditions is screened out from each operation evaluation value. This can screen out lower operation evaluation values ​​and reduce interference with the final result. Therefore, based on the time attenuation factor and the target operation evaluation value, the object identifier's attention to the target resource can be accurately calculated.

[0297] In one embodiment, the apparatus further includes an evaluation module; the evaluation module is configured to determine the number of valid operations of the object identifier for each target resource tag from the object operation data acquired within a preset time period; and determine a stable evaluation value of the object identifier for operating each target resource tag based on each valid operation number and the preset time period;

[0298] The operation data processing module 908 is further configured to determine the attention degree of the object identifier to the target resource based on the object operation frequency, the operation frequencies of each tag, the operation frequencies of each target, and the stability evaluation values.

[0299] In this embodiment, the number of valid operations performed by the object identifier on each target resource tag is determined from the object operation data acquired within a preset time period, thereby eliminating the adverse effects of invalid user operations on the results. This allows for the accurate calculation of the stable evaluation value of the object identifier's operations on each target resource tag based on each valid operation number and the preset time period. Based on the more accurate calculated stable evaluation value, the object identifier's attention to the target resource can be more accurately determined based on the object operation frequency, the operation frequency of each tag, the operation frequency of each target, and each stable evaluation value.

[0300] In one embodiment, the evaluation module is also used to divide the preset duration into multiple time periods through a preset time window, and determine the number of valid operations of the object identifier in the object operation data for each target resource tag in each time period; from the valid operation numbers corresponding to each time period, screen out the target operation number and the corresponding target time period that meet the valid operation conditions; based on the target operation number and the target time period, determine the stable evaluation value of the object identifier operating on each target resource tag.

[0301] In this embodiment, a preset time window is used to divide a preset duration into multiple time periods. The number of valid operations performed by the object identifier in the object operation data for each target resource tag in each time period is determined. From the valid operation counts corresponding to each time period, the target operation counts and corresponding target time periods that meet the valid operation conditions are screened out, thereby filtering out some data with a smaller number of valid operations and reducing the amount of data. Furthermore, the impact of the filtered valid operation counts on the final result is very small. Removing this portion of data can further improve processing efficiency, enabling the rapid calculation of a stable evaluation value for the object identifier's operation on each target resource tag based on the target operation counts and target time period.

[0302] In one embodiment, the evaluation module is further configured to determine, from object operation data acquired from different promotion channels within a preset time period, the number of valid operations of the object identifier on each target resource tag under each promotion channel; and, for each promotion channel, determine a stable evaluation value of the object identifier's operations on each target resource tag under the corresponding promotion channel based on the corresponding number of valid operations and the preset time period.

[0303] The operation data processing module 908 is further used to determine the attention of the object identifier to the target resource based on the object operation frequency, each tag operation frequency, each target operation frequency and each stability evaluation value corresponding to each promotion channel.

[0304] In this embodiment, the calculation of attention level combines data from various dimensions collected from different promotion channels over a certain period of time. From the object operation data of different promotion channels acquired within a preset time period, the number of valid operations performed by the object identifier on each target resource tag under each promotion channel is determined. For each promotion channel, the stable evaluation value of the object identifier's operations on each target resource tag under that channel is determined based on the corresponding number of valid operations and the preset time period. This allows for a more comprehensive consideration of factors affecting the accuracy of attention level from multiple aspects and dimensions, effectively improving the accuracy of attention level calculation.

[0305] In one embodiment, the target resource is promotion information; the device also includes a promotion module; the promotion module is used to screen out target object identifiers whose attention levels meet the push conditions based on the attention levels of each object identifier to the promotion information; and push associated information related to the promotion information to the target object identifiers.

[0306] In this embodiment, based on the attention of each object identifier to the promotion information, the target object identifiers whose attention meets the push conditions are screened out, and the associated information related to the promotion information is pushed to the target object identifiers, which can effectively realize the personalized push of information and is conducive to the effective dissemination and promotion of information.

[0307] In one embodiment, the target resource belongs to a member type object; the device also includes a distribution module; the distribution module is used to determine the membership activation probability of the object identifier for the member type object based on the object identifier's attention to the member type object; according to the member activation probability and the consumption resource value corresponding to the member type object, the target virtual gift resource is selected from the candidate virtual gift resources and distributed to the object identifier.

[0308] In this embodiment, based on the object identifier's attention to the member type object, the object identifier's membership activation probability for the member type object is accurately calculated. According to the member activation probability and the resource consumption value corresponding to the member type object, a suitable target virtual gift resource is selected from the candidate virtual gift resources and distributed to the object identifier. This can effectively realize personalized push of virtual gift resources, which is conducive to improving the membership activation probability.

[0309] In one embodiment, the issuance module is also used to determine the expected benefit value corresponding to each candidate virtual gift resource based on the member activation probability, the consumption resource value corresponding to the member type object, and the resource value contained in each candidate virtual gift resource; from the candidate virtual gift resources, select the target virtual gift resource whose expected benefit value meets the issuance conditions, and issue the target virtual gift resource according to the object identifier.

[0310] In this embodiment, the expected revenue value for each candidate virtual gift resource is accurately calculated based on the membership activation probability, the resource consumption value corresponding to the membership type object, and the resource value contained in each candidate virtual gift resource. This expected revenue value is then used as a condition for issuing the virtual gift resource. Virtual gift resources whose expected revenue value meets the issuance condition are selected and issued to the corresponding user, enabling personalized push of virtual gift resources.

[0311] For the specific definition of the operation data processing device, please refer to the definition of the operation data processing method above and will not be repeated here. Each module in the above-mentioned operation data processing device can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each of the above modules.

[0312] In one embodiment, a computer device is provided, which may be a terminal or a server. In this embodiment, a terminal is taken as an example, and its internal structure diagram may be as follows: Figure 10 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, an operation data processing method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad provided on the computer device housing, or an external keyboard, touchpad or mouse.

[0313] Those skilled in the art will understand that Figure 10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0314] In one embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0315] In one embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the steps in the above-mentioned method embodiments when executed by a processor.

[0316] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps of each of the above-described method embodiments.

[0317] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0318] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0319] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A method for processing operation data, characterized in that: The method comprises: Obtaining object operation data of the object identifier for the candidate resource tag, and obtaining resource operation data obtained by the operation on each target resource tag; the candidate resource tags include the target resource tag; Determining an object operation frequency of the object identifier based on the object operation data, and determining a tag operation frequency of each target resource tag being operated based on the resource operation data; The inverse of the object operation frequency is used as the object weight of the object identifier, and the inverse of each tag operation frequency is used as the tag weight of the corresponding target resource tag; determining a target operation frequency of the object identifier on each target resource tag of the target resource according to target operation data obtained by operating the object identifier on each target resource tag of the target resource; The attention degree of the object identifier to the target resource is determined based on the object weight, the weights of each tag, and the frequency of each target operation.

2. The method according to claim 1, characterized in that The step of obtaining object operation data of the object identifier for the candidate resource tag and obtaining resource operation data obtained by performing an operation on each target resource tag includes: Obtain object operation data for candidate resource tags of different versions for object identifiers, and obtain resource operation data obtained by operations on each target resource tag of different versions; different versions correspond to different timestamps; The determining the object operation frequency of the object identifier based on the object operation data, and determining the tag operation frequency of each target resource tag being operated based on the resource operation data, includes: determining an object operation frequency corresponding to the object identifier based on different versions of object operation data; The tag operation frequency of each target resource tag being operated is determined based on different versions of resource operation data.

3. The method according to claim 1, characterized in that The determining, based on the object weight, the weights of each tag, and the frequency of each target operation, the attention degree of the object identifier to the target resource includes: Determining, according to the object weight, the weight of each tag, and the frequency of each target operation, an operation evaluation value of the object identifier for each target resource tag; The attention degree of the object identifier to the target resource is determined based on the time decay factor and each of the operation evaluation values.

4. The method according to claim 3, characterized in that Determining the operation evaluation value of each target resource tag by the object identifier according to the object weight, each tag weight, and each target operation frequency includes: Determining a candidate operation frequency corresponding to each target resource tag of each candidate resource by the object identifier, the candidate resources including the target resource; Determining a sub-evaluation value of each target resource tag for a corresponding candidate resource based on the object weight, each tag weight, and the candidate operation frequency; Determining an operation evaluation value of the object identifier on each target resource tag based on each of the sub-evaluation values; The determining, based on the time decay factor and each of the operation evaluation values, the degree of attention of the object identifier to the target resource includes: A target operation evaluation value that meets the evaluation conditions is screened out from the operation evaluation values, and the attention degree of the object identifier to the target resource is determined based on the time decay factor and the target operation evaluation value.

5. The method according to claim 1, wherein The method further comprises: Determining the valid operation times of the object identifier for each target resource tag from the object operation data acquired within a preset time period; Determining a stable evaluation value of the object identifier operating each of the target resource tags based on each of the valid operation times and the preset duration; The determining, based on the object weight, the weights of each tag, and the frequency of each target operation, the attention degree of the object identifier to the target resource includes: The attention degree of the object identifier to the target resource is determined based on the object weight, the weights of each tag, the frequency of each target operation and the stability evaluation value.

6. The method according to claim 5, characterized in that Determining the valid operation count of the object identifier for each target resource tag from the object operation data acquired within the preset time period includes: Dividing the preset duration into a plurality of time periods through a preset time window, and determining the number of valid operations of the object identifier in the object operation data for each target resource tag in each time period; The determining, based on each of the valid operation times and the preset duration, a stable evaluation value of the object identifier operating each of the target resource tags includes: Filtering the target number of operations and the corresponding target time period that meet the valid operation conditions from the valid operation numbers corresponding to the time periods; A stable evaluation value of the object identifier operating on each target resource tag is determined according to the target number of operations and the target time period.

7. The method according to claim 5, characterized in that Determining the valid operation count of the object identifier for each target resource tag from the object operation data acquired within the preset time period includes: Determining the number of valid operations of the object identifier for each target resource tag under each promotion channel from the object operation data of different promotion channels acquired within a preset time period; The determining, based on each of the valid operation times and the preset duration, a stable evaluation value of the object identifier operating each of the target resource tags includes: For each promotion channel, determining a stable evaluation value of the object identifier operating each target resource tag under the corresponding promotion channel according to the number of valid operations and the preset duration corresponding to the corresponding promotion channel; The determining, based on the object weight, the weights of each tag, the frequency of each target operation, and the stability evaluation value, the attention degree of the object identifier to the target resource includes: Based on the object weight corresponding to each promotion channel, the weight of each tag, the frequency of each target operation and the stability evaluation value, the attention degree of the object identifier to the target resource is determined.

8. The method according to any one of claims 1 to 7, characterized in that The target resource is promotion information; the method further includes: Based on the attention level of each object identifier to the promotion information, target object identifiers whose attention levels meet the push conditions are selected; Push associated information related to the promotion information to the target object identifier.

9. The method according to any one of claims 1 to 7, characterized in that The target resource is a member-type object; the method further includes: Determining a membership activation probability of the object identifier for the membership type object based on the attention degree of the object identifier to the membership type object; According to the membership activation probability and the resource consumption value corresponding to the membership type object, a target virtual gift resource is selected from the candidate virtual gift resources and distributed to the object identifier.

10. The method according to claim 9, characterized in that The selecting a target virtual gift resource from candidate virtual gift resources and distributing it to the object identifier according to the membership activation probability and the resource consumption value corresponding to the membership type object includes: Determine the expected benefit value corresponding to each candidate virtual gift resource according to the membership activation probability, the consumption resource value corresponding to the membership type object, and the resource value contained in each candidate virtual gift resource; A target virtual gift resource that makes the expected benefit value meet a distribution condition is selected from the candidate virtual gift resources, and the target virtual gift resource is distributed according to the object identifier.

11. An operating data processing device, characterized in that: The device comprises: An acquisition module, configured to acquire object operation data of an object identifier for a candidate resource tag, and to acquire resource operation data obtained by an operation on each target resource tag; the candidate resource tags include the target resource tag; a frequency determination module, configured to determine an object operation frequency of the object identifier based on the object operation data, and determine a tag operation frequency of each target resource tag being operated based on the resource operation data; a weight determination module, configured to use the reciprocal of the object operation frequency as the object weight of the object identifier, and use the reciprocal of each tag operation frequency as the tag weight of the corresponding target resource tag; an operation module, configured to determine a target operation frequency of the object identifier on each target resource tag of the target resource according to target operation data obtained by operating the object identifier on each target resource tag of the target resource; The operation data processing module is used to determine the attention degree of the object identifier to the target resource based on the object weight, the weight of each tag and the frequency of each target operation.

12. The device according to claim 11, characterized in that The acquisition module is further used to acquire object operation data of the object identifier for candidate resource tags of different versions, and to acquire resource operation data obtained by operations performed on each target resource tag of different versions; different versions correspond to different timestamps; The frequency determination module is further configured to determine the object operation frequency corresponding to the object identifier based on different versions of object operation data; and determine the tag operation frequency of each target resource tag being operated based on different versions of resource operation data.

13. The device according to claim 11, characterized in that The operation data processing module is also used to determine the operation evaluation value of the object identifier for each target resource tag based on the object weight, each tag weight and each target operation frequency; and determine the attention of the object identifier to the target resource based on the time decay factor and each operation evaluation value.

14. The device according to claim 13, characterized in that The operation data processing module is also used to determine the candidate operation frequency corresponding to each target resource tag of each candidate resource for the object identifier, where the candidate resources include the target resource; determine the sub-evaluation value of each target resource tag for the corresponding candidate resource based on the object weight, the weight of each tag, and the candidate operation frequency; determine the operation evaluation value of the object identifier for each target resource tag based on each sub-evaluation value; filter out the target operation evaluation value that meets the evaluation conditions from each operation evaluation value, and determine the attention of the object identifier to the target resource based on the time attenuation factor and the target operation evaluation value.

15. The device according to claim 11, characterized in that The device further comprises: An evaluation module is configured to determine, from the object operation data acquired within a preset time period, the number of valid operations of the object identifier for each target resource tag; and determine a stable evaluation value of the object identifier for operating each target resource tag based on each valid operation number and the preset time period; The operation data processing module is further configured to determine the degree of attention of the object identifier to the target resource based on the object weight, each tag weight, each target operation frequency and each stability evaluation value.

16. The device according to claim 15, characterized in that The evaluation module is further configured to divide the preset duration into a plurality of time periods through a preset time window, and determine the number of valid operations of the object identifier in the object operation data for each target resource tag in each time period; From the valid operation times corresponding to each of the time periods, the target operation times and the corresponding target time periods that meet the valid operation conditions are screened out; based on the target operation times and the target time periods, the stable evaluation value of the object identifier operating on each of the target resource tags is determined.

17. The device according to claim 15, characterized in that The evaluation module is further configured to determine, from object operation data of different promotion channels acquired within a preset time period, the number of valid operations of the object identifier on each target resource tag under each promotion channel; and for each promotion channel, determine, based on the corresponding number of valid operations and the preset time period, a stable evaluation value of the object identifier's operation on each target resource tag under the corresponding promotion channel. The operation data processing module is further used to determine the attention of the object identifier to the target resource based on the object weight corresponding to each promotion channel, the weight of each tag, the frequency of each target operation and the stability evaluation value.

18. The device according to any one of claims 11 to 17, characterized in that The target resource is promotion information; the device further includes: The promotion module is configured to screen out target object identifiers whose attention levels meet push conditions based on the attention levels of each object identifier to the promotion information; and push associated information related to the promotion information to the target object identifiers.

19. The device according to any one of claims 11 to 17, characterized in that The target resource is a member-type object; the device further includes: The distribution module is used to determine the membership activation probability of the object identifier for the membership type object based on the attention of the object identifier to the membership type object; according to the membership activation probability and the consumption resource value corresponding to the membership type object, select the target virtual gift resource from the candidate virtual gift resources and distribute it to the object identifier.

20. The device according to claim 19, characterized in that The distribution module is also used to determine the expected benefit value corresponding to each candidate virtual gift resource based on the member activation probability, the consumption resource value corresponding to the member type object, and the resource value contained in each candidate virtual gift resource; from the candidate virtual gift resources, select the target virtual gift resource that makes the expected benefit value meet the distribution conditions, and distribute the target virtual gift resource according to the object identifier.

21. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: 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 storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

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

Images