Data processing method and device, electronic equipment and storage medium

Abstract

The invention discloses a data processing method and device, electronic equipment and a storage medium, and relates to the technical field of computers. A specific embodiment of the method comprises the steps of receiving a data processing instruction, collecting behavior information of a user within a preset time, and extracting behavior parameters and behavior objects in the behavior information to generate corresponding time sequence features; determining a behavior period and a behavior object of the user based on the time sequence characteristics to generate a corresponding recommendation task; in response to triggering of the recommendation task, obtaining corresponding service information based on the behavior object, extracting a to-be-recommended service parameter of the user, querying a business parameter of the behavior object, and adjusting the to-be-recommended service parameter to obtain a recommendation service of the user; and obtaining account information of the user, and sending the recommendation service to the account information. According to the embodiment, the problem that the accuracy of the preferential service is reduced due to the fact that the personalized preferential service is provided for the user through the preset preferential strategy and cannot be adapted and adjusted in time can be solved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a data processing method, device, electronic equipment and storage medium. Background Art

[0002] In the field of e-commerce, users can choose the items they need through e-commerce platforms. E-commerce platforms can also recommend items based on the different needs of different users, and can also provide users with preferential services so that users can choose corresponding items. In related technologies, preferential policies can usually be set in advance to provide users with personalized preferential services. However, preferential policies are set in advance and cannot be adjusted in time to market changes and user needs, thereby reducing the accuracy of preferential services. Summary of the invention

[0003] In view of this, an embodiment of the present invention provides a data processing method, device, electronic device and storage medium, which can solve the problem that personalized preferential services are provided to users through pre-set preferential policies, but cannot be adjusted in time, resulting in reduced accuracy of preferential services.

[0004] To achieve the above objective, according to one aspect of an embodiment of the present invention, a method for data processing is provided.

[0005] A data processing method according to an embodiment of the present invention includes: receiving a data processing instruction, obtaining corresponding user information, collecting user behavior information within a preset time, and extracting behavior parameters and behavior objects from the behavior information; extracting each preset parameter of the corresponding user behavior from the behavior parameters to convert them into a behavior label, clustering the behavior information based on the behavior label and the behavior object to obtain each behavior information set, so as to call a preset generation model to generate corresponding time series features; determining the user's behavior cycle and behavior object based on the time series features to generate a corresponding recommendation task; in response to the recommendation task being triggered, obtaining corresponding service information based on the behavior object, extracting the user's service parameters to be recommended, querying the business parameters of the behavior object, calling a preset adjustment strategy to adjust the service parameters to be recommended, and obtaining the recommended service for the user; and obtaining the user's account information to send the recommended service to the account information.

[0006] In one embodiment, the behavior cycle and behavior object of the user are determined based on the timing characteristics to generate a corresponding recommended task, including: extracting the behavior cycle and behavior object of the first preset behavior, and the behavior cycle and behavior object of each second preset behavior associated with the first preset behavior from the timing characteristics; for each second preset behavior, calculating the behavior correlation and object correlation with the first preset behavior according to the corresponding behavior cycle and behavior object to determine the total correlation, and generating the corresponding recommended task in response to the total correlation being greater than a preset correlation threshold.

[0007] In another embodiment, calling the preset generation model to generate corresponding timing features includes: calling the preset generation model, obtaining preset timing parameters to determine the corresponding timing period within the preset time, extracting the behavior parameters and behavior objects corresponding to each timing period in each behavior information set, and generating corresponding timing features.

[0008] In another embodiment, the obtaining of preset timing parameters to determine the corresponding timing period within the preset time includes: obtaining the timing type in the timing parameters to query the corresponding time node within the preset time period, and obtaining the timing duration of the timing type to determine the corresponding timing period based on the time node and the timing duration.

[0009] In another embodiment, the behavior cycle and behavior object of the user are determined based on the timing characteristics to generate a corresponding recommendation task, including: obtaining the historical behavior information of the user to determine the corresponding historical cycle from the historical behavior information based on the timing cycle to generate the historical timing characteristics of the historical cycle, and determine the target timing characteristics associated with the timing characteristics; obtaining the behavior object corresponding to the target timing characteristics to generate a corresponding recommendation task in combination with the timing cycle of the timing cycle.

[0010] In another embodiment, corresponding service information is obtained based on the behavior object, and service parameters to be recommended for the user are extracted, including: obtaining the item type to which the behavior object belongs, and querying the service information associated with the item type; obtaining the user's account information to filter service parameters from the service information to determine them as parameters to be recommended.

[0011] In another embodiment, a preset adjustment strategy is called to adjust the service parameters to be recommended, including: querying the corresponding adjustment strategy based on the item type, identifying the business parameters in the adjustment strategy, and obtaining the parameter value of the business parameter corresponding to the item type, so as to adjust the service parameters to be recommended according to the parameter value.

[0012] In another embodiment, after sending the recommended service to the account information, the method further includes: obtaining a usage status of the recommended service, calculating a corresponding usage rate, and adjusting the parameters of the service to be recommended based on the usage rate.

[0013] To achieve the above objective, according to another aspect of an embodiment of the present invention, a data processing device is provided.

[0014] A data processing device according to an embodiment of the present invention includes: a collection unit, which is used to receive data processing instructions, obtain corresponding user information, and collect user behavior information within a preset time; a generation unit, which is used to extract each preset parameter of the corresponding user behavior from the behavior parameters to convert it into a behavior label, cluster the behavior information based on the behavior label and the behavior object to obtain each behavior information set, extract the behavior parameters and behavior objects in the behavior information to call a preset generation model to generate corresponding time series features; the generation unit is used to determine the user's behavior cycle and behavior object based on the time series features to generate a corresponding recommendation task; an adjustment unit, which is used to respond to the recommendation task being triggered, obtain corresponding service information based on the behavior object, extract the user's service parameters to be recommended, query the business parameters of the behavior object, call a preset adjustment strategy to adjust the service parameters to be recommended, and obtain the recommended service for the user; a sending unit, which is used to obtain the user's account information to send the recommended service to the account information.

[0015] In one embodiment, the generation unit is specifically used to: extract the behavior cycle and behavior object of the first preset behavior, as well as the behavior cycle and behavior object of each second preset behavior associated with the first preset behavior from the timing features; for each second preset behavior, calculate the behavior correlation and object correlation with the first preset behavior according to the corresponding behavior cycle and behavior object to determine the total correlation, and generate a corresponding recommendation task in response to the total correlation being greater than a preset correlation threshold.

[0016] In another embodiment, the generation unit is specifically used to: call a preset generation model, obtain preset timing parameters to determine the corresponding timing period within the preset time, extract the behavior parameters and behavior objects corresponding to each timing period in each behavior information set, and generate corresponding timing features.

[0017] In another embodiment, the generating unit is specifically used to: obtain the timing type in the timing parameters to query the corresponding time node within the preset time period, obtain the timing duration of the timing type to determine the corresponding timing period based on the time node and the timing duration.

[0018] In another embodiment, the generating unit is used to: obtain the historical behavior information of the user, determine the corresponding historical period from the historical behavior information based on the time period, generate the historical time series feature of the historical period, and determine the target time series feature associated with the time series feature; obtain the behavior object corresponding to the target time series feature, and generate the corresponding recommendation task in combination with the time period of the time period.

[0019] In another embodiment, the adjustment unit is used to: obtain the item type to which the behavior object belongs, and query the service information associated with the item type; obtain the account information of the user to filter service parameters from the service information to determine the parameters to be recommended.

[0020] In another embodiment, the adjustment unit is used to: query the corresponding adjustment strategy based on the item type, identify the business parameters in the adjustment strategy, obtain the parameter value of the business parameter corresponding to the item type, and adjust the recommended service parameter according to the parameter value.

[0021] In yet another embodiment, the adjustment unit is used to: obtain a usage status of the recommended service, calculate a corresponding usage rate, and adjust the parameters of the service to be recommended based on the usage rate.

[0022] To achieve the above objective, according to another aspect of an embodiment of the present invention, an electronic device is provided.

[0023] An electronic device according to an embodiment of the present invention includes: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement the data processing method provided by the embodiment of the present invention.

[0024] To achieve the above objective, according to another aspect of an embodiment of the present invention, a computer readable medium is provided.

[0025] A computer-readable medium according to an embodiment of the present invention stores a computer program, and when the program is executed by a processor, the data processing method provided by the embodiment of the present invention is implemented.

[0026] To achieve the above objective, according to another aspect of an embodiment of the present invention, a computer program product is provided.

[0027] A computer program product according to an embodiment of the present invention includes a computer program, and when the program is executed by a processor, the data processing method provided by the embodiment of the present invention is implemented.

[0028] An embodiment of the above invention has the following advantages or beneficial effects: In the embodiment of the present invention, the behavior information of the user within a preset time period can be collected first to perform time series analysis to obtain the corresponding time series characteristics; the user's behavior rules and corresponding behavior objects can be determined through the time series characteristics, so the corresponding recommendation tasks can be generated based on this; after the recommendation task is triggered, the corresponding service information can be obtained based on the behavior object to extract the recommended service parameters applicable to the user, and the business parameters of the behavior object can also be queried to adjust the recommended service parameters to obtain the user's recommended service, so as to send the user's account information to achieve service recommendations for the user. In this way, in the embodiment of the present invention, when recommending services to users, the time series characteristics of the user's behavior can be determined first, so that the recommendation task can be determined according to the time series rules and behavior objects. When executing the recommendation task, after determining the recommended service parameters, they can be adjusted according to the implemented business parameters, so that the recommendation service can meet the personalized needs of users and real-time market changes, and improve the accuracy of service recommendations.

[0029] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention. Figure 1 is a schematic diagram of a main flow of a data processing method according to an embodiment of the present invention; Figure 2 is a schematic diagram of another main flow of a data processing method according to an embodiment of the present invention; Figure 3 is a schematic diagram of another main flow of a data processing method according to an embodiment of the present invention; Figure 4 is a schematic diagram of main units of a data processing apparatus according to an embodiment of the present invention; Figure 5 is an exemplary system architecture diagram to which embodiments of the present invention may be applied; Figure 6 It is a schematic diagram of the structure of a computer system suitable for implementing the embodiment of the present invention. DETAILED DESCRIPTION

[0031] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0032] It should be pointed out that the embodiments of the present invention and the features in the embodiments can be combined with each other without conflict. The acquisition, transmission, storage, use, and processing of data in the technical solution of this application are in compliance with the relevant provisions of national laws and regulations. In the embodiments of this application, some existing solutions in the industry such as certain software, components, and models may be mentioned, which should be considered as exemplary. Their purpose is only to illustrate the feasibility of the implementation of the technical solution of this application, but it does not mean that the applicant has or will necessarily use the solution.

[0033] An embodiment of the present invention provides a data processing system, which can be used in a scenario of recommending services to users in the e-commerce field, and specifically can be used in a scenario of recommending preferential services to users, such as a scenario of recommending coupon services.

[0034] The embodiment of the present invention provides a data processing method, which can be executed by a data processing system. Figure 1 As shown, the method includes the following steps.

[0035] S101: receiving a data processing instruction, obtaining corresponding user information, collecting user behavior information within a preset time, and extracting behavior parameters and behavior objects from the behavior information.

[0036] The data processing instructions may be sent by other systems, or may be automatically generated by the data processing system, such as periodically generating data processing instructions. The data processing instructions represent data processing for user service recommendations, which may include user information, such as user identification, user account information, etc. In the embodiment of the present invention, service recommendations may include multiple types, such as preferential service recommendations, item selection service recommendations, etc. Preferential service recommendations may specifically be coupon recommendations.

[0037] The preset time can be set according to the actual scenario, for example, within one month, three months, etc. In the field of e-commerce, user behavior information can be information about the user's operation behavior on the client page, such as click, browse, search and purchase behavior, etc. In this step, high-precision data acquisition technology can be used to achieve high-precision, high-speed and high-reliability acquisition of user behavior through sensors, signal processing and data transmission technology, and the collected data can be sent to the data processing system. The behavior parameters can be specifically the user's behavior operation parameters, such as the location, time, frequency of clicks, etc., and the behavior object can be the object of the behavior operation, such as the item information corresponding to the behavior operation.

[0038] S102: extracting preset parameters corresponding to user behaviors from the behavior parameters to convert them into behavior labels, clustering the behavior information based on the behavior labels and behavior objects to obtain sets of behavior information, and calling a preset generation model to generate corresponding time series features.

[0039] Among them, different user behaviors can be represented by corresponding labels. In this step, label conversion can be performed according to the parameter value of the preset parameter corresponding to each user behavior, that is, the behavior label of each user behavior is determined. The preset parameters can be set according to needs or scenarios, for example, they may include behavior type, behavior corresponding position, behavior corresponding page area, etc. In the embodiment of the present invention, codes corresponding to different parameter values ​​of each preset parameter can be set, and then the corresponding labels can be obtained by splicing the obtained codes. After obtaining the behavior label, the behavior information can be clustered in combination with the behavior object. The clustering method can be unlimited. After clustering, the behavior information set corresponding to each behavior can be obtained, and then the corresponding time series features can be generated.

[0040] It should be noted that, in the embodiment of the present invention, after clustering, the behavior information set with less data may be filtered to eliminate the user's erroneous operation behavior information or meaningless behavior information.

[0041] In the embodiment of the present invention, the temporal regularity of user behavior is determined by performing temporal analysis on the behavior information. Specifically, this step can generate the temporal features by: calling a preset generation model, obtaining preset temporal parameters to determine the corresponding temporal period within a preset time, extracting the behavior parameters and behavior objects corresponding to each temporal period in each behavior information set, and generating the corresponding temporal features.

[0042] The computational model may be pre-configured, for example, it may be an advanced time series analysis technique such as a Dynamic Time Warping (DTW) algorithm, etc.

[0043] Since user behavior usually has a temporal pattern within a period of time, the timing parameters can be preset to determine the timing period for the timing analysis. For example, the timing parameters can be weeks, months, holiday weeks, etc., so the specific time within the preset time can be determined to divide multiple timing periods, such as determining each week, each month, and each holiday week within the preset time as a timing period, and then the corresponding timing characteristics.

[0044] Specifically, in an embodiment of the present invention, the timing parameters may be specifically a timing type, which indicates the type of the timing period. Therefore, the timing type in the timing parameters may be obtained to query the corresponding time node within a preset time period, and the timing duration of the timing type may be obtained to determine the corresponding timing period based on the time node and the timing duration.

[0045] For example, the time series types may include weeks, months, holidays, etc., and a corresponding time series duration is set for each time series type. For example, the time series duration of a week is one week, the time series duration of a month is one month, and the time series duration of a holiday can be three days, one week, or one month.

[0046] It should be noted that in the embodiments of the present invention, after collecting the user's behavior information, it can be pre-processed to perform data cleaning and remove invalid or erroneous data points, such as filtering out abnormal MAC addresses or IMEI numbers, and then the data can be formatted and normalized for subsequent analysis. For example, the data processing system can automatically detect and exclude repeated or obviously abnormal behavior information, such as non-human click operation information.

[0047] S103: Determine the user's behavior cycle and behavior object based on the time series features to generate corresponding recommended tasks.

[0048] Among them, through the user's time series characteristics, the user's behavior pattern can be seen, and the user's behavior cycle and behavior object can be determined. The behavior object can represent the user's behavior operation object within the behavior cycle. Taking preferential service recommendation as an example, the purchase behavior usually has a greater impact on the preferential service recommendation, so the behavior object in this step can be the behavior object corresponding to the purchase behavior in the user behavior. The recommendation task represents the task of recommending services to the user, which may include user information, behavior cycle and behavior object. In this step, a task template can be set to generate a recommendation task.

[0049] Specifically, on the one hand, this step can be executed as follows: extracting the behavior cycle and behavior object of the first preset behavior, as well as the behavior cycle and behavior object of each second preset behavior associated with the first preset behavior from the time series features; for each second preset behavior, calculating the behavior correlation and object correlation with the first preset behavior according to the corresponding behavior cycle and behavior object to determine the total correlation, and generating a corresponding recommendation task in response to the total correlation being greater than a preset correlation threshold.

[0050] The first preset behavior and the second preset behavior can be set according to the scenario. For example, in the scenario of preferential service recommendation, the first preset behavior can be an ordering behavior, and the second preset behavior can include browsing, viewing, clicking, etc. In this step, the association between each second preset behavior and the first preset behavior can be analyzed to determine closely related behaviors, and then generate a recommendation task.

[0051] The behavior cycle can be determined according to the corresponding time sequence of the behavior, and specifically can be a time node, time interval, frequency of occurrence, etc. occurring within a preset time. The behavior object represents the operation object of the behavior, such as an operation object, etc. For each second preset behavior, its behavior correlation with the first preset behavior can be calculated according to the corresponding behavior cycle and the behavior information of the first preset behavior, and its behavior correlation and object correlation with the first preset behavior can be calculated according to the corresponding behavior object and the behavior object of the first preset behavior.

[0052] In this step, the matching degree between the behavior cycle of the first preset behavior and the behavior cycle of the second preset behavior can be calculated to determine the behavior association degree; and the similarity between the behavior object of the first preset behavior and the behavior object of the second preset behavior can be calculated to determine the object association degree. The total association degree can be a weighted average between the behavior association degree and the object association degree. If the total association degree is greater than the preset association degree threshold, it means that the second preset behavior is closely associated with the first preset behavior, so the recommended task corresponding to the user can be generated according to the behavior cycle and behavior object of the second preset behavior.

[0053] On the other hand, this step can be executed as follows: obtaining the user's historical behavior information, determining the corresponding historical period from the historical behavior information based on a preset time series period, generating the historical time series features of the historical period, and determining the target time series features associated with the time series features; obtaining the behavior object corresponding to the target time series features, and generating the corresponding recommendation task in combination with the time series period.

[0054] The historical behavior information can be obtained in advance, so the historical period corresponding to the time series period can be obtained from the historical behavior information according to the time series period, and then the historical time series features of the historical period are generated according to the historical behavior information in this historical period. Then, by calculating the correlation between the historical time series features and the time series features, if the correlation is close, it can be considered that the two form a law, so the corresponding recommendation task can be generated for service recommendation.

[0055] It should be noted that, in the embodiment of the present invention, some time series features can determine multiple corresponding historical periods, and one or more historical periods that are closer to the time series period can be selected, and then the association between the corresponding historical time series features and the time series features can be calculated respectively. The method for calculating the association in this step can be a dynamic time warping (DTW) algorithm.

[0056] The DTW algorithm is implemented by constructing a cost matrix, where each element of the matrix represents the distance between corresponding points in the two sequences (usually using the Euclidean distance). Through dynamic programming techniques, DTW calculates an optimal matching path that minimizes the cumulative distance of the entire sequence. For example, suppose there are two time series, one is the historical data of users buying gifts before a specific holiday, and the other is the user's purchase behavior before the subsequent holiday. The DTW algorithm can detect that even if the two sequences are not completely corresponding in time, the pattern of purchase behavior is similar. The implementation steps of the DTW algorithm may include: constructing a cost matrix: calculating the distance corresponding to each point in the two sequences and constructing a cost matrix; calculating the cumulative cost: constructing a cumulative cost matrix by accumulating the values ​​in the cost matrix to record the minimum cumulative distance to each point; finding the optimal path: starting from the end point of the cumulative cost matrix, backtracking to find the path with the minimum cumulative cost, that is, the optimal alignment path; calculating the DTW distance: the cumulative cost of the optimal path is the DTW distance between the two sequences.

[0057] S104: In response to the recommendation task being triggered, the corresponding service information is obtained based on the behavior object, the user's service parameters to be recommended are extracted, the business parameters of the behavior object are queried, and the preset adjustment strategy is called to adjust the service parameters to be recommended to obtain the user's recommended service.

[0058] Among them, the trigger condition can be set in the recommendation task, and can be set according to the behavior cycle, for example, it can be set to the starting time point of the next corresponding timing cycle. The behavior object can represent the item information that the user is interested in. In step S102, the behavior object of the recommendation task is determined, so the corresponding service information can be determined, and then the corresponding service parameters can be extracted to determine the service parameters to be recommended, and then the service parameters to be recommended are adjusted through the business parameters to obtain the user's recommended service.

[0059] Taking the example of service recommendation as a discount recommendation, the behavior object in this step can be an operation item, so the corresponding discount information can be determined, such as coupons, discount prices, etc. Therefore, this step can be specifically performed as follows: obtain the item type to which the behavior object belongs, query the service information associated with the item type; obtain the user's account information to filter the service parameters from the service information to determine the parameters to be recommended. The service information associated with the item type can specifically be the discount information applied to the item type. Since some discount information is associated with account information, such as account level, etc., the service parameters can be filtered based on the account information to determine the discount information applicable to the account information, that is, the parameters to be recommended.

[0060] It should be noted that, in the embodiment of the present invention, the determination of the data to be recommended can also be implemented according to a deep learning model, such as a deep learning model that may include a convolutional neural network (CNN) or a recurrent neural network (RNN) to process large-scale user behavior data and item feature data to generate personalized preferential strategies. The deep learning model is trained with historical user behavior data, and the model parameters are optimized through cross-validation and grid search techniques to improve the accuracy and generalization ability of the model. After the training is completed, the model can predict the probability of a user's response to different services based on the user's behavior characteristics, and then the service parameters of the service that the user is most likely to respond to can be selected according to the prediction results of the model to determine as the data to be recommended.

[0061] Furthermore, after determining the parameters to be recommended, they can be adjusted based on current market information and / or user information. Specifically, the following can be performed: querying the corresponding adjustment strategy based on the item type, identifying the business parameters in the adjustment strategy, obtaining the parameter value of the business parameter corresponding to the item type, and adjusting the service parameters to be recommended according to the parameter value.

[0062] The adjustment strategy is pre-set, and different adjustment strategies can be set for different item types. Since the service is usually adjusted based on current market information and / or user information, the business parameters required for adjustment can be set in the adjustment strategy. After determining the adjustment strategy, the business parameters in the adjustment strategy can be identified to obtain the parameter values ​​corresponding to the item type, and then make adjustments. For example, the business parameters can be specifically inventory, sales volume, etc. The adjustment method can be to reduce the intensity of preferential services when the inventory is small, and to increase the intensity of preferential services when the inventory is large.

[0063] For example, taking preferential services including discounts and promotions as an example, assuming that market analysis shows that there is excess inventory of a certain item and the preferential strategy needs to be adjusted quickly to clear the inventory, the L-BFGS algorithm can be used to adjust the discount intensity and the scope of promotion activities while taking into account historical sales data, user purchasing behavior and market trends.

[0064] It should be noted that a dynamic preferential policy adjuster can be set in the embodiment of the present invention to adjust the preferential service, and the final preferential policy can be determined according to the personalized preferential policy, business objectives and market conditions. Use advanced optimization algorithms, such as gradient descent, L-BFGS or conjugate gradient, to automatically adjust the learning rate and step size to quickly converge to the optimal solution. For example, if market analysis shows that a certain item is in excess inventory, the discount intensity of the item can be increased and used as a preferential policy for priority push. Using the L-BFGS algorithm, the discount intensity and the scope of promotional activities can be automatically adjusted while considering historical sales data, user purchasing behavior and market trends to find the most effective preferential policy.

[0065] S105: Acquire the user's account information to send the recommended service to the account information.

[0066] After the recommended service for the user is determined, it can be sent to the account information so that the user can use the recommended service in subsequent actions. Specifically, the recommended service can be sent to the user via email, mobile push or in-APP notification.

[0067] In the embodiment of the present invention, after the recommended service is sent to the account information, the usage of the recommended service can also be monitored to calculate the parameters such as the usage rate of the recommended service, and then the parameters of the recommended service can be adjusted so that the recommended service is more suitable for the user when possible. Specifically, in the embodiment of the present invention, the impact of different preferential services on user behavior can also be evaluated through A / B testing, analyzing which method can bring higher user conversion rate, and adjusting the strategy accordingly.

[0068] It should be noted that in the embodiment of the present invention, the optimization results can also be fed back to the system to continuously improve the timing analysis and discount generation algorithms to form a closed-loop optimization mechanism. For example, according to the user's response to the discount, the parameters of the recommendation algorithm are adjusted to improve the accuracy of future personalized recommendations.

[0069] In the embodiment of the present invention, risk assessment and anti-fraud risk control strategies can also be constructed to identify abnormal behaviors and take preventive measures. Specifically, abnormal behavior patterns can be identified through deep learning algorithms, device fingerprint technology, user behavior analysis, etc., such as autoencoders or long short-term memory networks (LSTM). Device fingerprint technology can use device fingerprint technology to collect unique identification information of devices, such as IP addresses, hardware configurations, operating system versions, etc.; device fingerprints can identify fraudulent behaviors such as using simulators or tampering with device information. Behavioral analysis can analyze user behavior patterns, such as login frequency, transaction patterns, usage time, etc., to identify abnormal behaviors; and cluster analysis can be used to divide user behaviors into normal users and potential fraud users. Risk assessment can assign a risk score to each user based on the user's behavioral characteristics and historical data. Accounts with high risk scores will be subject to stricter review and monitoring, and user behavior and transaction activities will be monitored in real time. Once an abnormal pattern is detected, an alarm will be triggered immediately.

[0070] In an embodiment of the present invention, when providing service recommendations for users, the temporal characteristics of user behavior can be determined first, so that the recommendation task can be determined based on the temporal rules and behavior objects. When executing the recommendation task, after determining the service parameters to be recommended, they can be adjusted according to the implemented business parameters, so that the recommendation service can meet the user's personalized needs and real-time market changes, thereby improving the accuracy of service recommendations.

[0071] Combine the following Figure 1 The embodiment shown in the figure specifically describes the method for data processing in the embodiment of the present invention. Figure 2 As shown, the method includes the following steps.

[0072] S201: receiving a data processing instruction, obtaining corresponding user information, collecting user behavior information within a preset time, and extracting behavior parameters and behavior objects from the behavior information.

[0073] S202: Obtain preset timing parameters to determine corresponding timing periods within a preset time, and generate corresponding timing features based on behavior parameters and behavior objects corresponding to each timing period.

[0074] S203: Determine the user's behavior cycle and behavior object based on the time series features to generate corresponding recommended tasks.

[0075] S204: In response to the recommendation task being triggered, the item type to which the behavior object belongs is obtained, and the service information associated with the item type is queried; the user's account information is obtained to filter the service parameters from the service information to determine the parameters to be recommended.

[0076] S205: querying the corresponding adjustment strategy based on the item type, identifying the business parameters in the adjustment strategy, and obtaining parameter values ​​of the business parameters corresponding to the item type, so as to adjust the recommended service parameters according to the parameter values.

[0077] S206: Acquire the user's account information to send the recommended service to the account information.

[0078] It should be noted that the data processing principle in the embodiment of the present invention is the same as Figure 1 The corresponding data processing principles in the illustrated embodiments are the same and will not be described in detail here.

[0079] Combine the following Figure 1 The embodiment shown in the figure specifically describes the method for data processing in the embodiment of the present invention. Figure 3 As shown, the method includes the following steps.

[0080] S301: receiving a data processing instruction, obtaining corresponding user information, collecting user behavior information within a preset time, and extracting behavior parameters and behavior objects from the behavior information.

[0081] S302: Obtain the timing type in the timing parameters to query the corresponding time node within the preset time period, obtain the timing duration of the timing type, and determine the corresponding timing period based on the time node and the timing duration.

[0082] S303: Generate corresponding time series features based on the behavior parameters and behavior objects corresponding to each time series period.

[0083] S304: Acquire the historical behavior information of the user, determine the corresponding historical period from the historical behavior information based on the time series period, generate the historical time series feature of the historical period, and determine the target time series feature associated with the time series feature.

[0084] S305: Obtaining a behavior object corresponding to the target timing feature to generate a corresponding recommendation task in combination with the timing period of the timing period.

[0085] S306: In response to the recommendation task being triggered, the corresponding service information is obtained based on the behavior object, the user's service parameters to be recommended are extracted, the business parameters of the behavior object are queried, and the preset adjustment strategy is called to adjust the service parameters to be recommended to obtain the user's recommended service.

[0086] S307: Acquire the user's account information to send the recommended service to the account information.

[0087] S308: Obtain the usage status of the recommended service, calculate the corresponding usage rate, and adjust the parameters of the service to be recommended based on the usage rate.

[0088] It should be noted that the data processing principle in the embodiment of the present invention is the same as Figure 1 The corresponding data processing principles in the illustrated embodiments are the same and will not be described in detail here.

[0089] In order to solve the problems existing in the prior art, an embodiment of the present invention provides a data processing device 400, such as Figure 4 As shown, the device 400 includes: a collection unit 401, which is used to receive data processing instructions, obtain corresponding user information, and collect user behavior information within a preset time; a generation unit 402, which is used to extract various preset parameters corresponding to user behavior from the behavior parameters to convert them into behavior labels, cluster the behavior information based on the behavior labels and the behavior objects to obtain various behavior information sets, extract the behavior parameters and behavior objects in the behavior information to call a preset generation model to generate corresponding time series features; the generation unit 402 is used to determine the user's behavior cycle and behavior object based on the time series features to generate a corresponding recommendation task; an integration unit 403, which is used to respond to the recommendation task being triggered, obtain corresponding service information based on the behavior object, extract the user's service parameters to be recommended, query the business parameters of the behavior object, call a preset adjustment strategy to adjust the service parameters to be recommended, and obtain the recommended service for the user; a sending unit 404, which is used to obtain the user's account information to send the recommended service to the account information.

[0090] It should be understood that the manner of implementing the embodiments of the present invention is different from the Figure 1 The method of the illustrated embodiment is the same and will not be described again here.

[0091] In one embodiment, the generation unit 402 is specifically used to: extract the behavior cycle and behavior object of the first preset behavior, as well as the behavior cycle and behavior object of each second preset behavior associated with the first preset behavior from the time series feature; for each second preset behavior, calculate the behavior correlation and object correlation with the first preset behavior according to the corresponding behavior cycle and behavior object to determine the total correlation, and generate a corresponding recommendation task in response to the total correlation being greater than a preset correlation threshold.

[0092] In another embodiment, the generation unit 402 is specifically used to: call a preset generation model, obtain preset timing parameters to determine the corresponding timing period within the preset time, extract the behavior parameters and behavior objects corresponding to each timing period in each behavior information set, and generate corresponding timing features.

[0093] In another embodiment, the generation unit 402 is specifically used to: obtain the timing type in the timing parameters to query the corresponding time node within the preset time period, obtain the timing duration of the timing type, and determine the corresponding timing period based on the time node and the timing duration.

[0094] In another embodiment, the generation unit 402 is used to: obtain the historical behavior information of the user, to determine the corresponding historical period from the historical behavior information based on the time period, to generate the historical time series feature of the historical period, and to determine the target time series feature associated with the time series feature; to obtain the behavior object corresponding to the target time series feature, to generate the corresponding recommendation task in combination with the time period of the time period.

[0095] In another embodiment, the adjustment unit 403 is used to: obtain the item type to which the behavior object belongs, and query the service information associated with the item type; obtain the account information of the user to filter service parameters from the service information to determine the parameters to be recommended.

[0096] In another embodiment, the adjustment unit 403 is used to: query the corresponding adjustment strategy based on the item type, identify the business parameters in the adjustment strategy, obtain the parameter value of the business parameter corresponding to the item type, and adjust the recommended service parameter according to the parameter value.

[0097] It should be understood that the manner of implementing the embodiments of the present invention is different from the Figure 1 , 2 , and 3 are the same as the embodiments shown in FIG. 3 and will not be described again here.

[0098] In an embodiment of the present invention, when providing service recommendations for users, the temporal characteristics of user behavior can be determined first, so that the recommendation task can be determined based on the temporal rules and behavior objects. When executing the recommendation task, after determining the service parameters to be recommended, they can be adjusted according to the implemented business parameters, so that the recommendation service can meet the user's personalized needs and real-time market changes, thereby improving the accuracy of service recommendations.

[0099] According to an embodiment of the present invention, the embodiment of the present invention further provides an electronic device and a readable storage medium.

[0100] The electronic device of an embodiment of the present invention comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the one processor, and the instructions are executed by the at least one processor so that the at least one processor executes the data processing method provided by the embodiment of the present invention.

[0101] Figure 5 An exemplary system architecture 500 is shown to which the data processing method or data processing apparatus according to the embodiment of the present invention can be applied.

[0102] like Figure 5As shown, system architecture 500 may include terminal devices 501, 502, 503, a network 504 and a server 505. Network 504 is used to provide a medium for communication links between terminal devices 501, 502, 503 and server 505. Network 504 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0103] The user can use the terminal devices 501, 502, 503 to interact with the server 505 through the network 504 to receive or send messages, etc. Various client applications can be installed on the terminal devices 501, 502, 503.

[0104] The terminal devices 501 , 502 , and 503 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0105] The server 505 may be a server that provides various services. The server may analyze and process the received data such as the product information query request, and feed back the processing result (such as product information - just an example) to the terminal device.

[0106] It should be noted that the data processing method provided in the embodiment of the present invention is generally executed by the server 505 , and accordingly, the data processing device is generally set in the server 505 .

[0107] It should be understood that Figure 5 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0108] Reference below Figure 6 , which shows a schematic diagram of the structure of a computer system 600 suitable for implementing an embodiment of the present invention. Figure 6 The computer system shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0109] like Figure 6 As shown, the computer system 600 includes a processing unit (CPU) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 to a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the system 600 are also stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0110] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage section 608 as needed.

[0111] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the processing unit (CPU) 601, the above-mentioned functions defined in the system of the present invention are executed.

[0112] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0113] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a unit, a program segment, or a part of a code, and the above-mentioned unit, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0114] The units involved in the embodiments of the present invention may be implemented by software or hardware. The units described may also be set in a processor, for example, it may be described as: a processor includes a generating unit, an adjusting unit, and a sending unit. The names of these units do not constitute a limitation on the units themselves in some cases, for example, the generating unit may also be described as a "unit for generating functions".

[0115] As another aspect, the present invention further provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device executes the data processing method provided by the present invention.

[0116] As another aspect, the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the data processing method provided by an embodiment of the present invention.

[0117] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for data processing, characterized in that: include: Receive data processing instructions, obtain corresponding user information, collect user behavior information within a preset time, and extract behavior parameters and behavior objects from the behavior information; Extracting preset parameters corresponding to the user behavior from the behavior parameters to convert them into behavior labels, clustering the behavior information based on the behavior labels and the behavior objects to obtain sets of behavior information, so as to call a preset generation model to generate corresponding time series features; Determine the user's behavior cycle and behavior object based on the time series feature to generate a corresponding recommended task; In response to the recommendation task being triggered, corresponding service information is obtained based on the behavior object, the parameters of the service to be recommended for the user are extracted, the business parameters of the behavior object are queried, and a preset adjustment strategy is called to adjust the parameters of the service to be recommended, so as to obtain the recommended service for the user; The account information of the user is obtained to send the recommended service to the account information.

2. The method according to claim 1, characterized in that Determining the user's behavior cycle and behavior object based on the time series feature to generate a corresponding recommended task includes: Extracting a behavior period and a behavior object of a first preset behavior, and a behavior period and a behavior object of each second preset behavior associated with the first preset behavior from the time series feature; For each second preset behavior, the behavior association and object association with the first preset behavior are calculated according to the corresponding behavior cycle and behavior object to determine the total association. In response to the total association being greater than a preset association threshold, a corresponding recommended task is generated.

3. The method according to claim 1, characterized in that The calling of the preset generation model to generate corresponding time series features includes: The preset generation model is called, the preset timing parameters are obtained to determine the corresponding timing period within the preset time, the behavior parameters and behavior objects corresponding to each timing period in each behavior information set are extracted, and the corresponding timing features are generated.

4. The method according to claim 3, characterized in that The obtaining of preset timing parameters to determine the corresponding timing period within the preset time includes: The timing type in the timing parameters is obtained to query the corresponding time node within the preset time period, and the timing duration of the timing type is obtained to determine the corresponding timing period based on the time node and the timing duration.

5. The method according to claim 3, characterized in that: Determining the user's behavior cycle and behavior object based on the time series feature to generate a corresponding recommended task includes: Acquire the historical behavior information of the user, determine the corresponding historical period from the historical behavior information based on the time series period, generate the historical time series feature of the historical period, and determine the target time series feature associated with the time series feature; The behavior object corresponding to the target time series feature is obtained to generate a corresponding recommendation task in combination with the time series period.

6. The method according to claim 1, characterized in that Acquiring corresponding service information based on the behavior object and extracting the service parameters to be recommended for the user include: Obtain the item type to which the behavior object belongs, and query the service information associated with the item type; The user's account information is obtained to filter service parameters from the service information to determine the parameters to be recommended.

7. The method according to claim 6, characterized in that Calling a preset adjustment strategy to adjust the parameters of the service to be recommended includes: The adjustment strategy corresponding to the item type is queried, the business parameters in the adjustment strategy are identified, and the parameter values ​​of the business parameters corresponding to the item type are obtained, so as to adjust the service parameters to be recommended according to the parameter values.

8. A data processing device, characterized in that: include: A collection unit, used to receive data processing instructions, obtain corresponding user information, and collect user behavior information within a preset time; A generating unit, configured to extract preset parameters corresponding to the user behavior from the behavior parameters to convert them into behavior labels, cluster the behavior information based on the behavior labels and the behavior objects to obtain sets of behavior information, extract the behavior parameters and behavior objects in the behavior information to call a preset generating model to generate corresponding time series features; The generating unit is used to determine the behavior cycle and behavior object of the user based on the time series feature to generate a corresponding recommended task; an adjustment unit, configured to, in response to the recommendation task being triggered, obtain corresponding service information based on the behavior object, extract the service parameters to be recommended for the user, query the business parameters of the behavior object, call a preset adjustment strategy to adjust the service parameters to be recommended, and obtain the recommended service for the user; The sending unit is used to obtain the account information of the user to send the recommended service to the account information.

9. The device according to claim 8, characterized in that The generating unit is specifically used for: The preset generation model is called, the preset timing parameters are obtained to determine the corresponding timing period within the preset time, the behavior parameters and behavior objects corresponding to each timing period in each behavior information set are extracted, and the corresponding timing features are generated.

10. The device according to claim 9, characterized in that The generating unit is specifically used for: The timing type in the timing parameters is obtained to query the corresponding time node within the preset time period, and the timing duration of the timing type is obtained to determine the corresponding timing period based on the time node and the timing duration.

11. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

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

13. A computer program product, comprising a computer program, characterized in that When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images