User behavior pattern mining and recommendation method, device, electronic device, and medium

Abstract

This application relates to the technical fields of data processing and cloud technology, and discloses a method, device, electronic device, and medium for mining and recommending user behavior patterns. The method for mining user behavior patterns includes: obtaining multiple behavioral indicator data of a user, wherein the behavioral indicator data includes time information and the behavioral data indicator value corresponding to the time information; determining the characteristic value in each behavioral data indicator value based on the time series information corresponding to the multiple behavioral indicator data, and using the behavioral indicator data corresponding to each characteristic value as characteristic indicator data; and mining the user's user behavior pattern based on each characteristic indicator data. The solution provided in this application is used to obtain user behavior patterns through data calculation, thereby improving the recognition accuracy of user behavior patterns.

Description

Technical Field

[0001] The present application relates to the technical field of data processing and cloud technology. Specifically, the present application relates to a method, device, electronic device, and medium for mining and recommending user behavior patterns. Background Art

[0002] Identifying user behavior patterns plays a positive role in providing users with more timely and accurate server push notifications and guiding developers to improve service functions. Analyzing user behavior data, such as user usage data on applications, can provide a basis for mining user behavior patterns.

[0003] Existing technologies often directly use collected user behavior data to cluster similar user behaviors. However, due to the high noise content in user behavior data, such as errors, user behavior patterns derived using traditional methods cannot accurately represent user behavior characteristics based on noisy user behavior data. In other words, the resulting user behavior patterns are not precise enough. Summary of the Invention

[0004] The purpose of this application is to solve at least one of the above-mentioned technical deficiencies, and the following technical solutions are proposed to improve the accuracy of identifying user behavior patterns.

[0005] In one aspect, the present application provides a method for mining user behavior patterns, comprising:

[0006] Acquire multiple behavior indicator data of the user, wherein the behavior indicator data includes time information and behavior data indicator values ​​corresponding to the time information;

[0007] Determine, based on time series information corresponding to the plurality of behavior indicator data, a characteristic value in each behavior data indicator value, and use the behavior indicator data corresponding to each characteristic value as characteristic indicator data;

[0008] Based on the data of various characteristic indicators, the user behavior patterns of users are mined.

[0009] Another aspect of the present application provides a recommendation method based on user behavior patterns, comprising:

[0010] Acquire multiple behavior indicator data of the user, the multiple behavior indicator data including behavior indicator data of the user corresponding to at least two information categories of the application;

[0011] Dividing the plurality of behavior indicator data according to information categories to obtain a plurality of behavior indicator data corresponding to each information category;

[0012] For each information category, based on the time series information corresponding to the plurality of behavior indicator data of the information category, determine the characteristic values ​​in the indicator values ​​of the behavior data corresponding to the information category, and use the characteristic values ​​corresponding to the information category as the characteristic indicator data corresponding to the information category;

[0013] For each information category, mining the user behavior pattern corresponding to the information category based on the characteristic indicator data of the information category, the user behavior pattern includes a time series pattern;

[0014] Based on the user's usage time of the application, at least one timing pattern matching the usage time is determined, and based on the determined timing pattern and the usage time, relevant information of the information category corresponding to the determined timing pattern is recommended to the user.

[0015] Another aspect of the present application provides a device for mining user behavior patterns, the device comprising:

[0016] A behavior indicator data acquisition module is used to acquire multiple behavior indicator data of the user, wherein the behavior indicator data includes time information and behavior data indicator values ​​corresponding to the time information;

[0017] a characteristic indicator data determination module, configured to determine a characteristic value in each behavior data indicator value based on time series information corresponding to a plurality of behavior indicator data, and use the behavior indicator data corresponding to each characteristic value as the characteristic indicator data;

[0018] The user behavior pattern mining module is used to mine the user behavior pattern of users based on various feature indicator data.

[0019] In another aspect of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for mining user behavior patterns shown in the first aspect of the present application is implemented.

[0020] Another aspect of the present application further provides a recommendation device based on user behavior patterns, comprising:

[0021] A second behavior indicator data acquisition module is used to acquire multiple behavior indicator data of the user, the multiple behavior indicator data including behavior indicator data of the user corresponding to at least two information categories of the application;

[0022] A division module, for dividing the plurality of behavior indicator data according to information categories, and obtaining a plurality of behavior indicator data corresponding to each information category;

[0023] A second characteristic indicator data determination module is configured to determine, for each information category, characteristic values ​​among the behavior data indicator values ​​corresponding to the information category based on time series information corresponding to the plurality of behavior indicator data of the information category, and use the characteristic values ​​corresponding to the information category as characteristic indicator data corresponding to the information category;

[0024] A second user behavior pattern mining module is configured to mine, for each information category, a user behavior pattern of the user corresponding to the information category based on each characteristic indicator data of the information category, wherein the user behavior pattern includes a time series pattern;

[0025] The second recommendation module is used to determine at least one timing pattern that matches the usage time based on the user's usage time of the application, and recommend relevant information of the information category corresponding to the determined timing pattern to the user based on the determined timing pattern and the usage time.

[0026] Another aspect of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for mining user behavior patterns shown in the first aspect of the present application is implemented.

[0027] The beneficial effects of the technical solution provided by this application are:

[0028] The user behavior pattern mining method provided in the present application determines characteristic indicator data that can characterize user behavior characteristics from user behavior indicator data, and mines user behavior patterns based on the characteristic indicator data. Since the characteristic indicator data includes characteristic values ​​that characterize user behavior characteristics and time information corresponding to characteristic values ​​that can determine the periodic characteristics of user behavior, user behavior patterns can be accurately determined based on the characteristic indicator data. Moreover, compared with the method of directly using behavior indicator data to determine user behavior patterns, the method of using characteristic indicator data screens and eliminates the behavior indicator data, thereby reducing the amount of data processing for mining user behavior patterns.

[0029] Additional aspects and advantages of the present application will be given in part in the following description, which will become apparent from the following description, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0031] Figure 1 A flowchart of a method for mining user behavior patterns provided in one embodiment of the present application;

[0032] Figure 2A schematic diagram of user behavior indicator data provided by an embodiment of the present application;

[0033] Figure 3 A schematic diagram of user behavior indicator data provided by another embodiment of the present application, which focuses on segmenting user behavior indicator data according to a specified time period;

[0034] Figure 4 A schematic diagram of characteristic points provided in one embodiment of the present application;

[0035] Figure 5 A timing diagram of a method for mining user behavior patterns provided by an embodiment of the present application;

[0036] Figure 6 A schematic diagram of the structure of a user behavior pattern mining device provided in an embodiment of the present application;

[0037] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0038] The following describes embodiments of the present application in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application and are not to be construed as limiting the present application.

[0039] It will be understood by those skilled in the art that, unless expressly stated otherwise, the singular forms "a", "an", "said" and "the" used herein may also include the plural forms. It should be further understood that the term "comprising" used in the specification of this application refers to the presence of features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, "connected" or "coupled" as used herein may include wireless connections or wireless couplings. The term "and / or" used herein includes all or any units and all combinations of one or more associated listed items.

[0040] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with their meanings in the context of the prior art, and will not be interpreted in an idealized or overly formal sense unless specifically defined as in the specification.

[0041] User behavior analysis refers to the process of collecting statistics and analyzing basic data on users’ use of related products or execution of specified operations, in order to discover patterns in users’ use of related products or execution of specified operations. With the development of electronic technology, mining and mastering user behavior patterns can help companies to specify marketing strategies, improve product experience, and attract users to use the products, thereby increasing the number of users. However, in the prior art, it is impossible to obtain accurate user behavior patterns based on the basic data collected on user behavior. Although there are many ways to mine user behavior patterns in the prior art, the results are not ideal. In order to solve one or more problems existing in the existing solutions for mining user behavior patterns and better meet actual needs, the present application provides a method for mining user behavior patterns.

[0042] The solution provided in the embodiments of the present application can be executed by any electronic device, such as a terminal device or a server, wherein the server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. The terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited to this. The terminal and the server can be directly or indirectly connected via wired or wireless communication, and this application does not limit this. With respect to the technical problems existing in the prior art, the user behavior pattern mining method, device, electronic device and storage medium provided in this application are intended to solve at least one of the above technical problems in the prior art.

[0043] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0044] The present application embodiment provides a possible implementation method, such as Figure 1As shown, a flowchart of a method for mining user behavior patterns is provided. The solution can be executed by any electronic device, and optionally, it can be executed on the server side. For example, the solution of the embodiment of the present application can be executed by a terminal device or server that can communicate with an application server. Of course, it can also be executed by an application server. The execution subject mines user behavior patterns by obtaining data related to user behavior in the database of the application server. Figure 1 As shown in , the method may include the following steps:

[0045] Step S110, obtaining multiple behavior indicator data of the user, wherein the behavior indicator data includes time information and behavior data indicator values ​​corresponding to the time information;

[0046] Step S120, determining a characteristic value in each behavior data indicator value based on the time series information corresponding to the plurality of behavior indicator data, and using the behavior indicator data corresponding to each characteristic value as characteristic indicator data;

[0047] Step S130: mining the user behavior pattern of the user based on each characteristic indicator data.

[0048] The solution provided by the present application can be applicable to but not limited to the following scenarios: when studying the behavior pattern of at least one user towards a certain product and / or a certain operation, an electronic device, such as a server, receives an analysis request for the user behavior pattern from a technician (such as an analyst of the user behavior pattern), and the server responds to the analysis request, and the electronic device obtains the user's behavior indicator data stored in the background (such as: a server or database that stores the user's behavior indicator data). The user's behavior indicator data is indicator data that characterizes the user's behavior. After the server obtains the user's behavior indicator data, it determines the characteristic indicator data that can characterize the user's behavior characteristics based on the user's behavior indicator data, wherein the characteristic indicator data is characteristic data that characterizes the user's behavior pattern. The user's user behavior pattern is mined based on the characteristic indicator data to achieve the purpose of mining the user's behavior pattern based on the behavior indicator data. The user behavior pattern is then sent to the request sender.

[0049] The user's behavior indicator data can be retrieved from a database storing the user's behavior data. The user's behavior indicator data at least includes time information and the behavior data indicator value corresponding to the time information. That is, the behavior indicator data includes: each behavior data indicator value, and the time corresponding to each behavior data indicator value. For example, the behavior of swiping a bus card may have a corresponding behavior indicator data value that can be the number of times the card is swiped, and the time information corresponding to each behavior data indicator value is the swiping time. The time information corresponding to each behavior data indicator value can be a moment or a time period. If it is a moment, the behavior indicator data indicates that the user swiped the bus card at that moment. If it is a time period, the behavior indicator data indicates the number of times the user swiped the bus card during that time period.

[0050] Combine Figure 2 As shown, Figure 2 This is a schematic diagram of user behavior indicator data provided by an embodiment of the present application. The schematic diagram is presented in the form of a two-dimensional coordinate system. The horizontal axis of the two-dimensional coordinate system is time. The time unit can be days, months, years, hours, minutes, seconds, etc. Figure 2 The time unit displayed can be days, and the vertical axis of the two-dimensional coordinate system is the number of user payments. The schematic diagram shows how the indicator data values ​​in the user's behavioral indicator data change over time. The behavioral indicator data are the coordinate points on the schematic diagram. For example, the number of user payments on the second day is 3, and the corresponding number of user payments on the third day is 2. The two behavioral indicator data are connected by a straight line, and the trend of the straight line represents the trend between the two behavioral indicator data. For example, the trend of the straight line connecting the number of user payments on the second day and the number of payments on the third day represents a decrease in the number of payments.

[0051] The timing information of each behavior indicator data is determined based on the time information in the behavior indicator data. The timing information can characterize the temporal sequence of each behavior indicator data. The timing information corresponding to multiple behavior indicator data includes the time information of the multiple behavior indicator data. According to the behavior data indicator value corresponding to each time information, the characteristic value in the behavior data indicator value is determined, and then the behavior indicator data corresponding to the characteristic value is used as the characteristic indicator data. The characteristic value is one of the behavior data indicator values, and the characteristic value can characterize the characteristics of the user behavior characteristics. The characteristic value can be the maximum / minimum value of the behavior data indicator values ​​corresponding to the multiple behavior indicator data. For example, the characteristic can be the peak period or trough period when the user performs a certain operation or uses a certain product. The characteristic indicator data includes each characteristic value and the time information corresponding to each characteristic value.

[0052] Feature indicator data that can characterize user behavior characteristics is screened out from the user's behavior indicator data to achieve screening of the behavior indicator data and reduce the impact of noise points on mining user behavior patterns. Moreover, mining user behavior patterns based on feature indicator data that characterize user behavior characteristics is conducive to obtaining accurate user behavior patterns. In addition, compared with the data volume of behavior indicator data, the data volume of feature indicator data is greatly reduced. Compared with the method of directly mining user behavior patterns based on behavior indicator data, mining user behavior patterns based on feature indicator data is conducive to quickly obtaining accurate user behavior patterns, reducing the amount of data processing in the process of mining user behavior patterns, and improving the efficiency of mining user behavior patterns.

[0053] Mining user behavior patterns based on feature indicator data. Because feature indicator data includes feature values ​​and the time corresponding to each feature value, the essence of mining user behavior patterns is based on the feature values ​​and the time corresponding to each feature value. Feature values ​​and the time corresponding to the feature values ​​can represent user behavior characteristics and temporal characteristics, respectively. Based on the user's feature values ​​and the time corresponding to the feature values, periodic features in user behavior patterns can be mined.

[0054] The user behavior pattern mining method provided in the present application determines characteristic indicator data that can characterize user behavior characteristics from user behavior indicator data, and mines user behavior patterns based on the characteristic indicator data. Since the characteristic indicator data includes characteristic values ​​that characterize user behavior characteristics and time information corresponding to characteristic values ​​that can determine periodic characteristics of user behavior, user behavior patterns can be accurately determined based on the characteristic indicator data. Moreover, compared with the method of directly using behavior indicator data to determine user behavior patterns, the method of using characteristic indicator data screens and eliminates the behavior indicator data, thereby reducing the amount of data for mining user behavior patterns.

[0055] If the distribution of user behavior data is relatively discrete, directly clustering the user behavior data may not produce effective clustering results, and thus it is impossible to find the patterns in the user behavior data; moreover, if there is a lot of user behavior data, directly clustering a large amount of user behavior data is not only difficult to cluster, but the amount of data processed by clustering is also large, and it is impossible to efficiently obtain the patterns in the user behavior data. The solution provided by this application first screens out characteristic indicator data that can characterize user behavior characteristics from multiple behavior indicator data, and then mines user behavior patterns through the characteristic indicator data that can characterize user behavior characteristics. On the one hand, it can accurately characterize user behavior characteristics based on characteristic indicator data, and on the other hand, it greatly reduces the amount of basic data for mining user behavior patterns, which is conducive to improving the efficiency of obtaining user behavior patterns.

[0056] In order to make the user behavior pattern mining solution and its technical effects provided by this application clearer, its specific implementation plan is described in detail with multiple examples below.

[0057] In an optional embodiment, based on the time series information corresponding to the plurality of behavior indicator data, the characteristic value in each behavior data indicator value is determined, which can be obtained by the following methods, including:

[0058] Based on time information, multiple behavioral indicator data are divided to obtain behavioral indicator data corresponding to multiple time periods;

[0059] For each time period, based on the time series information corresponding to each behavior indicator data belonging to the time period, the characteristic value in each behavior data indicator value of each time period is determined.

[0060] When the length of the time information corresponding to the behavior indicator data is long and the data indicator values ​​corresponding to the time information are large, directly processing the feature indicator data and mining the user behavior patterns for all the behavior indicator data will generate a large amount of data processing. To reduce the data processing amount, the optional embodiment of the present application provides the following solution to determine the feature values ​​in the behavior data indicator values, which may include:

[0061] Segmenting the plurality of behavior indicator data according to time series information corresponding to the plurality of behavior indicator data to obtain behavior indicator data corresponding to a plurality of time periods;

[0062] For each time period, based on the time series information corresponding to each behavior indicator data belonging to the time period, the characteristic value in each behavior data indicator value of each time period is determined.

[0063] The behavior indicator data is segmented according to a preset time period, and the behavior indicator data is segmented into behavior indicator data corresponding to a plurality of time periods. In this case, for the behavior indicator data within each time period, a characteristic value of each behavior indicator value within each time period is determined based on the time information of each behavior indicator data within the time period and the magnitude of the corresponding behavior indicator value. The behavior indicator data corresponding to the characteristic value is then used as the characteristic indicator data within the time period.

[0064] Combine Figure 3 As shown, Figure 3 A schematic diagram of user behavior indicator data provided by an embodiment of the present application is shown. Figure 3 and Figure 2 The same behavioral indicator data is shown. Figure 2 compared to, Figure 3 It also shows the behavioral indicator data of users divided according to the specified time period. Figure 2Shows the user's behavior indicator data divided by the behavior data indicator value. Figure 3 In the example, the behavior indicator data is segmented into days as the time period, and the behavior indicator value in each behavior indicator data is identified. For example, the behavior indicator value on the 4th day is 1, the behavior indicator value on the 5th day is 5, and so on. If the corresponding operation is the number of payment operations, the behavior indicator data indicates that 1 payment operation was performed on the 4th day and 5 payment operations were performed on the 5th day.

[0065] On the basis of the solution provided in this embodiment, mining the user behavior pattern of the user based on each characteristic indicator data provided in S130 can be performed in the following manner: mining the user behavior pattern of the user based on the characteristic indicator data corresponding to each time period.

[0066] Based on the characteristic indicator data corresponding to each time period in the behavioral indicator data, user behavior patterns are mined. If there is one characteristic indicator data corresponding to each time period, the characteristic indicator data of the same number as the time period are used to mine the user behavior pattern. Since the characteristic indicator data can characterize the user behavior characteristics within the time period, the characteristic indicator data corresponding to all time periods can characterize the overall user behavior characteristics and obtain the user behavior pattern.

[0067] The solution provided in the embodiment of the present application divides the behavior indicator data according to time periods, obtains characteristic indicator data within each time period, and uses the characteristic indicator data corresponding to each time period to mine user behavior patterns. This can efficiently obtain characteristic indicator data that can characterize user behavior characteristics, and is conducive to reducing the amount of data processing in mining user behavior patterns.

[0068] The present application also provides an optional embodiment for mining user behavior patterns based on various feature indicator data. This solution can be applied to solutions that use time periods to divide behavior indicator data, as well as solutions that do not divide behavior indicator data, and may include:

[0069] Determine the temporal characteristics of user behavior based on the time information of each characteristic indicator data;

[0070] Determine the behavioral characteristics of user behavior based on the time information and corresponding characteristic values ​​of each characteristic indicator data;

[0071] Mining user behavior patterns based on the temporal characteristics and behavioral characteristics.

[0072] Combine Figure 2 and Figure 3As shown, if the time information of the characteristic indicator data corresponds to the 2nd day, 5th day, 8th day, 11th day, 14th day, and 16th day respectively, then the temporal characteristics of the user behavior determined based on the time information can be a time period of 3 days or characteristic values ​​appearing on the 2nd day, 5th day, 8th day, 11th day, 14th day, and 16th day of the behavior indicator data.

[0073] Based on the time information of the characteristic indicator data and the corresponding characteristic value, the behavioral characteristics of the user behavior are determined. Figure 2 and Figure 3 As shown, the characteristic value corresponding to the characteristic indicator data is a local peak value of several adjacent time periods, indicating that the behavioral characteristics of the user behavior are: the execution peak period of the specified operation appears in the time information of the characteristic indicator data.

[0074] Mining user behavior patterns based on time series features and behavioral features. For example, if a specified operation has a peak execution period every three days, and if the specified operation is a payment operation, then the user behavior pattern is that the peak execution period of the payment operation occurs every three days.

[0075] Optionally, based on the time information of each characteristic indicator data, determining the temporal characteristics of the user behavior can be done in the following ways:

[0076] Based on the time series information of each feature indicator data, determining the time difference between the feature indicator data adjacent in time, the time series feature including each time difference;

[0077] Similarly, based on the time information and corresponding feature values ​​of each feature indicator data, determining the behavioral characteristics of the user behavior can be achieved in the following ways:

[0078] Based on the time information of each characteristic indicator data, the indicator difference between the characteristic values ​​adjacent in time is calculated, and the behavior characteristics include each indicator difference.

[0079] The scheme for determining timing characteristics and behavioral characteristics provided in the present application can be applicable to schemes for dividing behavioral indicator data into time periods or not. Therefore, the characteristic indicator data provided in this embodiment include adjacent characteristic indicator data in the scheme for not dividing the time period, and also include characteristic indicator data corresponding to adjacent time periods in the scheme for dividing the time period.

[0080] Combine Figures 2 to 3The diagram of behavioral indicator data illustrates this solution. The time series information of the characteristic indicator data is: day 2, day 5, day 8, day 11, day 14, day 16. The time differences between adjacent characteristic indicator data are determined as: 3 days, 3 days, 3 days, 3 days, 2 days, and so on. These time differences can be used as the time series characteristics of the characteristic indicator data. The indicator differences between adjacent behavioral data indicator values ​​are: 2, 1, 2, 1, and 2, and these indicator differences can be used as the user's behavioral characteristics.

[0081] The solution provided in the embodiment of the present application obtains the time difference and indicator difference between the characteristic indicator data. The time difference can be used to characterize the user's temporal characteristics, and the indicator difference can be used to characterize the user's behavioral characteristics. The temporal characteristics and behavioral characteristics in the user's behavioral indicator data are mined, which is conducive to obtaining accurate user behavior patterns based on the temporal characteristics and behavioral characteristics.

[0082] Figure 2 and Figure 3 It is shown that there is at most one characteristic indicator data in each time period, but the selection conditions of the characteristic indicator data can be adjusted. As the selection conditions of the characteristic indicator data are adjusted, there may be more than one characteristic indicator data in a time period. In this case, the present application provides an optional embodiment to determine the time difference between temporally adjacent characteristic indicator data, which may include:

[0083] For every two adjacent time periods, the time difference between the characteristic indicator data belonging to different periods in the two time periods is determined.

[0084] In this case, based on the time information of each characteristic index data, the index difference between the characteristic values ​​adjacent in time is calculated, which can be obtained by the following method:

[0085] For every two adjacent time periods, the index difference between the characteristic values ​​belonging to different periods in the two time periods is determined.

[0086] The essence of this solution is: obtaining the time difference between each characteristic indicator data corresponding to the current time period and each characteristic indicator data corresponding to the adjacent time period; obtaining the indicator difference between each characteristic indicator data in the current time period and each characteristic indicator data in the adjacent time period.

[0087] For example, if there is one characteristic indicator data A in the current time period a, and two characteristic indicator data B and C in the adjacent time period b of the current time period, then the time difference / indicator difference between the characteristic indicator data in the adjacent time periods is the time difference / indicator difference between characteristic indicator data A and B, and between A and C. If there are two characteristic indicator data A and D in the current time period a, and two characteristic indicator data B and C in the adjacent time period b of the current time period, then the distance between the characteristic indicator data in the adjacent time periods is the time difference / indicator difference between characteristic indicator data A and B, A and C, D and B, and D and C. If there are two characteristic indicator data A and D in the current time period a, and one characteristic indicator data B in the adjacent time period b of the current time period, then the distance between the characteristic indicator data in the adjacent time periods is the time difference / indicator difference between characteristic indicator data A and B, and between D and B.

[0088] The solution provided in the embodiment of the present application is a solution for determining the distance between each characteristic indicator data in adjacent time periods when there is more than one characteristic indicator data in a time period. It is beneficial to obtain a user behavior pattern with a periodicity that is smaller than the above-mentioned preset time period, and obtain more periodic features in the user behavior pattern. For example: when the preset time period is a day, the daily periodic features in the user behavior pattern can be obtained, and the weekly or annual periodic features can also be obtained.

[0089] After obtaining the time series features and behavioral features of user behavior, an optional embodiment of the present application further provides a solution for mining user behavior patterns based on the time series features and behavioral features, which may include:

[0090] A1, performing clustering processing on the time difference and / or index difference to obtain a clustering result;

[0091] A2: Determine the user behavior pattern based on the clustering result.

[0092] The embodiments of the present application provide at least the following three methods to obtain clustering results: first, clustering the time differences to obtain clustering results; second, clustering the index differences to obtain clustering results; third, clustering the time differences and index differences to obtain clustering results.

[0093] For example, the time difference or indicator difference between each characteristic indicator data is clustered according to time to obtain a clustering result. For example, the characteristic values ​​of characteristic indicator data A, B, and C correspond to the 1st day, the 3rd day, the 5th day, and the 9th day respectively. Then the time difference between temporally adjacent characteristic indicator data includes 2, 2, and 4. The time difference is clustered to obtain a clustering result. The clustering result can be: the same time difference is a clustering result, such as a time difference of 2 is a clustering result.

[0094] By clustering the distances between the characteristic indicator data in all adjacent time periods according to the time difference and indicator difference, and treating the same time difference or indicator difference as the same category, we can directly obtain the characteristics of the user's execution time or execution times for a specified operation.

[0095] The correspondence between clustering results and user behavior patterns is related to the specific operation performed. For example, if the operation performed is backend access, the clustering result indicates that the time difference between the two distances is the same category. In this case, the user behavior pattern may be a peak or trough in backend access occurring two hours apart. In the actual scenario where users use the application to make payments, clustering based on the time difference of payment operations may result in the following clustering results: Category 1 corresponds to a time difference of 5, Category 2 corresponds to a time difference of 7, Category 3 corresponds to a time difference of 30, and so on. Here, the time difference can be measured in days. A time difference of 5 indicates that similar payment behaviors occur five days apart. This payment behavior may be a payment peak period, and the corresponding user behavior pattern is that users experience a payment peak every five days.

[0096] The solution provided in the embodiment of the present application clusters the distances between each feature indicator data, thereby removing noise points that do not conform to the distance law. Moreover, based on the clustering results, the temporal characteristics or behavioral characteristics of the behavior indicator data can be directly determined, which is conducive to quickly obtaining user behavior patterns.

[0097] In an optional embodiment, determining a characteristic value in each behavior data indicator value based on the time series information corresponding to the plurality of behavior indicator data provided in S120 and using the behavior indicator data corresponding to each characteristic value as the characteristic indicator data can be implemented in the following manner, including:

[0098] B1, for any behavior data index value, if the behavior data index value is greater than the previous behavior data index value and greater than the next behavior data index value, then the behavior data index value is determined as a feature point;

[0099] B2, determining feature index data based on each feature point.

[0100] The solution provided by the embodiment of the present application determines characteristic indicator data based on the behavior data indicator value and the corresponding time, determines the behavior data indicator value corresponding to the previous moment and the next moment of a certain behavior indicator data based on the time information, compares the behavior data indicator value corresponding to the current moment with the behavior data indicator value corresponding to the previous moment and the behavior data indicator value corresponding to the next moment, and if the behavior data indicator value corresponding to the current moment is greater than the behavior data indicator value corresponding to the previous and next moments, then the behavior indicator data corresponding to the moment is determined as a characteristic point, that is, the behavior data indicator value time corresponding to the moment is determined as a characteristic point. Optionally, the moment mentioned here can also be a time period, and time can be used as a general term for moment and time period.

[0101] According to the above method, the characteristic points in the behavioral indicator data are obtained, combined with Figure 4 As shown, Figure 4 The figure shows a schematic diagram of feature points provided by an embodiment, and the behavior indicator data shown in the figure is Figure 3 The behavioral indicator data shown in Figure 4 The eigenvalues ​​corresponding to the feature points in are 3, 5, 6, 8, 7, and 9 respectively.

[0102] A feature point is a feature value that characterizes the user's behavior characteristics. The user's feature index data can be obtained by screening multiple feature points. The feature point can also be used as the user's feature index data to mine the user's behavior pattern based on the user's feature index data.

[0103] An optional embodiment provides a method for determining feature index data based on feature points, including:

[0104] B21, for any feature point, determine the index value change rate of the feature point based on the feature point and the behavior data index values ​​adjacent to the feature point in time;

[0105] B22, determining each feature point whose index value change rate is greater than the set value as feature index data.

[0106] For any feature point, based on the feature value of the feature point and the behavioral data index value adjacent to the feature point in time, calculate the index difference between the feature value of the feature point and the behavioral data index value corresponding to the adjacent time, calculate the time difference between the feature point and the adjacent time, and determine the index value change rate of the feature point based on the time difference and the index difference. For example: within 1 minute, the number of user payments increases from 10 to 50, and then decreases to 20. The behavioral data index value corresponding to the feature point is 50, and the index value change rate can be: 30 transactions within one minute.

[0107] The index value change rate is pre-set according to the actual situation, the index value change rate corresponding to each feature point is obtained, and each feature point whose index value change rate is greater than the set value is determined as the feature index data representing the user characteristics.

[0108] The indicator value change rate can be adjusted according to actual conditions. The number and position of characteristic indicator data can also be adjusted by adjusting the indicator value change rate. If the time period of the behavioral indicator data is long, the indicator value change rate can be increased and the number of characteristic indicator data can be reduced. This allows you to more intuitively see the temporal patterns of the behavioral indicator data over a longer time dimension.

[0109] The solution provided in the embodiment of the present application determines, as characteristic indicator data, behavior indicator data whose indicator value is greater than the indicator value of the previous and subsequent behavior data and whose indicator value change rate is greater than a set value. The screened characteristic indicator data retains the temporal characteristics of the behavior indicator data and satisfies the indicator value change rate, that is, satisfies the change rate trend of the behavior indicator data, and is more capable of characterizing the characteristics of user behavior.

[0110] In an optional embodiment, a solution for determining the rate of change of an index value of a feature point is provided, including:

[0111] Based on the feature point and the behavior data indicator values ​​adjacent to the feature point in time, a rate of change of the indicator value of the feature point is determined.

[0112] Obtain the time corresponding to the feature point and the adjacent time adjacent to the time, obtain the behavior data index value corresponding to the feature point and the behavior data index value corresponding to the adjacent time, and determine the index value change rate of the feature point based on the absolute value of the difference between the behavior data index value corresponding to the feature point and the behavior data index value corresponding to the adjacent time and the time difference.

[0113] If there are two behavior data index values ​​corresponding to adjacent times adjacent to the time, two corresponding index value change rates are obtained, and the larger index value change rate can be used as the index value change rate of the feature point.

[0114] In an optional embodiment, obtaining the user's behavior indicator data may be performed in the following manner, including:

[0115] C1, obtaining a user behavior pattern analysis request, where the analysis request includes a specified time period and / or a specified operation type;

[0116] C2: Based on the analysis request, obtain user behavior indicator data corresponding to the specified time period and / or specified operation type.

[0117] The user in C1 can be any user who has performed a specified operation type, and the background has stored the user's data for the specified time period, or the data of the specified operation type. The server receives a request for analyzing the user's behavior pattern, which can include an analysis request for each operation type performed by the user within the specified time period, an analysis request for the user to perform a specified operation type, or an analysis request for the user to perform a specified operation type within the specified time period.

[0118] The server parses the analysis request, obtains the time period and / or specified operation type contained in the analysis request, obtains the user's behavior indicator data corresponding to the specified time period and / or specified operation type from the background or database, and executes steps S110 to S130 to mine user behavior patterns based on the behavior indicator data.

[0119] After mining the user behavior patterns according to the solutions provided in the above embodiments, data analysis can be performed based on the user behavior patterns, or corresponding marketing strategies can be formulated based on the user behavior patterns.

[0120] In order to better illustrate the mining method of user behavior patterns provided by this application, combined with Figure 5 The following describes the method for mining user behavior patterns using a time sequence diagram.

[0121] A user, such as a developer, requests a user behavior pattern of a user using a product or performing a specified operation from the server. The server receives the request, parses the request using an analysis program in the memory, obtains the user information, specified time period, and / or specified operation in the request, and retrieves the user's behavior indicator data (corresponding to the user behavior indicator data) from the data warehouse storing the user behavior data based on the user information, specified time period, and / or specified operation. Figure 5 User behavior data in ).

[0122] After retrieving the behavioral indicator data related to the user's use of the product or specified operation, the analysis program is executed to implement the following operations: according to the preset time period, such as hours, days, weeks, months, etc., the behavioral indicator data is divided into multiple time periods, the corresponding behavioral indicator data in each time period is obtained, and the characteristic indicator data in each time period is filtered out according to the preset filtering method (corresponding to Figure 5The preset screening method can be: within the current time period, if the characteristic value at a certain moment is greater than the values ​​on both sides thereof, a local peak appears, and the rate of change of the peak is greater than a certain value, then the peak at that moment is used as the characteristic value, and the corresponding characteristic indicator data is determined based on the characteristic value. The distance between the corresponding characteristic indicator data in adjacent time periods is calculated, such as the time difference and / or the indicator difference, and the time and characteristic value of the characteristic indicator data are saved. Finally, all the distances corresponding to the characteristic indicator data corresponding to the adjacent time periods are clustered to obtain a time series pattern that can characterize the user's time series characteristics and behavioral characteristics, determine the user's behavior pattern based on the time series pattern, and send the user behavior pattern to the request sender.

[0123] It is worth noting that the behavior indicator data in the above embodiment may be the behavior indicator data of the user corresponding to the application. The acquisition of multiple behavior indicator data of the user provided in this application may be obtained in the following manner:

[0124] Based on the user's user identification, multiple behavior indicator data of the user are obtained from the data warehouse of the application program, wherein the data warehouse is used to associate and store the user identification of each user with the behavior indicator data corresponding to each user.

[0125] The user warehouse stores behavioral indicator data for at least one application corresponding to a user, including the user ID corresponding to the user using the at least one application and the behavioral indicator data associated with the user ID. Pre-storing multiple behavioral indicator data corresponding to a user's applications in the data warehouse facilitates direct retrieval of behavioral indicator data for a user's use of a particular application from the data warehouse, improving the efficiency of obtaining behavioral indicator data.

[0126] The behavior indicator data provided in any of the above embodiments can be further subdivided. If the behavior indicator data includes behavior indicator data corresponding to at least two information categories, in the user behavior pattern mining method provided in this application, step S120 provides determining a feature value in each behavior data indicator value based on the time series information corresponding to the multiple behavior indicator data, and using the behavior indicator data corresponding to each feature value as the feature indicator data, which can also be achieved in the following manner:

[0127] Dividing the plurality of behavior indicator data according to information categories to obtain a plurality of behavior indicator data corresponding to each information category;

[0128] For each information category, based on the time series information corresponding to multiple behavior indicator data of the information category, the characteristic values ​​in the behavior data indicator values ​​corresponding to the information category are determined, and the characteristic values ​​corresponding to the information category are used as the characteristic indicator data corresponding to the information category.

[0129] In practice, each application typically provides users with data from multiple information categories. For example, a food delivery app might provide users with data on food, fruit, and medicine; a multimedia app might provide users with data from multiple information categories, such as music and videos. To obtain more detailed user behavior patterns for this application, we first divide the multiple behavioral indicator data by information category, generating multiple behavioral indicator data sets corresponding to each information category.

[0130] It should be noted that the embodiment of the present application does not limit the method for dividing information categories in applications. The classification can be performed according to actual needs. The classification method for different categories of applications can also be different. For example, for a user's payment behavior on a payment application, the information categories under this behavior may include: payment for food expenses, payment for transportation expenses, payment for clothing expenses, etc. One of the information categories can be further subdivided, such as payment for food expenses can be divided into payment for meal expenses, payment for fruit expenses, payment for dessert expenses, etc. The user's multiple behavior indicator data are then divided according to this information category to obtain multiple behavior indicator data corresponding to each information category.

[0131] In order to obtain the user behavior patterns corresponding to each information category and provide users with more accurate and personalized recommendations, in an embodiment of the present application, after obtaining multiple behavioral indicator data corresponding to each information category, time series information corresponding to the multiple behavioral indicator data corresponding to each information category can be obtained. For example, the time information of paying for food expenses can be arranged in sequence to obtain the time series information corresponding to the behavioral indicator data corresponding to that information category. The characteristic values ​​of each behavioral data indicator value corresponding to the information category can be determined according to the scheme for determining characteristic values ​​in the behavioral data indicator values ​​provided in the above embodiment, and each characteristic value corresponding to the information category can be used as the characteristic indicator data corresponding to the information category. For example, if the amount of food expenses paid by the user on a certain Saturday is much greater than the expenses on Friday and Sunday, the food expenses paid on that Saturday can be used as one of the characteristic indicator data for paying for food expenses.

[0132] On this basis, the user behavior patterns of users provided in step S130 of this application are mined based on various feature indicator data, including:

[0133] For each information category, a user behavior pattern corresponding to the information category is mined based on each characteristic indicator data of the information category, wherein the user behavior pattern includes a time series pattern.

[0134] After obtaining the characteristic indicator data corresponding to each information category, the user behavior patterns corresponding to that information category are mined based on this characteristic indicator data. For example, the characteristic indicator data for paying for food expenses is: a characteristic value appears every Saturday, that is, there is a peak period for paying for food expenses every 7 days. For the information category of paying for food expenses, the time series pattern is a 7-day cycle, and the characteristic value occurs every Saturday. When making information recommendations based on this time series pattern, for example, food-related information can be recommended to users every Friday or Saturday. On this basis, combined with the mining of characteristic values, that is, analyzing the data on food payment expenses, the user's payment pattern for food expenses can be obtained. For the four characteristic points that appear in each month, the payment expenses decrease in sequence. When making information recommendations, based on the time information of the recommendation moment, such as the week of the current month, food information with appropriate prices is recommended to the user.

[0135] After obtaining the user behavior pattern according to the solution provided in the above embodiment, a method for mining the user behavior pattern provided in an optional embodiment of the present application further includes:

[0136] Based on the user's usage time of the application, at least one timing pattern matching the usage time is determined, and based on the determined timing pattern and the usage time, relevant information of the information category corresponding to the determined timing pattern is recommended to the user.

[0137] An example is as follows: based on the user's usage time, such as noon, afternoon, evening, etc., at least one timing pattern of the corresponding information category in the application that matches the usage time can be determined, and dietary information corresponding to the timing pattern corresponding to the usage time can be recommended to the user, such as: recommending dinner, afternoon tea, midnight snack, etc.

[0138] There are multiple timing patterns corresponding to an application. For example, for a takeaway application, the corresponding information category data includes catering data, fruit data, medicine data, etc. Each type of data has a corresponding timing pattern. First, based on the user's usage time of the application and these timing patterns, the timing pattern that matches the usage time is determined, and then the relevant information of the corresponding information category is determined for the matching timing pattern. For example, if the current usage time is 3 pm, among the timing patterns corresponding to catering data, fruit data, and medicine data, only the timing pattern of catering data matches the usage time, and the behavior of afternoon tea at 3 o'clock exists in the timing pattern corresponding to catering data, then the relevant information of afternoon tea is recommended to the user.

[0139] It is understandable that when there is no timing pattern that matches the usage time, relevant information can be recommended according to the recommendation strategy of the prior art or according to other preconfigured recommendation strategies.

[0140] Based on the user behavior pattern mining method provided in the above embodiment, an optional embodiment of the present application further provides a recommendation method based on user behavior patterns, including:

[0141] D1, obtaining multiple behavior indicator data of the user, wherein the multiple behavior indicator data include behavior indicator data of at least two information categories of the user corresponding to the application;

[0142] D2, dividing the plurality of behavioral indicator data according to information categories to obtain a plurality of behavioral indicator data corresponding to each information category;

[0143] D3, for each information category, based on the time series information corresponding to the multiple behavior indicator data of the information category, determine the characteristic values ​​in the behavior data indicator values ​​corresponding to the information category, and use the characteristic values ​​corresponding to the information category as the characteristic indicator data corresponding to the information category;

[0144] D4, for each information category, mining the user behavior pattern corresponding to the information category based on the characteristic indicator data of the information category, wherein the user behavior pattern includes a time series pattern;

[0145] D5, based on the user's usage time of the application, determine at least one timing pattern that matches the usage time, and based on the determined timing pattern and the usage time, recommend relevant information of the information category corresponding to the determined timing pattern to the user.

[0146] Multiple behavioral indicator data comprising at least two information categories can be obtained from a data warehouse storing user identifiers and user behavioral indicator data corresponding to applications. Each application typically provides users with data in multiple information categories. For example, regarding a user's payment behavior on a payment application, information categories for such behavior may include: payment for food, payment for transportation, payment for clothing, etc. An information category can be further subdivided, such as payment for food can be divided into payment for meals, payment for fruit, payment for dessert, etc.

[0147] To obtain more detailed user behavior patterns for applications, multiple behavioral indicator data can be divided according to information categories, resulting in multiple behavioral indicator data corresponding to each information category. The division of information categories can be based on actual needs. The division method for different categories of applications can vary, and this embodiment of the application does not impose any restrictions.

[0148] In order to obtain the user's behavior patterns corresponding to each information category and provide more accurate and personalized recommendations to the user, in an embodiment of the present application, after obtaining multiple behavioral indicator data corresponding to each information category, time series information corresponding to the multiple behavioral indicator data corresponding to each information category can be obtained. For example, the time information of paying for food expenses can be sorted in order to obtain the time series information corresponding to the payment of food expenses. The characteristic values ​​of each behavior data indicator value corresponding to each information category can be determined according to the scheme for determining characteristic values ​​in behavior data indicator values ​​provided in the above embodiment. For example, if the amount of food expenses paid by a user on a certain Saturday is much greater than that paid on Friday and Sunday, the food expenses paid on that Saturday can be used as a characteristic value of the food expenses paid.

[0149] On this basis, for each information category, the user behavior pattern corresponding to the information category is mined based on the characteristic indicator data of the information category, wherein the user behavior pattern includes a time series pattern.

[0150] After obtaining the characteristic indicator data corresponding to each information category, we then use this characteristic indicator data to mine the user behavior patterns corresponding to that information category. For example, if the characteristic indicator data for paying for food expenses is a single eigenvalue that appears every Saturday, this indicates a peak period for paying for food expenses every seven days. For this information category, the time series pattern is a seven-day cycle, with eigenvalues ​​occurring every Saturday. Based on this, combined with eigenvalue mining—that is, analyzing the data on food payment expenses—we can derive the user's payment pattern for food expenses: for each of the four eigenvalues ​​that appear each month, the payment decreases over time.

[0151] After obtaining the user behavior pattern, based on the user's usage time of the application, at least one timing pattern that matches the usage time is determined, and based on the determined timing pattern and usage time, relevant information of the information category corresponding to the determined timing pattern is recommended to the user.

[0152] For example, a user in a food delivery app has two information categories. One is for paying for food, and the corresponding user behavior pattern is: a payment peak occurs every Saturday. It is detected that the user uses the food delivery app on Fridays. The other is for paying for fruit, and the corresponding user behavior pattern is: fruit payments occur once every Thursday. Based on the time the user uses the food delivery app, a timing pattern matching this usage time is first determined. Based on this matching timing pattern, information categories corresponding to this timing pattern are recommended to the user. For example, if the current usage time is Friday and the timing pattern matching this usage time is determined to be a timing pattern for paying for food, food-related information is recommended to the user. Furthermore, food information matching the user's payment can be recommended based on the week of the month in which the current usage time falls. Usage time can be further refined, and based on the time period of the user's use of the relevant app, such as lunch, afternoon, or evening, food information corresponding to this usage time and timing pattern is recommended to the user, such as dinner, afternoon tea, or midnight snack.

[0153] It is understandable that when there is no timing pattern that matches the usage time, relevant information can be recommended according to the recommendation strategy of the prior art or according to other preconfigured recommendation strategies.

[0154] The recommendation scheme provided in the embodiment of the present application recommends relevant information of the information category corresponding to the usage time and user behavior pattern to the user based on the user behavior pattern and the user's usage time of the application, thereby achieving accurate recommendation, which is conducive to increasing the probability of the recommended information being adopted by the user and can improve the user experience.

[0155] In order to better illustrate the mining method and recommendation method provided by this application, the following two examples are used to further illustrate the solution provided by the embodiment of this application:

[0156] Example 1

[0157] The application scenario of this embodiment is payment operations. The behavioral indicator data of payment operations includes payment time and payment amount / number of payments. The characteristic value in the behavioral indicator data is the payment peak, which is the peak of the payment amount or the peak of the number of payments within a preset time period. The user behavior pattern can be the periodic regularity of payment operations over time, the periodic regularity of payment amount and time, or the periodic regularity of payment number over time. The details are as follows:

[0158] Obtain the user's payment time and number of payments, segment the payment time by hour, and calculate the number of payments occurring per hour. Determine the peak number of payments based on the number of payments occurring in adjacent hours. Calculate the temporal distance between two adjacent payment peaks to obtain the payment cycle in which the peak occurs, which can be every eight hours. Calculate the temporal distance between two adjacent payment peaks to obtain a pattern in the number of payments, which can be two more payments than the previous one every eight hours. Based on this payment cycle and payment pattern, recommend information that matches the user's payment cycle and payment pattern.

[0159] Example 2

[0160] The application scenario of this embodiment is to recommend information to users using a takeout app. This scenario includes the following two types of behavioral indicator data: behavioral indicator data corresponding to food payment and fruit payment. The behavioral indicator data for food payment includes payment time and payment amount. The corresponding characteristic value is the peak value of the payment amount. The time series pattern is the temporal pattern of food payment. The behavioral indicator data corresponding to fruit payment is similar and will not be repeated here. The details are as follows:

[0161] If the user is currently using a food delivery app on Friday, and considering the time series pattern of food payments (peak payment every Saturday), and the time series pattern of fruit payments (peak payment every Thursday), the app determines that the time series pattern matching this usage time is the peak payment pattern every Saturday, and the information category corresponding to this time series pattern is food payments, then food-related information is recommended to the user. This recommendation scheme increases the likelihood of users adopting recommended information, improves the conversion rate of recommended information, reduces the time users spend in the app, and improves the user experience.

[0162] Based on the same principle as the method provided in the embodiment of the present application, the embodiment of the present application also provides a user behavior pattern mining device 600, such as Figure 6 As shown, the device may include: a behavior indicator data acquisition module 610, a feature indicator data determination module 620, and a user behavior pattern mining module 630, wherein:

[0163] The behavior indicator data acquisition module 610 is used to acquire multiple behavior indicator data of the user, wherein the behavior indicator data includes time information and behavior data indicator values ​​corresponding to the time information;

[0164] A characteristic indicator data determination module 620 is configured to determine a characteristic value in each behavior indicator value based on time series information corresponding to the plurality of behavior indicator data, and use the behavior indicator data corresponding to each characteristic value as the characteristic indicator data;

[0165] The user behavior pattern mining module 630 is used to mine the user behavior pattern of users based on various feature indicator data.

[0166] The user behavior pattern mining device provided in the present application determines characteristic indicator data that can characterize user behavior characteristics from user behavior indicator data, and mines user behavior patterns based on the characteristic indicator data. Since the characteristic indicator data includes characteristic values ​​that characterize user behavior characteristics and time information corresponding to characteristic values ​​that can determine periodic characteristics of user behavior, user behavior patterns can be accurately determined based on the characteristic indicator data. Moreover, compared with the method of directly using behavior indicator data to determine user behavior patterns, the method of using characteristic indicator data screens and eliminates the behavior indicator data, thereby reducing the amount of data for mining user behavior patterns.

[0167] Optionally, the plurality of behavior indicator data include behavior indicator data of the user corresponding to at least two information categories of the application; the characteristic indicator data determination module 620 is specifically configured to:

[0168] Dividing the plurality of behavior indicator data according to information categories to obtain a plurality of behavior indicator data corresponding to each information category;

[0169] For each information category, based on the time series information corresponding to the plurality of behavior indicator data of the information category, determine the characteristic values ​​in the indicator values ​​of the behavior data corresponding to the information category, and use the characteristic values ​​corresponding to the information category as the characteristic indicator data corresponding to the information category;

[0170] The user behavior pattern mining module 630 is specifically used to:

[0171] For each information category, mining the user behavior pattern corresponding to the information category based on the characteristic indicator data of the information category, the user behavior pattern includes a time series pattern;

[0172] The user behavior pattern mining device 600 also includes a recommendation module;

[0173] The recommendation module is used to determine at least one timing pattern that matches the usage time based on the user's usage time of the application, and recommend relevant information of the information category corresponding to the determined timing pattern to the user based on the determined timing pattern and the usage time.

[0174] Optionally, the user behavior pattern mining module 630 is specifically configured to:

[0175] Based on the time information, the plurality of behavior indicator data are divided to obtain behavior indicator data corresponding to a plurality of time periods;

[0176] For each time period, determining a characteristic value in each behavior data indicator value of each time period based on the time series information corresponding to each behavior indicator data belonging to the time period;

[0177] Mining the user behavior pattern of the user based on each of the characteristic indicator data includes:

[0178] Based on the characteristic indicator data corresponding to each of the time periods, the user behavior pattern of the user is mined.

[0179] Optionally, the user behavior pattern mining module 630 is further configured to:

[0180] Determining the temporal characteristics of user behavior based on the time information of each characteristic indicator data;

[0181] Determining the behavioral characteristics of the user behavior based on the time information and corresponding characteristic values ​​of each of the characteristic indicator data;

[0182] The user behavior pattern is mined based on the time series features and the behavior features.

[0183] Optionally, the user behavior pattern mining module 630 is further configured to: determine a time difference between temporally adjacent feature indicator data based on time series information of each feature indicator data, wherein the time series feature includes each time difference;

[0184] Determining the behavioral characteristics of the user behavior based on the time information and the corresponding characteristic value of each characteristic indicator data includes:

[0185] Based on the time information of each of the characteristic indicator data, an indicator difference between temporally adjacent characteristic values ​​is calculated, and the behavior feature includes each of the indicator differences.

[0186] Optionally, the user behavior pattern mining module 630 is further configured to:

[0187] For every two adjacent time periods, determine the time difference between the characteristic indicator data belonging to different periods in the two time periods;

[0188] The calculating the index difference between temporally adjacent feature values ​​based on the time information of each feature index data includes:

[0189] For every two adjacent time periods, the index difference between the characteristic values ​​belonging to different periods in the two time periods is determined.

[0190] Optionally, the user behavior pattern mining module 630 is further configured to:

[0191] Performing clustering processing on the time difference and / or the indicator difference to obtain a clustering result;

[0192] A user behavior pattern is determined based on the clustering result.

[0193] Optionally, the feature indicator data determination module 620 is configured to:

[0194] For any of the behavior data index values, if the behavior data index value is greater than the previous behavior data index value and greater than the next behavior data index value, the behavior data index value is determined as a feature point:

[0195] The feature index data is determined based on each of the feature points.

[0196] Optionally, the characteristic indicator data determination module 620 is further configured to:

[0197] For any of the feature points, determining a rate of change of the index value of the feature point based on the feature point and the behavior data index values ​​adjacent to the feature point in time;

[0198] Each feature point whose index value change rate is greater than a set value among the feature points is determined as the feature index data.

[0199] Optionally, the behavior indicator data acquisition module 610 is specifically configured to:

[0200] Obtaining a user's behavior pattern analysis request, wherein the analysis request includes a specified time period and / or a specified operation type;

[0201] Based on the analysis request, user behavior indicator data corresponding to the specified time period and / or the specified operation type is obtained.

[0202] Optionally, the behavior indicator data is the user's behavior indicator data corresponding to the application program, and the behavior indicator data acquisition module 610 is specifically configured to:

[0203] Based on the user identifier of the user, a plurality of behavior indicator data of the user is obtained from the data warehouse of the application, wherein the data warehouse is used to associate and store the user identifier of each user with the behavior indicator data corresponding to each user. The user behavior pattern mining device of the embodiment of the present application can execute the user behavior pattern mining method provided by the embodiment of the present application, and its implementation principle is similar. The actions performed by each module and unit in the user behavior pattern mining device in each embodiment of the present application correspond to the steps in the user behavior pattern mining method in each embodiment of the present application. For the detailed functional description of each module of the user behavior pattern mining device, please refer to the description of the corresponding user behavior pattern mining method shown in the previous text, which will not be repeated here.

[0204] An embodiment of the present application further provides a recommendation device based on user behavior patterns, including:

[0205] A second behavior indicator data acquisition module is used to acquire multiple behavior indicator data of the user, wherein the multiple behavior indicator data include behavior indicator data of the user corresponding to at least two information categories of the application;

[0206] A division module, for dividing the plurality of behavior indicator data according to information categories, and obtaining a plurality of behavior indicator data corresponding to each information category;

[0207] A second characteristic indicator data determination module is configured to determine, for each information category, characteristic values ​​among the behavior data indicator values ​​corresponding to the information category based on time series information corresponding to the plurality of behavior indicator data of the information category, and use the characteristic values ​​corresponding to the information category as characteristic indicator data corresponding to the information category;

[0208] A second user behavior pattern mining module is configured to mine, for each information category, a user behavior pattern of the user corresponding to the information category based on each characteristic indicator data of the information category, wherein the user behavior pattern includes a time series pattern;

[0209] The second recommendation module is used to determine at least one timing pattern that matches the usage time based on the user's usage time of the application, and recommend relevant information of the information category corresponding to the determined timing pattern to the user based on the determined timing pattern and the usage time.

[0210] Based on the same principle as the method shown in the embodiment of the present application, an electronic device is also provided in the embodiment of the present application, which may include but is not limited to: a processor and a memory; the memory is used to store a computer program; the processor is used to execute the method for mining user behavior patterns shown in any optional embodiment of the present application by calling the computer program. Compared with the prior art, the present application determines characteristic indicator data that can characterize the user behavior characteristics from the user's behavior indicator data, and mines the user behavior pattern based on the characteristic indicator data. Since the characteristic indicator data includes each characteristic value that characterizes the user behavior characteristics, and the time information corresponding to each characteristic value that can determine the periodic characteristics of the user behavior, the user behavior pattern can be accurately determined based on the characteristic indicator data. Moreover, compared with the method of directly using the behavior indicator data to determine the user behavior pattern, the method of using the characteristic indicator data filters and eliminates the behavior indicator data, thereby reducing the amount of data for mining the user behavior pattern.

[0211] In an alternative embodiment, an electronic device is provided, such as Figure 7 As shown, Figure 7The electronic device 4000 shown may be a server, including a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, for example, via a bus 4002. Optionally, the electronic device 4000 may further include a transceiver 4004. It should be noted that in actual applications, the number of transceivers 4004 is not limited to one, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present application.

[0212] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 4001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0213] Bus 4002 may include a path for transmitting information between the above components. Bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. Bus 4002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0214] The memory 4003 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to these.

[0215] The memory 4003 is used to store application code for executing the solution of the present application, and the execution is controlled by the processor 4001. The processor 4001 is used to execute the application code stored in the memory 4003 to implement the content shown in the above method embodiment.

[0216] Among them, electronic devices include but are not limited to: mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 7 The electronic device shown is merely an example and should not limit the functions and scope of use of the embodiments of the present application.

[0217] The server provided in this application can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The terminal can be a smartphone, tablet computer, laptop computer, desktop computer, smart speaker, smart watch, etc., but is not limited to these. The terminal and the server can be directly or indirectly connected through wired or wireless communication, and this application does not limit this.

[0218] Among them, the user behavior pattern mining method provided in this application can also be implemented through cloud computing. Cloud computing refers to the delivery and use model of IT infrastructure, which means obtaining required resources through the network in an on-demand and easily scalable manner; in a broad sense, cloud computing refers to the delivery and use model of services, which means obtaining required services through the network in an on-demand and easily scalable manner. Such services can be related to IT and software, the Internet, or other services. Cloud computing is the product of the integration of the development of traditional computer and network technologies such as grid computing, distributed computing, parallel computing, utility computing, network storage technologies, virtualization, and load balancing.

[0219] Cloud computing has rapidly grown, driven by the internet, real-time data streams, the diversification of connected devices, and the growing demand for search services, social networks, mobile commerce, and open collaboration. Unlike previous parallel and distributed computing approaches, the emergence of cloud computing will fundamentally revolutionize the entire internet and enterprise management model.

[0220] The user behavior pattern mining method provided in this application can realize the automatic mining of user behavior patterns based on the user's behavior indicator data. This solution can also be implemented through artificial intelligence cloud services, which are generally referred to as AIaaS (AI as a Service, Chinese for "AI as a Service"). This is a mainstream service mode of artificial intelligence platforms. Specifically, the AIaaS platform will split several common AI services and provide independent or packaged services in the cloud. This service model is similar to opening an AI theme mall: all developers can access and use one or more artificial intelligence services provided by the platform through the API interface. Some senior developers can also use the AI ​​framework and AI infrastructure provided by the platform to deploy and operate their own exclusive cloud artificial intelligence services. In this application, the AI ​​framework and AI infrastructure provided by the platform can be used to implement the user behavior pattern mining method provided in this application.

[0221] An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer-readable storage medium is run on a computer, the computer can execute the corresponding contents of the aforementioned method embodiment.

[0222] It should be understood that although the steps in the flowcharts of the accompanying drawings are shown in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the flowcharts of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0223] It should be noted that the computer-readable medium mentioned above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. The computer-readable storage medium can 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 can 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 this application, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. This propagated data signal can take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0224] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0225] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device executes the method shown in the above embodiment.

[0226] According to one aspect of the present application, a computer program product or computer program is provided, comprising 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 user behavior pattern mining method provided in the various optional implementations described above.

[0227] Computer program code for performing the operations of the present application may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0228] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of 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 than that 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 and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0229] The modules described in the embodiments of this application may be implemented in software or hardware. The name of a module does not necessarily limit the module itself. For example, a user behavior pattern mining module may be described as a "module for mining user behavior patterns based on various feature indicator data."

[0230] The above description is merely a preferred embodiment of the present application and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this application is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but also encompasses other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the above-mentioned disclosed concepts. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A method for mining user behavior patterns, characterized in that: include: Acquire multiple behavior indicator data of the user, wherein the behavior indicator data includes time information and behavior data indicator values ​​corresponding to the time information; Determining, based on time series information corresponding to the plurality of behavior indicator data, a characteristic value in each of the behavior data indicator values, and using the behavior indicator data corresponding to each of the characteristic values ​​as characteristic indicator data, including: for any of the behavior data indicator values, if the behavior data indicator value is greater than its previous behavior data indicator value and greater than its next behavior data indicator value, determining the behavior data indicator value as a characteristic point; for any of the characteristic points, determining, based on the characteristic point and behavior data indicator values ​​temporally adjacent to the characteristic point, a rate of change of the indicator value of the characteristic point; and determining, among the characteristic points, each characteristic point whose rate of change of the indicator value is greater than a set value as the characteristic indicator data; Based on each of the characteristic indicator data, the user behavior pattern of the user is mined, including: based on the time series information of each of the characteristic indicator data, determining the time difference between temporally adjacent characteristic indicator data, and determining the time series characteristics of the user behavior based on the time difference; based on the time information of each of the characteristic indicator data, calculating the indicator difference between the characteristic values ​​corresponding to temporally adjacent characteristic indicator data, and determining the behavioral characteristics of the user behavior based on the indicator difference; mining the user behavior pattern based on the time series characteristics and the behavioral characteristics.

2. The method according to claim 1, characterized in that The plurality of behavior indicator data include behavior indicator data of the user corresponding to at least two information categories of the application; The determining, based on the time series information corresponding to the plurality of behavior indicator data, a characteristic value in each of the behavior data indicator values, and using the behavior indicator data corresponding to each of the characteristic values ​​as characteristic indicator data, includes: Dividing the plurality of behavior indicator data according to information categories to obtain a plurality of behavior indicator data corresponding to each information category; For each information category, based on the time series information corresponding to the plurality of behavior indicator data of the information category, determine the characteristic values ​​in the indicator values ​​of the behavior data corresponding to the information category, and use the characteristic values ​​corresponding to the information category as the characteristic indicator data corresponding to the information category; Mining the user behavior pattern of the user based on each of the characteristic indicator data includes: For each information category, mining the user behavior pattern of the user corresponding to the information category based on each characteristic indicator data of the information category, wherein the user behavior pattern includes a time series pattern; the method further includes: Based on the usage time of the application by the user, at least one timing pattern matching the usage time is determined; based on the determined timing pattern and the usage time, relevant information of the information category corresponding to the determined timing pattern is recommended to the user.

3. The method according to claim 1, characterized in that The determining, based on the time series information corresponding to the plurality of behavior indicator data, a characteristic value in each of the behavior data indicator values ​​includes: Based on the time information, the plurality of behavior indicator data are divided to obtain behavior indicator data corresponding to a plurality of time periods; For each time period, determining a characteristic value in each behavior data indicator value of each time period based on the time series information corresponding to each behavior indicator data belonging to the time period; The characteristic value is the maximum value or the minimum value among the behavior data indicator values ​​corresponding to the plurality of behavior indicator data, and the characteristic indicator data includes each of the characteristic values ​​and the time information corresponding to each of the characteristic values; Mining the user behavior pattern of the user based on each of the characteristic indicator data includes: Based on the characteristic indicator data corresponding to each of the time periods, the user behavior pattern of the user is mined.

4. The method according to any one of claims 1 to 3, characterized in that The characteristic value corresponding to the characteristic indicator data is a local peak value of several adjacent time periods, and the behavioral characteristic of the user behavior is the execution peak period of the specified operation appearing in the time information of the characteristic indicator data.

5. The method according to claim 1, wherein The plurality of behavior indicator data includes behavior indicator data corresponding to a plurality of time periods, the characteristic indicator data includes characteristic indicator data corresponding to each time period, and a time period includes at least one characteristic indicator data; when each time period corresponds to one characteristic indicator data, the user behavior pattern is mined using the same number of characteristic indicator data as the time period; The determining of the time difference between temporally adjacent feature indicator data based on the time series information of each feature indicator data includes: For every two adjacent time periods, determine the time difference between the characteristic indicator data belonging to different periods in the two time periods; The calculating the index difference between temporally adjacent feature values ​​based on the time information of each feature index data includes: For every two adjacent time periods, the index difference between the characteristic values ​​belonging to different periods in the two time periods is determined.

6. The method according to claim 1, characterized in that The mining of the user behavior pattern based on the time series features and the behavior features includes: Performing clustering processing on the time difference and / or the indicator difference to obtain a clustering result; A user behavior pattern is determined based on the clustering result.

7. The method according to claim 1, characterized in that The indicator value change rate is a fixed value or a dynamic value, and is used to adjust the quantity and position of characteristic indicator data when the indicator value change rate is configured as a dynamic value.

8. The method according to claim 1, characterized in that The obtaining of user behavior indicator data includes: Obtaining a user's behavior pattern analysis request, wherein the analysis request includes a specified time period and / or a specified operation type; Based on the analysis request, user behavior indicator data corresponding to the specified time period and / or the specified operation type is obtained.

9. The method according to claim 1, characterized in that The behavior indicator data is the behavior indicator data of the user corresponding to the application, and the acquiring of multiple behavior indicator data of the user includes: Based on the user identification of the user, a plurality of behavior indicator data of the user is obtained from a data warehouse of the application, wherein the data warehouse is used to store the user identification of each user and the behavior indicator data corresponding to each user in an associated manner.

10. A recommendation method based on user behavior patterns, characterized in that: include: Acquire a plurality of behavior indicator data of a user, wherein the plurality of behavior indicator data includes behavior indicator data of the user corresponding to at least two information categories of an application; Dividing the plurality of behavior indicator data according to information categories to obtain a plurality of behavior indicator data corresponding to each information category; For each information category, based on the time series information corresponding to the plurality of behavior indicator data of the information category, determine the characteristic values ​​in the indicator values ​​of the behavior data corresponding to the information category, and use the characteristic values ​​corresponding to the information category as the characteristic indicator data corresponding to the information category; For each information category, using the user behavior pattern mining method according to any one of claims 1 to 9, mining the user behavior pattern of the user corresponding to the information category based on each characteristic indicator data of the information category, wherein the user behavior pattern includes a time series pattern; Based on the usage time of the application by the user, at least one timing pattern matching the usage time is determined; based on the determined timing pattern and the usage time, relevant information of the information category corresponding to the determined timing pattern is recommended to the user.

11. A device for mining user behavior patterns, characterized in that: include: A behavior indicator data acquisition module is used to acquire multiple behavior indicator data of the user, wherein the behavior indicator data includes time information and behavior data indicator values ​​corresponding to the time information; A characteristic indicator data determination module is configured to determine, based on time series information corresponding to a plurality of behavior indicator data, a characteristic value in each behavior data indicator value, and use the behavior indicator data corresponding to each characteristic value as the characteristic indicator data, including: for any behavior data indicator value, if the behavior data indicator value is greater than its previous behavior data indicator value and greater than its next behavior data indicator value, then determining the behavior data indicator value as a characteristic point; for any characteristic point, based on the characteristic point and the behavior data indicator values ​​adjacent to the characteristic point in time, determining the indicator value change rate of the characteristic point; and determining each characteristic point whose indicator value change rate is greater than a set value among the characteristic points as the characteristic indicator data; The user behavior pattern mining module is used to mine the user behavior pattern of the user based on each feature indicator data, including: determining the time difference between temporally adjacent feature indicator data based on the time series information of each feature indicator data, and determining the temporal characteristics of the user behavior based on the time difference; calculating the indicator difference between the characteristic values ​​corresponding to temporally adjacent feature indicator data based on the time information of each feature indicator data, and determining the behavioral characteristics of the user behavior based on the indicator difference; mining the user behavior pattern based on the temporal characteristics and the behavioral characteristics.

12. The device according to claim 11, characterized in that The plurality of behavior indicator data include behavior indicator data of the user corresponding to at least two information categories of the application; The characteristic indicator data determination module is specifically used to: Dividing the plurality of behavior indicator data according to information categories to obtain a plurality of behavior indicator data corresponding to each information category; For each information category, based on the time series information corresponding to the plurality of behavior indicator data of the information category, determine the characteristic values ​​in the indicator values ​​of the behavior data corresponding to the information category, and use the characteristic values ​​corresponding to the information category as the characteristic indicator data corresponding to the information category; User behavior pattern mining module, specifically used for: For each information category, mining the user behavior pattern corresponding to the information category based on the characteristic indicator data of the information category, the user behavior pattern includes a time series pattern; A device for mining user behavior patterns, including a recommendation module; The recommendation module is used to determine at least one timing pattern that matches the usage time based on the user's usage time of the application, and recommend relevant information of the information category corresponding to the determined timing pattern to the user based on the determined timing pattern and the usage time.

13. The device according to claim 11, characterized in that User behavior pattern mining module, specifically used for: Based on the time information, the plurality of behavior indicator data are divided to obtain behavior indicator data corresponding to a plurality of time periods; For each time period, determining a characteristic value in each behavior data indicator value of each time period based on the time series information corresponding to each behavior indicator data belonging to the time period; The characteristic value is the maximum value or the minimum value among the behavior data indicator values ​​corresponding to the plurality of behavior indicator data, and the characteristic indicator data includes each of the characteristic values ​​and the time information corresponding to each of the characteristic values; The user behavior pattern mining module is specifically used to: Based on the characteristic indicator data corresponding to each of the time periods, the user behavior pattern of the user is mined.

14. The device according to any one of claims 11 to 13, characterized in that The characteristic value corresponding to the characteristic indicator data is a local peak value of several adjacent time periods, and the behavioral characteristic of the user behavior is the execution peak period of the specified operation appearing in the time information of the characteristic indicator data.

15. The device according to claim 11, characterized in that The plurality of behavior indicator data includes behavior indicator data corresponding to a plurality of time periods, the characteristic indicator data includes characteristic indicator data corresponding to each time period, and a time period includes at least one characteristic indicator data; when each time period corresponds to one characteristic indicator data, the user behavior pattern is mined using the same number of characteristic indicator data as the time period; The user behavior pattern mining module is also used to: For every two adjacent time periods, determine the time difference between the characteristic indicator data belonging to different periods in the two time periods; The calculating the index difference between temporally adjacent feature values ​​based on the time information of each feature index data includes: For every two adjacent time periods, the index difference between the characteristic values ​​belonging to different periods in the two time periods is determined.

16. The device according to claim 11, characterized in that The user behavior pattern mining module is also used to: Performing clustering processing on the time difference and / or the indicator difference to obtain a clustering result; A user behavior pattern is determined based on the clustering result.

17. The device according to claim 11, characterized in that The indicator value change rate is a fixed value or a dynamic value, and is used to adjust the quantity and position of characteristic indicator data when the indicator value change rate is configured as a dynamic value.

18. The device according to claim 11, characterized in that The behavioral indicator data acquisition module is specifically used to: Obtaining a user's behavior pattern analysis request, wherein the analysis request includes a specified time period and / or a specified operation type; Based on the analysis request, user behavior indicator data corresponding to the specified time period and / or the specified operation type is obtained.

19. The device according to claim 11, characterized in that The behavior indicator data is the user's behavior indicator data corresponding to the application. The behavior indicator data acquisition module is specifically used to: Based on the user identification of the user, a plurality of behavior indicator data of the user is obtained from a data warehouse of the application, wherein the data warehouse is used to store the user identification of each user and the behavior indicator data corresponding to each user in an associated manner.

20. A recommendation device based on user behavior patterns, characterized in that include: A second behavior indicator data acquisition module is used to acquire multiple behavior indicator data of the user, wherein the multiple behavior indicator data include behavior indicator data of the user corresponding to at least two information categories of the application; A division module, for dividing the plurality of behavior indicator data according to information categories, and obtaining a plurality of behavior indicator data corresponding to each information category; A second characteristic indicator data determination module is configured to determine, for each information category, characteristic values ​​among the behavior data indicator values ​​corresponding to the information category based on time series information corresponding to the plurality of behavior indicator data of the information category, and use the characteristic values ​​corresponding to the information category as characteristic indicator data corresponding to the information category; a second user behavior pattern mining module, configured to, for each information category, mine the user behavior pattern of the user corresponding to the information category based on each characteristic indicator data of the information category using the user behavior pattern mining device according to any one of claims 11 to 19, wherein the user behavior pattern includes a time series pattern; The second recommendation module is used to determine at least one timing pattern that matches the usage time based on the user's usage time of the application, and recommend relevant information of the information category corresponding to the determined timing pattern to the user based on the determined timing pattern and the usage time.

21. An electronic device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method according to any one of claims 1 to 10 is implemented.

22. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 10 is 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