User behavior path construction method and device, and storage medium

Abstract

The invention discloses a user behavior path construction method and device and a storage medium, and relates to the technical field of digital processing. According to the method, a target behavior event is determined based on preset configuration information, an initial behavior path of a user is constructed based on the target behavior event, the preset configuration information comprises a user account corresponding to the behavior event, then the initial path of each user is divided into path segments based on the preset layer number, each path segment corresponds to one layer, and the user behavior path is constructed based on the path segments. And finally, summarizing the path segments of each user according to the hierarchy based on the corresponding relationship between the path segments and the hierarchy, and constructing a hierarchical path of the user. According to the method, the complexity of the user behavior path can be reduced, the effective behavior path of each user is aggregated according to the hierarchy, and the key path is displayed, so that an operator can concentrate on the user behavior path which is more likely to generate value, and the accuracy of user behavior path analysis is further improved.

Description

Technical Field

[0001] The present application relates to the field of digital processing technology, and in particular to a method, device and storage medium for constructing a user behavior path. Background Art

[0002] With the rapid development of the IPTV (Interactive Personal Television) industry over the years, user growth has now entered a bottleneck period and is no longer in the bonus stage of rapid user growth. At present, only statistics on the number of clicks, exposures, playbacks, subscriptions, and other behaviors, such as the number of people, times, and duration, are not enough to fully reflect the real behavior and preferences of users, nor can they effectively guide operational decisions. Instead, other analytical methods are needed to identify the most frequently visited paths and the links that are most likely to lose users, so as to improve operational results.

[0003] At present, in the user behavior path analysis, the online time, that is, all the behavior events from power on to power off, are generally connected as nodes in chronological order to represent a user behavior path, and then the behavior path of each user is analyzed. However, the behavior path of IPTV users is usually long, because users may watch multiple programs, browse different menus, search for content, and even interact socially in a period of time. Most users will also carry out some preparation stages that cannot reflect user behavior patterns when starting IPTV services, such as selecting language, logging in to an account, browsing recommended content, etc. Therefore, the user behavior path constructed by the above method in the IPTV industry is too long and contains many irrelevant events. It is difficult for operators to find valuable path nodes from multiple complex user behavior paths, which will reduce the accuracy of user behavior analysis.

[0004] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of this application is to provide a method, device and storage medium for constructing a user behavior path, aiming to solve the technical problem of low accuracy of user behavior analysis caused by the path constructed by the traditional behavior path construction method being too long and complicated.

[0006] To achieve the above objectives, the present application proposes a method for constructing a user behavior path, the method comprising: Determine a target behavior event based on preset configuration information, and construct an initial behavior path of the user based on the target behavior event, wherein the preset configuration information includes a user account corresponding to the behavior event; Based on a preset number of layers, split the initial path of each user into path segments, each path segment corresponding to a layer; Based on the correspondence between the path segments and the levels, the path segments of each user are aggregated according to the levels to construct a hierarchical path for the user.

[0007] In one embodiment, the step of determining a target behavior event based on preset configuration information and constructing an initial behavior path of the user based on the target behavior event includes: Determining the target behavior event corresponding to the user based on the user account; Based on the start time of the target behavior event, sort each of the target behavior events to obtain a user's behavior event sequence; Based on the behavior event sequence of each user, the target behavior event is used as a node to construct the initial behavior path of the user.

[0008] In one embodiment, the step of aggregating the path segments of each user according to the hierarchy based on the correspondence between the path segments and the hierarchy to construct the hierarchical path of the user includes: Determine a sequence label of the path segment in the behavior event sequence, wherein the sequence label corresponds one-to-one to the order of the hierarchy; Based on the sequence label, traverse the path segment of each user to determine the level corresponding to the path segment; The path segments of each user are allocated to the corresponding levels to construct the hierarchical path of the user.

[0009] In one embodiment, the step of allocating the path segments of each user to the corresponding hierarchy and constructing the hierarchical path of the user comprises: Determine the number of occurrences of the path segment in the same level, and obtain the number of sessions of the path segment; Acquire a preset number of path segments based on the preset configuration information, where the preset number of path segments represents a threshold value of the number of the path segments displayed in each level; Based on the number of sessions of the path segment and the preset number of path segments, the valid path segments that need to be displayed in each level are determined to obtain the hierarchical path of the user.

[0010] In one embodiment, the step of constructing the initial behavior path of the user based on the behavior event sequence of each user and taking the target behavior event as a node includes: Based on the preset configuration information, obtaining preset time ranges corresponding to different behavior event types; If the duration of the target behavior event is outside the preset time range, deleting the target behavior event from the behavior event sequence of the user; Based on the behavior event sequence of each user, the remaining target behavior events are used as nodes to construct the initial behavior path of the user.

[0011] In one embodiment, the step of constructing the initial behavior path of the user based on the behavior event sequence of each user and taking the target behavior event as a node further includes: Determining the number of behavior event types in the behavior event sequence of each of the users within a preset time window; If the number of the behavior event types exceeds a preset type threshold, deleting the behavior event sequence of the user; Based on the remaining behavior event sequences of each of the users, the target behavior event is used as a node to construct the initial behavior path of the user.

[0012] In one embodiment, the step of constructing the initial behavior path of the user based on the behavior event sequence of each user and taking the target behavior event as a node further includes: Acquire a path window period based on the preset configuration information, wherein the path window period is used to determine a maximum interval duration between two behavior events in a behavior path; Determine the time difference between the end time of the preceding behavior event and the start time of the succeeding behavior event in the behavior event sequence of each user; If the time difference is greater than the path window period, the user's behavior event sequence is decomposed into two initial behavior paths based on the preceding event and the succeeding event.

[0013] In one embodiment, before the step of splitting the initial path of each user into path segments based on a preset number of layers, the method further includes: Determine a start event and an end event in the target behavior event based on the preset configuration information; Determining a retention length of the initial behavior path based on the preset number of layers; The initial behavior path is truncated based on the retained length to obtain a new initial behavior path, wherein the truncation position is located after the start event and / or before the end event.

[0014] In addition, to achieve the above-mentioned purpose, the present application also proposes a device for constructing a user behavior path, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the method for constructing a user behavior path as described above.

[0015] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the method for constructing the user behavior path as described above are implemented.

[0016] The present application provides a method for constructing a user behavior path. The method first determines a target behavior event based on preset configuration information, and constructs the user's initial behavior path based on the target behavior event. The preset configuration information includes a user account corresponding to the behavior event. Then, based on a preset number of layers, the initial path of each user is split into path segments, where each path segment corresponds to a layer. Finally, based on the correspondence between the path segments and the layers, the path segments of each user are summarized according to the layers to construct the user's hierarchical path.

[0017] This method determines the target behavior event through preset configuration information, screens out key events related to user behavior patterns, constructs the user's initial behavior path, and then splits these paths into shorter and easier-to-analyze path segments based on the preset number of layers. Based on the correspondence between path segments and layers, the path segments of each user are summarized by layer to construct the user's hierarchical path. This hierarchical method allows operators to see the user's behavior patterns at different levels more clearly, thereby more accurately identifying the user's most frequently visited paths and the links that are most likely to churn, thereby improving the accuracy of user behavior analysis and helping operators provide more targeted guidance for operational decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0020] Figure 1 A flowchart diagram of the first embodiment of the method for constructing a user behavior path of the present application; Figure 2 A flowchart diagram of the second embodiment of the method for constructing a user behavior path of the present application; Figure 3 A flowchart diagram of the third embodiment of the method for constructing a user behavior path of the present application; Figure 4A flowchart diagram of a fourth embodiment of a method for constructing a user behavior path of the present application; Figure 5 A schematic diagram of a method flow for constructing a user behavior path in an embodiment of the present application; Figure 6 A schematic diagram of the device structure of the hardware operating environment involved in the method for constructing a user behavior path in an embodiment of the present application.

[0021] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0022] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0023] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0024] After experiencing rapid growth, the IPTV industry has now entered a bottleneck period of user growth and is no longer enjoying the dividends of rapid user growth. In this case, traditional user behavior indicators such as the number of clicks, exposures, plays, and subscriptions, times, and duration are no longer sufficient to fully reflect user behavior and preferences, and cannot effectively guide operational decisions. Therefore, more advanced analysis methods are needed to identify the most frequently visited paths and the links that are most likely to churn, in order to improve operational results.

[0025] At present, in the user behavior path analysis, the online time, that is, all the behavior events from power on to power off, are generally connected as nodes in chronological order to represent a user behavior path, and then the behavior path of each user is analyzed. However, the behavior path of IPTV users is usually long, because users may watch multiple programs, browse different menus, search for content, and even interact socially in a period of time. Most users will also carry out some preparation stages that cannot reflect user behavior patterns when starting IPTV services, such as selecting language, logging in to an account, browsing recommended content, etc. Therefore, the user behavior path constructed by the above method in the IPTV industry is too long and contains many irrelevant events. It is difficult for operators to find valuable path nodes from multiple complex user behavior paths, which will reduce the accuracy of user behavior analysis.

[0026] In view of the above problems, the present application proposes a method for constructing a user behavior path. First, the target behavior event is determined based on preset configuration information, and the user's initial behavior path is constructed based on the target behavior event. The preset configuration information includes the user account corresponding to the behavior event. Then, based on the preset number of layers, each user's initial path is split into path segments, where each path segment corresponds to a level. Finally, based on the correspondence between the path segments and the levels, each user's path segments are summarized according to the levels to construct the user's hierarchical path.

[0027] The above method determines the target behavior event through preset configuration information, screens out key events related to the user behavior pattern, constructs the user's initial behavior path, and then splits these paths into shorter and easier to analyze path segments based on the preset number of layers. Based on the correspondence between the path segments and the levels, the path segments of each user are summarized by level to construct the user's hierarchical path. This hierarchical approach allows operators to see the user's behavior patterns at different levels more clearly, thereby more accurately identifying the user's most frequently visited paths and the links that are most likely to churn, thereby improving the accuracy of user behavior analysis and helping operators provide more targeted guidance for operational decisions.

[0028] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, etc. The following takes the construction device of the user behavior path as an example to illustrate this embodiment and the following embodiments.

[0029] Based on this, the first embodiment proposed in this application provides a method for constructing a user behavior path. Figure 1 In this embodiment, the method for constructing the user behavior path includes steps S10 to S30: Step S10, determining a target behavior event based on preset configuration information, and constructing an initial behavior path of the user based on the target behavior event, wherein the preset configuration information includes a user account corresponding to the behavior event.

[0030] It should be noted that the preset configuration information is some information about the behavioral events that need to participate in the path analysis, such as the user account corresponding to the behavioral event, the type of behavioral event, and the reasonable time range corresponding to different event types. Under the same type setting, dimension information can be added. For example, under the same "watch live" type of behavioral event, there can be two dimensional behavioral events of "watch live channel A" and "watch live channel B". The start event type and the end event type can also be specified. The user account is the unique identifier used by the user in the IPTV system to record the user's behavior and attributes. Behavioral events refer to various operations or activities performed by users in the IPTV system, such as exposure, clicks, browsing, playback (live, on-demand, replay, etc.), purchase, ordering, etc. These operations or activities are recorded to form user behavior data. The initial behavioral path refers to the event chain formed by connecting the target behavioral events in chronological order after filtering out the target behavioral events that need to be analyzed based on the user account, reflecting the user's behavioral trajectory in the IPTV system. The initial behavioral path is classified and divided according to the account information of the IPTV user to facilitate subsequent data analysis and processing.

[0031] First, collect user behavior data from the IPTV system database, including user account, behavior event type, event occurrence time, and other information. Filter target behavior events from the database based on user accounts and group them by user accounts. For each group, sort the target behavior events in chronological order based on the timestamps of the target behavior events. Connect the sorted target behavior events to form the user's initial behavior path. Each path contains the user's behavior trajectory within a certain period of time.

[0032] Optionally, user behavior event data can be collected from the IPTV platform through probe embedding. Specifically, data collection tools (probes) are pre-set at various functional points or key pages of the IPTV platform. When users perform corresponding operations such as clicking a button or watching a video, these probes will capture and record the user's behavior data. Alternatively, user behavior data can be obtained from other third-party systems. These third-party systems can provide other services or information related to the IPTV platform, which can be used as a reference for user behavior path analysis.

[0033] Optionally, the user behavior event data and user account information collected from the IPTV platform are imported into ClickHouse, and ad hoc queries are performed through ClickHouse to obtain the target behavior events. ClickHouse is an open source, high-performance column-based database management system for online analysis and processing queries. When selecting the user's target behavior event, different dimensions are added for analysis according to actual needs. For example, if exposure data is selected as the analysis object and the regional dimension is added, the behavior paths of users in different regions before or after the exposure behavior can be observed. This cross-dimensional analysis method can more comprehensively reveal the user's behavioral characteristics and provide operators with more accurate operational strategy recommendations.

[0034] Optionally, in addition to grouping by user account, the target behavior events corresponding to the user account can be further grouped according to the type of behavior event, such as watching programs, clicking advertisements, etc., and an initial behavior path can be generated to clearly reflect the user's preference and participation in different types of behavior events. Alternatively, different user accounts can be further grouped according to the attributes of the user account, such as region, age, gender, etc., and an initial behavior path can be generated to analyze the behavioral differences of different user groups in the IPTV system.

[0035] Optionally, step S10 further includes steps S11 to S13: Step S11, determining the target behavior event corresponding to the user based on the user account.

[0036] Step S12: sort each of the target behavior events based on the start time of the target behavior event to obtain a user's behavior event sequence.

[0037] Step S13, based on the behavior event sequence of each user, the target behavior event is used as a node to construct the initial behavior path of the user.

[0038] Exemplarily, for a set of target behavior events that have been grouped according to user accounts, the specific time points at which the target behavior events in each group occur are obtained to determine the sequence of events, and the user's target behavior events are arranged in chronological order to form a user's behavior event sequence. Each behavior event is taken as a node in the path and connected in chronological order to form the user's initial behavior path.

[0039] Step S20: split the initial path of each user into path segments based on a preset number of layers, each path segment corresponding to a layer.

[0040] When analyzing user behavior paths, in order to simplify the analysis process or improve analysis efficiency, a display level limit, namely the preset number of levels, is pre-set. The preset number of levels determines which level of behavior events will be displayed when displaying the user behavior path. Based on the preset number of levels, the initial behavior path is split to obtain several path segments.

[0041] Optionally, a path segment may be composed of two adjacent target behavior events, and each path segment corresponds to a level.

[0042] For example, a display level limit is preset according to the analysis requirements. This limit can be a fixed value or a dynamic value calculated based on a rule or algorithm, such as determined based on the average number of behavioral events of all users. For each user's initial behavior path, the entire path is traversed in chronological order, and during the traversal process, the initial behavior path is split according to the number of layers.

[0043] To better understand the solution provided in this example, this example is further explained in combination with specific application scenarios.

[0044] Assuming that the preset number of layers is 6, and the initial behavior path of a certain user is: A->B->C->D->E->F->G, the above initial behavior path is split into the following 5 path segments: A->B, B->C, C->D, D->E, E->F, F->G. Optionally, the preset number of layers and the number of behavior events in the path segment can be adjusted according to user needs. For example, assuming that the preset number of layers is 3, the initial behavior path of the above user can be split into the following 3 path segments: A->B->C, C->D->E, E->F->G.

[0045] Step S30: Based on the correspondence between the path segments and the levels, the path segments of each user are aggregated according to the levels to construct a hierarchical path for the user.

[0046] In IPTV user behavior analysis, although the behavior path of a single user provides direct information about the user's viewing habits, its value is often limited because a single path may only reflect the user's viewing choices in a specific time period and is not universal. In order to understand user behavior more deeply, the paths of multiple users can be aggregated and analyzed. In step S20, the initial behavior path of each user has been truncated based on the preset number of layers. Then, these path segments are assigned to their corresponding levels, and the path segments of all users are summarized and analyzed at each level to construct the user's hierarchical path.

[0047] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can refer to the above introduction, and will not be repeated later. Figure 2 , step S30 includes steps S31 to S33: Step S31, determining the sequence labels of the path segments in the behavior event sequence, wherein the sequence labels correspond one-to-one to the order of the levels.

[0048] Step S32: traverse the path segment of each user based on the sequence label to determine the level corresponding to the path segment.

[0049] Exemplarily, first, define a sequence label for each path segment in the behavior event sequence, indicating the position of the path segment in the behavior event sequence. For each user's behavior path, starting from the first path segment, sequential labels are assigned in the order in which they appear. For each user, start traversing all of its path segments, and determine the level to which each path segment belongs based on the sequence label. Among them, each sequence label corresponds to the order of the levels one by one, that is, the nth path segment of each user corresponds to the nth level.

[0050] Step S33: Allocate the path segment of each user to the corresponding hierarchy to construct the hierarchical path of the user.

[0051] For the currently traversed path segment, it is added to the corresponding level according to its sequence label. Finally, each user's initial behavior path is converted into a hierarchical representation, and the path segments in each level are summarized to construct the hierarchical path of all users.

[0052] Exemplarily, it is assumed that the initial path of the user obtained after step S10 is as follows: Path 1 (User 1): A->B->C->D->E; Path 2 (User 2): A->B->C->E->F; Path 3 (User 3): A->B->C->D->T; Path 4 (User 4): A->B->C->D->E; Path 5 (User 5): F->G->H->E->F.

[0053] After step S20, the user's path segment is obtained as follows: User 1: A->B, B->C, C->D, D->E; User 2: A->B, B->C, C->E, E->F; User 3: A->B, B->C, C->D, D->T; User 4: A->B, B->C, C->D, D->T; User 5: F->G, G->H, H->E, E->F.

[0054] The hierarchical path of the user obtained after step S30 is as follows: First layer: A->B, A->B, A->B, A->B, F->G; Second layer: B->C, B->C, B->C, B->C, G->H; The third layer: C->D, C->E, C->D, C->D, H->E; Fourth layer: D->E, E->F, D->T, D->T, E->F.

[0055] Optionally, step S33 also includes steps S331 to S333: Step S331, determining the number of occurrences of the path segment in the same level, and obtaining the number of sessions of the path segment.

[0056] It should be noted that the number of sessions is used to indicate the number of users participating in a certain path segment at a certain level. A path segment with a large number of sessions indicates that the path segment is frequently visited by users, and a path segment with a small number of sessions indicates that the path segment is prone to user loss.

[0057] Taking the above example as an example, the number of sessions for each path segment at each level is as follows, where the number in brackets after the path segment indicates the number of sessions for that path segment: First layer: A->B (4), F->G (1); Second layer: B->C (4), G->H (1); Third layer: C->D (3), C->E (1), H->F (1); Fourth layer: D->E (1), E->F (2), D->T (2).

[0058] Step S332: obtaining a preset number of path segments based on the preset configuration information, wherein the preset number of path segments represents a threshold value of the number of the path segments displayed in each level.

[0059] Step S333: Based on the number of sessions of the path segment and the preset number of path segments, determine the valid path segment that needs to be displayed in each level to obtain the hierarchical path of the user.

[0060] It should be noted that the preset configuration information also includes some rules for constructing the hierarchical path of the user, such as the number of levels of the hierarchical path, the number of path segments to be displayed in each layer, etc. Among them, the preset number of path segments is a pre-set limit on the number of path segments that need to be displayed in each layer when analyzing the user's behavior path in order to simplify the analysis process or improve the analysis efficiency. For example, if the preset number of path segments is 1, it means that only one path segment is selected for display in each layer of the user's hierarchical path.

[0061] Exemplarily, the path segments are sorted according to the number of sessions, and the path segment with the largest number of sessions is selected for display. Taking the above example, the hierarchical path of the user is obtained: First layer: A->B; Second layer: B->C; Third layer: C->D; Fourth layer: D->E.

[0062] Optionally, when there are multiple preset display nodes and there are multiple path segments with the same number of sessions, multiple path segments with fewer sessions can be combined into one path segment for display to obtain the user's hierarchical path. Taking the above example, assuming that the number of preset path segments is 2: First layer: A->B, F->G; Second layer: B->C, G->H; The third layer: C->D, {C->E, H->F}; Fourth layer: E->F, D->T.

[0063] Optionally, a weight is set for a specific type of behavior event in the preset configuration information, and the path segments to be displayed in each level are determined based on the weight of the behavior event in the path segment, the number of sessions of the path segment, and the preset number of path segments to obtain the hierarchical path of the user. For example, the path segments are first sorted based on the number of sessions. If there are multiple path segments with the same number of sessions, these path segments with the same number of sessions are sorted based on the size of the event weight, and the first preset number of path segments are selected for display to obtain the hierarchical path of the user.

[0064] Optionally, after obtaining the user's hierarchical path, the number of lost sessions in each path segment in each layer can be calculated by counting the number of sessions in all path segments in the current layer and the number of sessions in all path segments in the next layer. Operations personnel can analyze the behavioral event nodes or reasons for user loss based on the number of lost sessions combined with other analysis models.

[0065] Based on the above embodiments of the present application, in the third embodiment of the present application, the same or similar contents as those in the above embodiments can be referred to the above introduction, and will not be described in detail later. Figure 3 , step S13 also includes steps S131 to S133: Step S131, based on the preset configuration information, obtaining preset time ranges corresponding to different behavior event types.

[0066] It should be noted that the preset configuration information also includes preset time ranges corresponding to different event types, where the preset time range is a reasonable time threshold set for each behavior event type, which is used to determine whether the behavior event is valid or worth recording. The preset time range can be determined based on business logic and user behavior patterns. For example, the reasonable time range for watching a video may be a few minutes to a few hours, while the reasonable time range for clicking an advertisement may be a few seconds.

[0067] For each target behavior event in the user behavior event sequence, parse the timestamp of each target behavior event, determine the start and end time of the target behavior event, and calculate its duration.

[0068] Step S132: if the duration of the target behavior event is outside the preset time range, the target behavior event is deleted from the behavior event sequence of the user.

[0069] Step S133, based on the behavior event sequence of each user, the remaining target behavior events are used as nodes to construct the initial behavior path of the user.

[0070] The duration of each target behavior event is compared with the corresponding preset time range. If the duration of the target behavior event is outside the preset time range, such as too short or too long, it may indicate an incorrect operation or abnormal behavior, and the target behavior event is deleted from the user behavior event sequence. Based on the filtered valid behavior event sequence, each target behavior event is used as a node and connected in chronological order to form the user's initial behavior path.

[0071] For example, suppose there is an online video platform where users' behaviors such as watching videos, browsing recommended content, and clicking on ads will be recorded. First, a reasonable preset time range is set for each behavior event type. The preset time range for watching videos is 5 minutes to 2 hours, the preset time range for browsing recommended content is 10 seconds to 5 minutes, and the preset time range for clicking on ads is 1 to 5 seconds.

[0072] Suppose a user's behavior event sequence is as follows: Watch video A (start time: 08-00-00, end time: 09-15-00); Browse recommended content (start time: 09-36-00, end time: 09-38-00); Click on the ad (start time: 09-38-03, end time: 09-38-06); Watch video B (start time: 09-39-00, end time: 09-39-10); Watch Video C (start time: 10-30-00, end time: 11-15-00).

[0073] Next, the duration of each target behavior event is compared with the corresponding preset time range. The duration of watching video A and watching video C is 1 hour and 15 minutes and 45 minutes respectively, which are valid within the preset time range. The duration of browsing recommended content and clicking ads is also within the preset time range and is valid. The duration of watching video B is only 10 seconds, which is not within the preset time range and may be an error operation, so it is deleted from the behavior event sequence.

[0074] Finally, based on the filtered target behavior event sequence, we generate the user's initial behavior path: watching video A->browsing recommended content->clicking on an ad->watching video C. In the actual user's initial behavior path, each node can also include more information, such as event type, content description, user account, etc.

[0075] Optionally, step S13 further includes steps S134 to S136: Step S134, determining the number of behavior event types in the behavior event sequence of each user within a preset time window.

[0076] Step S135: if the number of the behavior event types exceeds a preset type threshold, the behavior event sequence of the user is deleted.

[0077] Step S136, based on the remaining behavior event sequences of each of the users, the target behavior event is used as a node to construct the initial behavior path of the user.

[0078] First, set a preset time window, such as one day, one hour, or any other time period, to analyze the user's behavior within that time period. Traverse the event sequence, determine whether the event belongs to the preset time window based on the timestamp of the event, and count different types of events. Compare the number of behavioral event types obtained with the preset type threshold. If the number of types exceeds the threshold, it is considered that the user's viewing behavior changes frequently or there are operations such as frequent channel switching, so it is deleted. For each remaining behavioral event sequence, take each target behavioral event as a node and connect them in chronological order to form the user's initial behavior path.

[0079] For example, assuming that the preset type threshold within a preset window, such as within 1 hour, is 5, a user's behavior event sequence is collected as follows: Open the IPTV app -> browse the program list -> watch live program A -> pause live program A -> click on an ad -> post a comment -> switch the channel to live program B -> return to watching live program A -> watch program C on demand -> search for content D -> close the IPTV app.

[0080] Next, we traverse the behavioral event sequence and count the number of different types of events: opening / closing the application once; browsing the program list once; watching the live broadcast twice; pausing the live broadcast once; clicking on an ad once; posting a comment once; ordering a program once; and searching for content once.

[0081] Among them, although watching live programs A and B are two different programs, they can be regarded as the same type here, that is, watching live broadcasts, because the focus of user path analysis is on the number of behavior types, rather than the specific program content. The statistical results show that the number of behavioral event types of this user reached 8 within 1 hour, exceeding the preset threshold of 5 types. This behavior event sequence is considered to be an abnormal event sequence and deleted, or marked, and then all time series that exceed the preset type threshold are merged together as an abnormal set for further analysis.

[0082] Optionally, the preset time window can be a fixed time such as one day, or can be dynamically adjusted based on the user's historical behavior data or real-time behavior data. For example, for active users, a shorter time window can be set to capture more detailed behavior changes; for inactive users, a longer time window can be set to accumulate sufficient behavior data.

[0083] Optionally, step S13 further includes steps S137 to S139: Step S137: acquiring a path window period based on the preset configuration information, wherein the path window period is used to determine a maximum interval duration between two behavior events in a behavior path.

[0084] It should be noted that the path window period refers to the maximum interval allowed between different types of behavior events in determining the user behavior path. This window period is used to distinguish different parts of a user's continuous behavior. If the time interval between two behavior events exceeds the path window period, it may mean that the user's behavior has been interrupted or shifted. For example, assuming the path window period is 2 hours, if the user browses the program list and then starts watching the live broadcast 2 hours later, then these two behaviors will be regarded as two independent parts of the behavior path.

[0085] Step S138, determining the time difference between the end time of the preceding behavior event and the start time of the succeeding behavior event in the behavior event sequence of each user.

[0086] Step S139: if the time difference is greater than the path window period, then based on the preceding event and the succeeding event, the user's behavior event sequence is decomposed into the two initial behavior paths.

[0087] Traverse each user's behavior event sequence, obtain the start and end time of each event, and calculate the time difference between two adjacent events. Once the time difference is found to be greater than the path window period, the sequence is disconnected at the current position, and the two disconnected parts of the behavior event sequence are regarded as two independent initial behavior paths.

[0088] Optionally, if the end time of the preceding event is consistent with the start time of the succeeding event, and the types of the preceding event and the succeeding event are consistent, the preceding event and the succeeding event are regarded as one node.

[0089] For example, suppose there is a sequence of behavioral events of a user when using an IPTV platform, as follows: Open the IPTV application (start time 08-00-00); Browse movie channels (start time: 08-02-00, end time: 08-03-00); Play movie A (start time: 08-05-00, end time: 08-30-00); Play movie B (start time: 08-30-00, end time: 09-04-00); Close the IPTV application (timestamp: 09-05-00).

[0090] In this example, "Play Movie A" and "Play Movie B" have the same timestamp, and their types are both Play Movie. These two events can be merged into one node. The merged behavior event sequence is as follows: Open the IPTV application (start time 08-00-00); Browse movie channels (start time: 08-02-00, end time: 08-03-00); Play movie (start time: 08-05-00, end time: 09-04-00); Close the IPTV application (start time: 09-05-00).

[0091] In this merged event sequence, "play movie" is a merged node.

[0092] In this embodiment, an accurate user initial behavior path is generated by sorting behavior events, filtering duration, limiting the number of event types, and considering the time interval between behavior events. This method improves data quality and reduces the generation of invalid or abnormal behavior paths, thereby improving the efficiency of user behavior analysis and helping operators understand user behavior more accurately.

[0093] Based on the above embodiments of the present application, in the fourth embodiment of the present application, the same or similar contents as those in the above embodiments can be referred to the above introduction, and will not be described in detail later. Figure 4 Before step S20, the method for constructing the user behavior path further includes steps S40 to S60: Step S40: determining a start event and an end event in the target behavior event based on the preset configuration information.

[0094] Exemplarily, the preset configuration information is first read to identify which types of behavior events are included in the start event and the end event, traverse the initial behavior path, find the behavior events that match the start event and the end event, and record the locations of these events.

[0095] Step S50, determining the retention length of the initial behavior path based on the preset number of layers.

[0096] Step S60: truncate the initial behavior path based on the retained length to obtain a new initial behavior path, wherein the truncation position is located after the start event and / or before the end event.

[0097] It should be noted that in the IPTV industry, long and complex behavior paths contain a large number of information points, many of which may be irrelevant or redundant. This information will increase the difficulty of analysis, and when the behavior path is too long, the user's true intention and behavior pattern may become difficult to identify, making it difficult for IPTV operators to extract valuable information from such a behavior path. Therefore, when conducting user behavior path analysis in the IPTV industry, it is necessary to reasonably control the length of the behavior path, that is, the number of behavior events that need to be retained in the initial behavior path.

[0098] Optionally, the number of behavior events in the path segment is adjusted according to user needs, thereby adjusting the relationship between the retention length and the preset number of layers. For example, if the initial behavior path is to be decomposed into the form of layers and individual behavior events, the retention length = the preset number of layers.

[0099] For each user's initial behavior path, we start from the start event and traverse the entire path in chronological order. During the traversal, we determine whether the number of current behavior events has reached the retention length. If so, we perform a truncation operation and delete the subsequent behavior events without displaying them.

[0100] Optionally, based on the needs or purposes of path analysis, cutoff conditions can be set according to the types of behavioral events. For example, in order to identify key factors that affect user conversion, a cutoff operation can be set after a certain type of event such as purchase or order is reached, focusing on analyzing the entire process of users from browsing, selection to final conversion.

[0101] Optionally, a time window is set to only display the user's behavior path within the time window. When traversing behavior events outside the time window, a truncation operation is performed to capture the user's behavior characteristics within a certain period of time, reduce data redundancy during the analysis process, and improve analysis efficiency.

[0102] To better understand the solution provided in this example, this example is further explained in combination with specific application scenarios.

[0103] Assume that the retention length of the initial behavior path is determined to be the preset number of layers + 1. The behavior events of 5 users are grouped according to their user accounts, and the following 5 initial behavior paths are obtained: Path 1 (User 1): A->B->C->D->E->F->G->T; Path 2 (User 2): A->B->C->E->F->G->T; Path 3 (User 3): A->B->C->D->T->F->G->T; Path 4 (User 4): A->B->C->D->E->F->G; Path 5 (User 5): A->B->C->E->F->G->R.

[0104] Assuming that the preset number of layers is 4, the retention length is determined to be 5, that is, the above initial path is truncated according to the preset number of layers to obtain a new initial behavior path: Path 1 (User 1): A->B->C->D->E; Path 2 (User 2): A->B->C->E->F; Path 3 (User 3): A->B->C->D->T; Path 4 (User 4): A->B->C->D->E; Path 5 (User 5): A->B->C->E->F.

[0105] Optionally, the effective range of the initial behavior path is determined according to the positions of the start event and the end event. If the number of events in the effective range exceeds the reserved length, the event path pieces of the reserved length are intercepted as the new initial behavior path. If the number of events in the effective range does not exceed the reserved length, these event paths are directly used as the new initial behavior path.

[0106] For example, assume that the start events are A and D, and the end events are F and K. There is an initial behavior path: A->B->C->D->E->F->G->H->I->J->K. The above initial behavior path is first divided according to the start event position and the end event position: A->B->C->D, E, F->G->H->I->J->K. Assuming that the retention length is 4, the above path is truncated to obtain new initial behavior paths: A->B->C->D and F->G->H->I.

[0107] Optionally, if the number of events within the effective range exceeds the reserved length, the event path elements of the reserved length may be truncated as new initial behavior paths, that is, in the above example, the final initial behavior paths are A->B->C->D and H->I->J->K.

[0108] Optionally, the event path in the middle may be intercepted as a new initial behavior path, that is, in the above example, the final initial behavior paths are A->B->C->D and G->H->I->J.

[0109] For example, to help understand the implementation process of the method for constructing a user behavior path obtained by combining this embodiment with the above embodiments, please refer to Figure 5 , Figure 5 A brief flowchart of a method for constructing a user behavior path is provided, specifically: First, obtain the user's behavior event, read the pre-set path window period, preset number of layers and preset number of path segments. Then, according to the data of the user's behavior event, determine whether the path window period, preset number of layers and preset number of path segments need to be modified. For example, during the long live broadcast of a large-scale event, users generally watch for a long time, and the above parameter restrictions can be appropriately adjusted, such as increasing the preset number of layers. If it is determined that the modification is to be made, after modifying the above parameter restrictions, dynamically filter out the target behavior events that need to be analyzed according to the preset configuration information, including querying by user account, querying by behavior event type or querying by behavior event time, etc. Then, group the filtered target behavior events according to the user account, sort them according to their start time, and obtain the behavior event sequence of each user. Determine whether the number of types of target behavior events in each user's behavior event sequence exceeds the threshold. If it exceeds the threshold, it is determined that the user has abnormal operations such as frequent switching, and it is determined that the analysis value of the user's behavior path is not great, and the behavior event sequence of this type of user is deleted. When the number of target behavior event types in the behavior event sequences of all users is within the preset type threshold, the same target behavior events in the behavior event sequences of each user are merged, the start event or the end event is found, the path length is truncated according to the preset number of layers, and then the path is decomposed into layers and path segments for aggregation. The path segments are displayed in each layer, the layers beyond the preset number of layers are merged, and the path segments beyond the preset number of path segments are folded to obtain the user's hierarchical path, and then the number of lost sessions of the path segment in each layer is calculated, and the user's hierarchical path is displayed to the operator so that the operator can analyze the user's behavior pattern through the user's hierarchical path.

[0110] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the method for constructing the user behavior path of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0111] The present application provides a device for constructing a user behavior path, and the device for constructing a user behavior path includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for constructing a user behavior path in the above-mentioned embodiment one.

[0112] Reference below Figure 6, which shows a schematic diagram of the structure of a device for constructing a user behavior path suitable for implementing an embodiment of the present application. The device for constructing a user behavior path in an embodiment of the present application may include but is not limited to mobile terminals such as laptop computers, PADs (Portable Application Description: tablet computers), and fixed terminals such as digital TVs and desktop computers. Figure 6 The user behavior path construction device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0113] like Figure 6 As shown, the user behavior path construction device may include a processing device 1001 (such as a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM: Read Only Memory) 1002 or the program loaded from the storage device 1003 to the random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the user behavior path construction device are also stored. The processing device 1001, ROM1002 and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the user behavior path construction device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows the user behavior path construction device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or provided instead.

[0114] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0115] The user behavior path construction device provided by the present application adopts the user behavior path construction method in the above embodiment, which can solve the technical problem of low accuracy of user behavior analysis caused by the path constructed by the traditional behavior path construction method being too long and complicated. Compared with the prior art, the beneficial effects of the user behavior path construction device provided by the present application are the same as the beneficial effects of the user behavior path construction method provided by the above embodiment, and the other technical features in the user behavior path construction device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0116] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0117] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0118] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, wherein the computer-readable program instructions are used to execute the method for constructing a user behavior path in the above-mentioned embodiment.

[0119] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.

[0120] The computer-readable storage medium may be included in the user behavior path construction device; or may exist independently without being assembled into the user behavior path construction device.

[0121] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the user behavior path construction device, the user behavior path construction device can write computer program codes for performing the operations of the present application in one or more programming languages ​​or a combination thereof. The programming languages ​​include object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, or as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can 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 can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0122] The flowcharts 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 flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0123] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0124] The readable storage medium provided by the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned method for constructing user behavior paths, and can solve the technical problem of low accuracy of user behavior analysis caused by the paths constructed by the traditional behavior path construction method being too long and complicated. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as the beneficial effects of the method for constructing user behavior paths provided by the above-mentioned embodiments, and will not be repeated here.

[0125] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for constructing a user behavior path, characterized in that: The method comprises: Determine a target behavior event based on preset configuration information, and construct an initial behavior path of the user based on the target behavior event, wherein the preset configuration information includes a user account corresponding to the behavior event; Based on a preset number of layers, split the initial path of each user into path segments, each path segment corresponding to a layer; Based on the correspondence between the path segments and the levels, the path segments of each user are aggregated according to the levels to construct a hierarchical path for the user.

2. The method for constructing a user behavior path according to claim 1, characterized in that: The step of determining a target behavior event based on preset configuration information and constructing an initial behavior path of the user based on the target behavior event includes: Determining the target behavior event corresponding to the user based on the user account; Based on the start time of the target behavior event, sort each of the target behavior events to obtain a user's behavior event sequence; Based on the behavior event sequence of each user, the target behavior event is used as a node to construct the initial behavior path of the user.

3. The method for constructing a user behavior path according to claim 2, characterized in that: The step of aggregating the path segments of each user according to the hierarchy based on the correspondence between the path segments and the hierarchy to construct the hierarchical path of the user comprises: Determine a sequence label of the path segment in the behavior event sequence, wherein the sequence label corresponds one-to-one to the order of the hierarchy; Based on the sequence label, traverse the path segment of each user to determine the level corresponding to the path segment; The path segments of each user are allocated to the corresponding levels to construct the hierarchical path of the user.

4. The method for constructing a user behavior path according to claim 3, characterized in that: The step of allocating the path segments of each user to the corresponding hierarchy to construct the hierarchical path of the user comprises: Determine the number of occurrences of the path segment in the same level, and obtain the number of sessions of the path segment; Acquire a preset number of path segments based on the preset configuration information, where the preset number of path segments represents a threshold value of the number of the path segments displayed in each level; Based on the number of sessions of the path segment and the preset number of path segments, the valid path segments that need to be displayed in each level are determined to obtain the hierarchical path of the user.

5. The method for constructing a user behavior path according to claim 2, characterized in that: The step of constructing the initial behavior path of the user based on the behavior event sequence of each user and taking the target behavior event as a node comprises: Based on the preset configuration information, obtaining preset time ranges corresponding to different behavior event types; If the duration of the target behavior event is outside the preset time range, deleting the target behavior event from the behavior event sequence of the user; Based on the behavior event sequence of each user, the remaining target behavior events are used as nodes to construct the initial behavior path of the user.

6. The method for constructing a user behavior path according to claim 2, characterized in that: The step of constructing the initial behavior path of the user based on the behavior event sequence of each user and taking the target behavior event as a node also includes: Determining the number of behavior event types in the behavior event sequence of each of the users within a preset time window; If the number of the behavior event types exceeds a preset type threshold, deleting the behavior event sequence of the user; Based on the remaining behavior event sequences of each of the users, the target behavior event is used as a node to construct the initial behavior path of the user.

7. The method for constructing a user behavior path according to claim 2, characterized in that: The step of constructing the initial behavior path of the user based on the behavior event sequence of each user and taking the target behavior event as a node also includes: Acquire a path window period based on the preset configuration information, wherein the path window period is used to determine a maximum interval duration between two behavior events in a behavior path; Determine the time difference between the end time of the preceding behavior event and the start time of the succeeding behavior event in the behavior event sequence of each user; If the time difference is greater than the path window period, the user's behavior event sequence is decomposed into two initial behavior paths based on the preceding event and the succeeding event.

8. The method for constructing a user behavior path according to claim 1, characterized in that: Before the step of splitting the initial path of each user into path segments based on the preset number of layers, the method further includes: Determine a start event and an end event in the target behavior event based on the preset configuration information; Determining a retention length of the initial behavior path based on the preset number of layers; The initial behavior path is truncated based on the retained length to obtain a new initial behavior path, wherein the truncation position is located after the start event and / or before the end event.

9. A device for constructing a user behavior path, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the method for constructing a user behavior path as described in any one of claims 1 to 8.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the method for constructing a user behavior path as described in any one of claims 1 to 8 are implemented.

Images