User behavior data acquisition method for fission marketing platform

Abstract

The invention discloses a user behavior data acquisition method for a fission marketing platform, and relates to the technical field of data acquisition. According to the method, a task linkage coding field is established through a task node numbering system and an inter-task association rule matrix, so that a user behavior state has path identification and jump tracking capabilities; a user relation network diagram is further combined, a user behavior collaborative index table is constructed, a behavior collaborative collection triggering mechanism based on the fission relation among multiple users is achieved, and the limitation of traditional single-point triggering collection is broken through; and a task chain type acquisition path diagram is constructed by identifying a jump boundary of a key path in a task chain and splicing data segments, so that the continuity, the integrity and the semantic consistency of acquisition are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of data collection technology, and in particular to a method for collecting user behavior data for a viral marketing platform. Background Technology

[0002] With the rapid development of viral marketing strategies on internet platforms, the collection and mining of user behavior data has become a core means to improve the platform's precision targeting capabilities and user profile construction. Traditional data collection methods primarily rely on real-time tracking of user behavior triggers or periodic scanning to extract operation logs. However, these methods often lack a comprehensive understanding of the contextual relationships within the user behavior chain, failing to fully reveal the collaborative relationships between users and the dynamic dependencies between their behavioral paths. Especially in viral marketing platforms, user viral paths often involve multi-level task chains, associated incentive mechanisms, and cross-user state-triggered behaviors. Without the ability to collect data in a coordinated manner based on the viral path structure, it is difficult to fully capture the complex evolution of behavior, limiting the granularity and depth of subsequent analysis.

[0003] CN103593376B discloses a method and apparatus for collecting user behavior data. This method matches user actions with filtering and statistical strategies to generate keywords and categorizes and stores data under the same keywords. While this method offers some real-time performance and matching accuracy, enabling classification as the behavior occurs, it still relies heavily on single-point triggering and static strategy matching. It lacks the ability to identify and model behavioral chain structures, making it particularly challenging to efficiently collect and track collaborative behaviors when faced with complex, multi-level interconnected relationships among users. Furthermore, the solution fails to provide the capability to construct a data chain-like collection path from the starting task node to the ending node, resulting in a weak structured representation of behavioral sequences and hindering in-depth subsequent behavior tracing and correlation analysis.

[0004] CN115904469A discloses a method for collecting end-to-end trajectory data for user interaction, focusing on collecting system execution trajectory information at the system's underlying layer through a data tracking mechanism, particularly focusing on the classification and aggregation of anomalies, performance bottlenecks, and fault-related events. This method is suitable for platform performance monitoring and system behavior auditing scenarios, but its focus remains limited to system-level operational trajectory information. It fails to effectively support the linkage collection mechanism between behavioral nodes based on marketing strategy logic, and it does not construct a collaborative indexing mechanism for user behavior relationship networks with fission characteristics. Therefore, it is difficult to adapt to the refined requirements of current fission platforms for task-level behavior tracking and closed-loop collection of user interaction behavior data. Summary of the Invention

[0005] In view of the problems existing in current user behavior data collection technologies, this invention is proposed.

[0006] Therefore, the problem to be solved by this invention is how to automatically trigger the data collection task of the bound user when the behavior state jumps, so as to significantly improve the coverage and timeliness of user behavior data under the fission link.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0008] In a first aspect, the present invention provides a method for collecting user behavior data for a viral marketing platform, comprising: numbering task nodes based on a preset user viral behavior path and establishing a task association rule matrix to form a task linkage coding field; extracting user pairs with valid viral links based on a user relationship network graph and constructing a user behavior collaborative index table; when a user's task status field jumps, triggering a collection task instruction for the bound user according to the collaborative user pair recorded in the user behavior collaborative index table, specifying the start and end task node numbers and field dimension parameter set for behavior data collection; for the task linkage coding field corresponding to the triggered user, extracting behavior data within the corresponding time period according to the start and end boundaries in the task association rule matrix and the task node numbers before and after the current jump status, and uniformly storing the behavior data in a user behavior segmentation structure to construct a task chain collection path.

[0009] As a preferred embodiment of the user behavior data collection method for a viral marketing platform described in this invention, the establishment of the task association rule matrix includes: collecting a preset sample set of user viral behavior paths; marking the involved behavioral events in three dimensions according to event type, triggering order, and path branch number to construct a set of behavioral event triples; filtering a set of viral incentive behaviors with viral guidance significance from the set of behavioral event triples; setting task number identifiers based on key task nodes in the user viral behavior path and establishing a mapping table between key task nodes and task numbers; analyzing the logical dependencies between task nodes in multiple user behavior paths based on the order of task number identifiers and the correlation of conversion rates, constructing a task association rule matrix, and setting a task instruction block structure.

[0010] As a preferred embodiment of the user behavior data collection method for a viral marketing platform described in this invention, the formation of the task linkage encoding field includes: constructing a task state chain based on a mapping table and the task number group and jump boundary range recorded in the task instruction block structure; encoding the task node sequence in each user viral behavior path; and generating a corresponding task linkage encoding field to identify the user's current task state and triggerable boundary.

[0011] As a preferred embodiment of the user behavior data collection method for a viral marketing platform described in this invention, the construction of the task state chain includes: extracting the task node numbers of all users in the same viral behavior path according to a mapping table, and arranging them in ascending order according to the behavior event timestamps to generate an initial task sequence as a set of candidate task state points; traversing consecutive task pairs in the initial task sequence. <T i ,T j >, call the task instruction block structure to extract the jump boundary range [Δt] min ,Δt max If T satisfies: j -T i ∈

[0012] [Δt min ,Δt max If ], then record the corresponding task jump relationship. <T i →T j >To the valid jump set; to include all task jump relationships in the valid jump set. <T i →T j Cluster the tasks by their starting task node number, construct a state transition table, and generate a state chain segment for each task transition relationship, representing the user's state in task T. i Afterwards, you will be able to transfer to the T mission. j The system has the ability to link all state chain segments in sequence according to the actual user path, generating a complete task state chain.

[0013] As a preferred embodiment of the user behavior data collection method for a viral marketing platform described in this invention, the construction of the user behavior collaborative index table includes: traversing user nodes in the user relationship network graph, filtering out the edge set containing viral conversion behavior records, extracting user pairs that meet the closed-loop behavior path conditions, and generating an effective viral link set; based on the effective viral link set, extracting the interaction task number sequence generated by each pair of users in the original behavior sequence, combining it with the task linkage code field, determining whether the task temporal linkage condition is met, and constructing a collaborative user pair set; for each pair of collaborative users in the collaborative user pair set, extracting the current state of the task linkage code field, and generating a user behavior collaborative index table.

[0014] As a preferred embodiment of the user behavior data collection method for a viral marketing platform described in this invention, the task-sequence linkage conditions include: for each pair of users a U b > Extract the task node number subsequence within the specified time window from the original behavior sequence, denoted as B. a and B b Each record is labeled with its corresponding timestamp and task node number; in user U​a Task node number subsequence B a In the process, consecutive task pairs are extracted based on the task linkage code field. <T i ,T j > Determine user U b Task node number subsequence B b Does it follow user U in terms of time? a Complete T i T occurred afterward j The action event; if there is a time immediately following the action event, then determine whether the jump interval of the action event pair meets the time window limit configured in the task instruction block: Δt min ≤(t bj -t ai )≤Δt max Inside, where t ai For user U a Task T i timestamp, t bj For user U b Task T j Timestamp; Determine user U b Task T j If the behavior type of the behavior entry is an element in the set of fission incentive behaviors specified in the configuration, then mark the user's interaction with the behavior. a U b >

[0015] For collaborative users.

[0016] As a preferred embodiment of the user behavior data collection method for a viral marketing platform described in this invention, the extraction of behavior data within a corresponding time period includes: after a jump occurs in the task linkage code field corresponding to the triggered user, identifying the task node numbers before and after the jump from the task state chain, and locating the corresponding jump boundary range in the task association rule matrix; determining the collection time interval based on the jump boundary and the time constraints in the task instruction block structure, and filtering all behavior data records within the user collection time interval whose task numbers are within the jump boundary range; for each behavior data record, extracting specified fields from the field dimension parameter set, and combining them into a behavior fragment structure in ascending order of time, including the historical trajectory sequence of the task linkage code field, which is uniformly stored in the user behavior segmentation structure for subsequent path structure splicing.

[0017] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, they implement the steps of the user behavior data collection method for a viral marketing platform as described in the first aspect of the present invention.

[0018] ​Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, they implement the steps of the user behavior data collection method for a viral marketing platform as described in the first aspect of the present invention.

[0019] The beneficial effects of this invention are as follows: By establishing a task linkage coding field through a task node numbering system and a task association rule matrix, this invention enables user behavior states to have path recognition and jump tracking capabilities. Furthermore, by combining a user relationship network graph, a user behavior collaborative index table is constructed, realizing a multi-user behavior collaborative collection triggering mechanism based on fission relationships, breaking through the limitations of traditional single-point trigger collection. Additionally, by identifying the jump boundaries of key paths in the task chain and splicing data segments, a task chain-style collection path graph is constructed, thereby effectively improving the continuity, completeness, and semantic consistency of the collection. In summary, this invention can achieve accurate capture and closed-loop tracking of user behavior in complex fission links, improving the platform's efficiency in organizing and analyzing user behavior data. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating the user behavior data collection method used in viral marketing platforms. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] As mentioned in the background section, traditional data collection methods primarily rely on real-time tracking triggered by user behavior or periodic scanning to extract operation logs. However, these methods often lack a comprehensive understanding of the contextual relationships within the user behavior chain, failing to fully reveal the collaborative relationships between users and the dynamic dependencies between their behavioral paths. This is especially true in viral marketing platforms, where user growth paths frequently involve multi-level task chains, associated incentive mechanisms, and cross-user state-triggered behaviors. Without the ability to collect data in a coordinated manner based on the viral path structure, it is difficult to fully capture the complex evolution of behavior, thus limiting the granularity and depth of subsequent analysis.

[0026] Figure 1 This is a flowchart illustrating a user behavior data collection method for a viral marketing platform according to an embodiment of the present invention. Figure 1 As shown, the user behavior data collection methods used in viral marketing platforms include:

[0027] S1: Based on the preset user fission behavior path, assign task node numbers and establish a matrix of association rules between tasks to form a task linkage coding field.

[0028] In this embodiment of the invention, establishing the inter-task association rule matrix includes:

[0029] A. Collect a preset sample set of user fission behavior paths, and label the relevant behavior events in three dimensions according to event type, triggering order and path branch number to construct a set of behavior event triplets.

[0030] It should be noted that the user viral behavior path refers to a set of viral task processes summarized by the platform based on high-frequency behavioral patterns observed in users' past participation in viral propagation. This set of processes reflects the complete path from initial task triggering to final conversion. This invention resolves the ambiguity of task nodes in different path branches by assigning a unique task number to each task node. The task numbering method employs a three-element numbering mechanism, generating a unique task number by combining three dimensions: task trigger type, task trigger sequence number, and task branch code. This ensures that the number not only possesses behavioral recognition capabilities but also accommodates the structural complexity of path branches.

[0031] B. Select event sequences with viral guidance significance from the behavioral event triplet (such as first forwarding, invitation response, secondary propagation), set task number identifiers based on key task nodes in the user viral behavior path, and establish a mapping table between nodes and numbers.

[0032] Specifically, each behavioral event is first tagged in three dimensions according to its event type (e.g., task start, task completion, task forwarding), the chronological order of the event, and the branch number within the path, generating a set of behavioral event triples. Each behavioral event triple has a structure of (event type, behavioral sequence number, path branch number), forming the basic data unit for task identification and path modeling. This set of triples not only preserves the task evolution order in the original path but also retains path branch information, effectively reflecting the path's topology and task transformation logic. This tagging method allows for the logical aggregation of similar nodes in different fission task paths, facilitating subsequent task number mapping and matrix analysis.

[0033] Subsequently, within the constructed set of triplets, an event filtering operation needs to be performed to extract key behavioral sequences with viral marketing value. For example, actions triggered by a user's initial forwarding, invitation response, or secondary dissemination by others typically have a strong conversion orientation in the viral marketing path and should therefore be identified and extracted as the basis for task node numbering. For each extracted event sequence, based on its position in the user's viral marketing path, i.e., the structural status of the task node, a task number is determined, and a one-to-one mapping relationship is established between task nodes and their numbers, generating a task number mapping table.

[0034] C. In multiple user behavior paths, based on the order of task number identification and the correlation of conversion rate, analyze the logical dependencies between task nodes, construct a matrix of association rules between tasks, and set up a task instruction block structure to record jump boundaries and constraints.

[0035] Preferably, to further analyze the transition logic between task nodes, this invention extracts the task number sequence from the behavior paths of multiple users and determines whether there is a logical dependency between tasks based on the sequential relationship between the numbers and the statistical correlation between task conversion rates. Task conversion rate correlation refers to whether the probability of task B occurring is statistically significantly dependent given that task A is completed. If this probability exceeds a preset threshold, a valid logical transition edge is considered to exist between task A and task B. All logical edges are recorded in the inter-task association rule matrix, which uses task numbers as node identifiers and logical transition edges as matrix relationship elements.

[0036] In addition, to improve the controllability of path jumps, this invention also proposes to set up a task instruction block structure. This structure is an extensible data encapsulation unit, which contains information such as boundary conditions, time constraints, logical relationships between tasks, and behavioral type restrictions that may occur between each task pair.

[0037] Unlike traditional path-jumping models, this invention does not directly establish deterministic connections between two task nodes. Instead, it sets up logical judgments for jumps. For example, if task node A wants to jump to task node C, but node C has specific requirements regarding action time or user type, then the jump is considered valid only if the jump boundary conditions are met. Through this mechanism, dynamic control and policy constraints on path jumps are achieved, providing the necessary logical judgment basis for the subsequent construction of task state chains.

[0038] Furthermore, the formation of the task linkage coding field includes:

[0039] Based on the mapping table and the task number group and jump boundary range recorded in the task instruction block structure, a task state chain is constructed. The task node sequence in each user fission behavior path is encoded to generate a corresponding task linkage code field, which is used to identify the user's current task state and triggerable boundary.

[0040] A preferred approach to constructing the task state chain includes:

[0041] Based on the mapping table, extract the task node numbers of all tasks in the same user's fission behavior path, and sort them in ascending order according to the behavior event timestamp to generate an initial task sequence as a set of candidate task status points;

[0042] Traverse consecutive task pairs in the initial task sequence <T i ,T j >, call the task instruction block structure to extract the jump boundary range [Δt] min ,Δt max If T satisfies: j -T i ∈[Δt min ,Δt max If ], then record the corresponding task jump relationship. <T i →T j >To the valid jump set;

[0043] All task jump relationships in the valid jump set <T i →T j Cluster the tasks by their starting task node numbers to construct a state transition table. Each record in the table represents the set of next state nodes that can be jumped to from the current task, along with the corresponding transition strategy information. A state chain segment is generated for each task transition relationship, representing the user's state in task T. i Afterwards, you will be able to transfer to the T mission. j The ability to coordinate;

[0044] All state chain segments are combined sequentially according to the triggering order in the actual user path to generate a complete task state chain.

[0045] After constructing the user task state chain, the platform needs to perform structured encoding on the sequence of task nodes formed in each user's viral behavior path. This sequence of task nodes is an ordered set of task trajectories formed according to task node numbers, timestamp order, and legal jump relationships, representing the user's actual behavior path in the platform's task path. For this sequence of task nodes, the platform needs to traverse each jump unit recorded in the task state chain. For each jump path segment, key fields such as the starting task number, ending task number, allowed time window in the task instruction block, and triggered behavior type are extracted, and these fields are combined into standardized state segment encoding units.

[0046] S2: Based on the user relationship network graph, extract user pairs with valid fission links and construct a user behavior collaborative index table.

[0047] In this embodiment of the invention, the construction of the user behavior collaborative index table includes:

[0048] S2.1: Traverse the user nodes in the user relationship network graph, filter out the set of edges with records of fission conversion behavior, extract user pairs that meet the closed loop condition of behavior path, and generate a set of valid fission links.

[0049] Among them, the user relationship network graph is a directed graph structure, where nodes represent platform user entities and edges represent the behavioral propagation path relationships between users during the fission process.

[0050] Based on the selected set of behavioral edges, user pairs that meet the closed-loop behavioral path criteria need to be further extracted. A closed-loop behavioral path refers to a scenario where one user acts as the task source node, and another user completes the task transformation, forming a valid back-chain or forward-chain relationship in time and behavioral path. For example, user E invites user D to participate in task B1, and then D completes task B1, in turn activating E's task C node; this structure constitutes a closed-loop path. The platform needs to identify these user pairs and summarize them into a valid fission link set based on the task linkage code field and the task number and timestamp information in the behavior log. This valid fission link set will serve as the foundational data set for subsequent collaborative judgment and index table construction.

[0051] S2.2: Based on the effective fission link set, extract the interaction task number sequence generated by each pair of users in the original behavior sequence, combine it with the task linkage code field, determine whether the task timing linkage condition is met, and construct a collaborative user pair set.

[0052] In this embodiment of the invention, the task timing linkage conditions include:

[0053] For each pair of users a U b ​> Extract the task node number subsequence within the specified time window from the original behavior sequence, denoted as B. a and B b Each line is marked with its corresponding timestamp and task node number.

[0054] In user U a Task node number subsequence B a In the process, consecutive task pairs are extracted based on the task linkage code field. <T i ,T j > Determine user U b Task node number subsequence B b Does it follow user U in terms of time? a Complete T i T occurred afterward j Behavioral events.

[0055] If a time immediately follows an action event, determine whether the jump interval of the action event pair meets the time window limit configured in the task instruction block: Δt min ≤(t bj -t ai )≤Δt max Inside, where t ai For user U a Task T i timestamp, t bj For user U b Task T j Timestamp; Determine user U b Task T j If the behavior type of the behavior entry is an element in the set of fission incentive behaviors specified in the configuration, then mark the user's interaction with the behavior. a U b For collaborative users.

[0056] S2.3: For each pair of collaborative users in the collaborative user pair set, extract the current status of the task linkage code field and generate a user behavior collaborative index table.

[0057] It should be noted that the task state chain structure is used to describe the set and sequence of jumpable task nodes formed by a user in a single task path, which is used to guide the boundary identification and chain collection of subsequent jump events; while the task time-series linkage condition judgment is oriented towards multi-user collaboration relationships. Based on the synchronization of behavioral events and the alignment of jump numbers between different users, a linkage collaborative index structure is established. The two have different construction dimensions and uses, independent parameter systems, and mutually supportive logic.

[0058] ​As can be seen, the purpose of this step is to filter out user pairs with actual fission path value based on the user relationship network graph built within the platform, and thereby construct a user behavior collaborative index table with behavioral temporal linkage structure characteristics. This operation no longer uses individual user behavior as the sole modeling dimension, but rather jointly models multiple user behavior sequences based on fission guidance logic, thereby achieving a quantitative representation of the relationship between fission link propagation path and collaborative state. Through the construction of this index table structure, the platform can perform advanced data collection and analysis tasks such as fission behavior structure identification, jump path closed-loop judgment, and collaborative behavior evolution trend prediction at multiple user dimensions.

[0059] S3: When a user's task status field changes, the collection task instruction for the bound user is triggered based on the collaborative user pair recorded in the user behavior collaborative index table, specifying the start and end task node numbers and field dimension parameter set for behavior data collection.

[0060] Specifically, when a user's task status field undergoes a valid jump (based on the jump constraints in the task status chain and task instruction block structure), the platform triggers the data collection task instruction of the collaborative user bound to that user, according to the collaborative user pair recorded in the user behavior collaborative index table.

[0061] The data collection command must specify the following two key configuration parameters: the start and end task node numbers for data collection and the extraction of field dimension parameter sets.

[0062] Specifically, the platform extracts the task node numbers before and after the jump based on the user's task linkage code field status in the current jump event, and searches the user behavior collaboration index table to see if there is a matching collaborative user jump path. If it exists, it is considered a valid collaborative link: the starting task number is set as the task trigger node of the jump event; the ending task number is set as the node number in the collaborative user's task trajectory that forms a backlink or response linkage with the jump; at the same time, it verifies whether this number pair satisfies the jump logical edge and time boundary registered in the task association rule matrix.

[0063] Based on the event type combination of the jump task pair, the platform searches for the set of fields required for this type of task collection in the task instruction block structure and constructs a set of field dimension parameters. This set of field parameters includes, but is not limited to: behavior timestamp (used to determine path synchronization), task number and path branch number (used for jump direction resolution), behavior event type and behavior source identifier (used to identify the triggering method), and propagation level and user type tag (used for fission level determination).

[0064] S4: For the task linkage code field corresponding to the triggered user, based on the start and end boundaries in the task association rule matrix, extract the behavior data within the corresponding time period according to the task node number before and after the current jump state, and store the behavior data in a unified user behavior segmentation structure to construct a task chain collection path.

[0065] S4.1: Extract behavioral data within the corresponding time period.

[0066] After the task linkage code field corresponding to the triggered user jumps, the task node numbers before and after the jump are identified from the task status chain, and the corresponding jump boundary range is located in the task association rule matrix.

[0067] Based on the jump boundaries and time constraints in the task instruction block structure, the data collection time interval is determined, and all behavioral data records within the user's data collection time interval and whose task numbers fall within the jump boundary range are filtered. Specifically, all behavioral event records that occur within this time interval and whose task numbers are within the jump boundary field must be filtered. The filtering operation must be executed based on three key conditions: first, a timestamp constraint (the occurrence time of the behavioral record must be within the jump time window); second, a task number constraint; and third, a user identifier consistency condition (the behavioral data must belong to the behavioral trajectory under the same user identifier).

[0068] For each behavioral data record, the specified fields in the field dimension parameter set are extracted and combined into a behavioral segment structure in ascending order of time. This includes the historical trajectory sequence of the task linkage encoding field, which is uniformly stored in the user behavior segment structure for subsequent path structure splicing.

[0069] S4.2: Task chain-style data collection path.

[0070] The system reads all behavior segment structures from the user behavior segmentation structure and clusters them according to the task node numbers before and after the jump, obtaining a candidate jump segment set, which is internally arranged in ascending order of timestamp. The clustering operation uses jump number pairs as the core clustering factor. The platform needs to traverse all behavior segment structures in the user segmentation structure and perform clustering based on the start and end numbers of the task linkage code field recorded in the structure. Each cluster set is called a candidate jump segment set, representing all behavior trajectory segments that may represent "task A jumps to task B". Each set needs to be arranged in ascending order of the behavior segment's timestamp to ensure temporal continuity in subsequent trajectory splicing.

[0071] For each candidate jump segment, determine the continuity factor of two consecutive behavior segments, using the following rules:

[0072] The task linkage code field trajectory at the end position is extracted from the previous row of the fragment structure, and compared with the task linkage code field trajectory at the start position extracted from the next row of the fragment structure.

[0073] In the task instruction block structure, check if the jump direction exists in the registered set of allowed jumps or the set of step jumps. If it is in either set, the jump is considered to satisfy the trajectory direction continuity condition.

[0074] Simultaneously, the interval between the start timestamp of the next line segment structure and the end timestamp of the previous line segment structure is calculated. The minimum and maximum time interval limits for jumps of the same type of task are read from the task instruction block structure. If the time interval between the two segments is within the limit range, the trajectory time continuity condition is considered to be met.

[0075] When the jump direction satisfies the direction validity and the time interval satisfies the time validity, the two behavioral segment structures are deemed to have linkage trajectory continuity, are marked as high consistency segment segments, and are recorded as continuous path units.

[0076] Based on the collaborative user pairs recorded in the user behavior collaborative index table, multiple continuous path unit structures are used to extract the intersection of task linkage encoding fields between users, construct a path node cross graph, and form a task chain collection path graph.

[0077] Specifically, in the task node intersection graph, the platform establishes a multi-path intersection structure based on the jump direction and weights the intersection points to indicate the frequency of collaborative behaviors, forming a logical fusion of user linkage trajectories. Finally, the task node intersection graph structure is combined with the continuous path units of a single user to generate a complete task chain-style data collection path graph, which drives the platform's subsequent intelligent decision-making operations such as generating fission strategies, predicting jump behaviors, and deploying incentive strategies.

[0078] This embodiment also provides a computer device applicable to the user behavior data collection method for a viral marketing platform, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the user behavior data collection method for a viral marketing platform as proposed in the above embodiment.

[0079] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0080] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the user behavior data collection method for a viral marketing platform as proposed in the above embodiments.

[0081] In summary, this invention establishes a task linkage coding field through a task node numbering system and a task association rule matrix, enabling user behavior states to have path recognition and jump tracking capabilities. Furthermore, by combining a user relationship network graph, a user behavior collaborative index table is constructed, realizing a multi-user behavior collaborative collection triggering mechanism based on fission relationships, breaking through the limitations of traditional single-point trigger collection. Additionally, by identifying key path jump boundaries in the task chain and splicing data segments, a task chain-style collection path graph is constructed, effectively improving the continuity, completeness, and semantic consistency of the collection. In conclusion, this invention can achieve accurate capture and closed-loop tracking of user behavior in complex fission chains, improving the platform's efficiency in organizing and analyzing user behavior data.

[0082] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for collecting user behavior data for a viral marketing platform, characterized in that: include: Based on the preset user fission behavior path, the task nodes are numbered and a matrix of association rules between tasks is established to form a task linkage code field. Based on the user relationship network graph, user pairs with valid fission links are extracted, and a user behavior collaborative index table is constructed. When a user's task status field changes, the collection task instruction for the bound user is triggered based on the collaborative user pair recorded in the user behavior collaborative index table, specifying the start and end task node numbers and field dimension parameter set for behavior data collection. For the task linkage code field corresponding to the triggered user, based on the start and end boundaries in the task association rule matrix, extract the behavior data within the corresponding time period according to the task node number before and after the current jump state, and store the behavior data in a unified user behavior segmentation structure to construct a task chain collection path.

2. The user behavior data collection method for a viral marketing platform as described in claim 1, characterized in that: The establishment of the inter-task association rule matrix includes: Collect a preset sample set of user fission behavior paths, and label the relevant behavior events in three dimensions according to event type, triggering order and path branch number to construct a set of behavior event triplets; In the behavioral event triplet set, a set of fission incentive behaviors with fission guidance significance is selected, and task number identifiers are set according to the key task nodes in the user fission behavior path, and a mapping table between key task nodes and task numbers is established. Based on the correlation between task number identification order and conversion rate in multiple user behavior paths, the logical dependencies between task nodes are analyzed, a matrix of association rules between tasks is constructed, and a task instruction block structure is set.

3. The user behavior data collection method for a viral marketing platform as described in claim 2, characterized in that: The formation of the task linkage coding field includes: Based on the mapping table and the task number group and jump boundary range recorded in the task instruction block structure, a task state chain is constructed. The task node sequence in each user fission behavior path is encoded to generate a corresponding task linkage code field, which is used to identify the user's current task state and triggerable boundary.

4. The user behavior data collection method for a viral marketing platform as described in claim 3, characterized in that: The construction of the task state chain includes: Based on the mapping table, extract the task node numbers of all tasks in the same user's fission behavior path, and sort them in ascending order according to the behavior event timestamp to generate an initial task sequence as a set of candidate task status points; Traverse consecutive task pairs in the initial task sequence <T i ,T j >, call the task instruction block structure to extract the jump boundary range [Δt] min ,Δt max If T satisfies: j -T i ∈[Δt min ,Δt max If ], then record the corresponding task jump relationship. <T i →T j >To the valid jump set; All task jump relationships in the valid jump set <T i →T j Cluster the tasks by their starting node number, construct a state transition table, and generate a state chain segment for each task transition relationship, representing the user's state in task T. i Afterwards, you will be able to transfer to the T mission. j The ability to coordinate; All state chain segments are combined sequentially according to the triggering order in the actual user path to generate a complete task state chain.

5. The user behavior data collection method for a viral marketing platform as described in claim 1, characterized in that: The construction of the user behavior collaborative index table includes: Traverse the user nodes in the user relationship network graph, filter out the set of edges with records of fission conversion behavior, extract user pairs that meet the closed loop condition of behavior path, and generate a set of effective fission links; Based on the effective fission link set, extract the interaction task number sequence generated by each pair of users in the original behavior sequence, combine it with the task linkage code field to determine whether the task timing linkage condition is met, and construct a collaborative user pair set; For each pair of collaborative users in the collaborative user pair set, extract the current status of the task linkage encoding field to generate a user behavior collaborative index table.

6. The user behavior data collection method for a viral marketing platform as described in claim 5, characterized in that: The task timing linkage conditions include: For each pair of users a U b > Extract the task node number subsequence within the specified time window from the original behavior sequence, denoted as B. a and B b Each line is marked with its corresponding timestamp and task node number;​ In user U a Task node number subsequence B a In the process, consecutive task pairs are extracted based on the task linkage code field. <T i ,T j > Determine user U b Task node number subsequence B b Does it follow user U in terms of time? a Complete T i T occurred afterward j behavioral events; If a time immediately follows an action event, determine whether the jump interval of the action event pair meets the time window limit configured in the task instruction block: Δt min ≤(t bj -t ai )≤Δt max Inside, where t ai For user U a Task T i timestamp, t bj For user U b Task T j Timestamp; Determine user U b Task T j If the behavior type of the behavior entry is an element in the set of fission incentive behaviors specified in the configuration, then mark the user's interaction with the behavior. a U b For collaborative users.​ 7. The user behavior data collection method for a viral marketing platform as described in claim 1, characterized in that: The extraction of behavioral data within the corresponding time period includes: After the task linkage code field corresponding to the triggered user jumps, the task node numbers before and after the jump are identified from the task status chain, and the corresponding jump boundary range is located in the task association rule matrix. Based on the jump boundary and the time constraints in the task instruction block structure, the collection time interval is determined, and all behavioral data records within the user collection time interval and whose task numbers are within the jump boundary range are filtered. For each behavioral data record, the specified fields in the field dimension parameter set are extracted and combined into a behavioral segment structure in ascending order of time. This includes the historical trajectory sequence of the task linkage encoding field, which is uniformly stored in the user behavior segment structure for subsequent path structure splicing.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the user behavior data collection method for a viral marketing platform as described in any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the user behavior data collection method for a viral marketing platform as described in any one of claims 1 to 7.

Images