Method, device, equipment and storage medium for dynamically determining interactive qualification level

Abstract

The present application discloses a method, device, equipment and storage medium for dynamically determining an interaction qualification level, the method comprising: obtaining the first-stage behavior data of a target user, converting the first-stage behavior data into a first graph representation data of a preset data dimension, and obtaining a first-stage behavior prediction result; determining a threshold interval corresponding to the first-stage behavior prediction result, determining the first-stage interaction qualification level of the target user based on the threshold interval, obtaining the second-stage behavior data of the target user determined based on the first-stage interaction qualification level; determining the second graph representation data of the second-stage behavior data to determine the second-stage interaction qualification level of the target user in the second stage, wherein the second stage is a time-series continuous stage with the first stage. The present application solves the technical problem that it is difficult to accurately determine the interaction qualification level of a user in the prior art.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology in financial technology (Fintech), and in particular to a method, device, equipment and storage medium for dynamically determining an interaction qualification level. Background Art

[0002] With the continuous development of financial technology, especially Internet technology finance, more and more technologies are applied in the financial field, but the financial industry also puts forward higher requirements for technology. For example, the financial industry also has higher requirements for the dynamic determination of interactive qualification levels.

[0003] At present, the user's interaction qualification level is usually determined by the order rate prediction result determined by the user behavior data distributed in a specific dimension. However, in most scenarios, the distribution of user behavior feature dimensions changes dynamically. If the user's interaction qualification level is determined in the above way, the user's order rate prediction result will be inaccurate, which will lead to inaccurate determination of the user's interaction qualification level. Summary of the invention

[0004] The main purpose of the present application is to provide a method, device, equipment and storage medium for dynamically determining an interaction qualification level, aiming to solve the technical problem in the prior art that it is difficult to accurately determine a user's interaction qualification level.

[0005] To achieve the above object, the present application provides a method for dynamically determining an interaction qualification level, the method comprising:

[0006] Acquire the first-stage behavior data of the target user, convert the first-stage behavior data into first graph representation data of a preset data dimension, perform prediction processing on the first graph representation data, and obtain the first-stage behavior prediction result of the first graph representation data;

[0007] Determine a threshold interval corresponding to the first-stage behavior prediction result, determine a first-stage interaction qualification level of a target user based on the threshold interval, and obtain second-stage behavior data of the target user determined based on the first-stage interaction qualification level;

[0008] Determine second graph representation data of the second-stage behavior data, and determine the second-stage interaction qualification level of the target user based on the second graph representation data, wherein the second stage is a temporally continuous stage with the first stage.

[0009] Optionally, the steps of determining a threshold interval corresponding to the first-stage behavior prediction result, determining a first-stage interaction qualification level of a target user based on the threshold interval, and obtaining second-stage behavior data of the target user determined based on the first-stage interaction qualification level include:

[0010] Based on the numerical value of the first-stage behavior prediction result, determine the first-stage interaction qualification level of the first-stage target user;

[0011] The first interaction file of the first-stage interaction qualification level is determined and parsed to obtain the second-stage behavior data.

[0012] Optionally, the step of determining and parsing the first interaction file of the first-stage interaction qualification level to obtain the second-stage behavior data includes:

[0013] a first interaction document for determining the first stage interaction qualification level;

[0014] Determine the channel type of the target user;

[0015] The first interaction file is parsed, and second-stage behavior data is determined based on the channel type and the parsed first interaction file.

[0016] Optionally, the step of obtaining the first-stage behavior data of the target user and converting the first-stage behavior data into a first graph representation data of a preset data dimension includes:

[0017] Acquire the first-stage behavior data of the target user, and determine the first behavior graph data of the first-stage behavior data;

[0018] Determine the encoding vector of the first behavior graph data, perform dimensionality reduction processing on the encoding vector to map it to the preset data dimension, and obtain first graph representation data.

[0019] Optionally, the step of determining the first behavior graph data of the first stage behavior data includes:

[0020] Generate each first user behavior node of the target user based on the first stage behavior data;

[0021] Determining a first conversion order between the first user behavior nodes to connect the first user behavior nodes to obtain initial user behavior graph data;

[0022] Determine first conversion frequency data between the first user behavior nodes, and determine the first behavior graph data based on the first conversion frequency data and the initial user behavior graph data.

[0023] Optionally, the initial user behavior graph data includes a first node connection edge of a first user behavior node;

[0024] The step of determining the first behavior graph data based on the first conversion frequency data and the initial user behavior graph data comprises:

[0025] Determine, based on the first conversion frequency data, a first user behavior conversion frequency corresponding to the first node connection edge;

[0026] Determining a first connection edge weight of the first node connection edge based on the first user behavior conversion frequency and a preset target conversion frequency corresponding to the first node connection edge;

[0027] In the initial user behavior graph data, the first connection edge weight is assigned to the first node connection edge to obtain the first behavior graph data.

[0028] Optionally, after the step of determining the second graph representation data of the second-stage behavior data and determining the second-stage interaction qualification level of the target user based on the second graph representation data, the method further comprises:

[0029] A second interaction file of the second-stage interaction qualification level is determined, so that the target user can perform various user behaviors based on the second interaction file.

[0030] Optionally, the steps of determining a threshold interval corresponding to the first-stage behavior prediction result, determining a first-stage interaction qualification level of a target user based on the threshold interval, and obtaining second-stage behavior data of the target user determined based on the first-stage interaction qualification level include:

[0031] Determine a threshold interval corresponding to the first-stage behavior prediction result, and determine the first-stage interaction qualification level of the target user based on the threshold interval;

[0032] Obtaining the channel type of the target user with the first-stage interaction qualification level;

[0033] If the channel type is a public account channel, outputting preset selected articles to the target user based on the public account channel;

[0034] The second-stage behavior data of the target user determined based on the preset selected articles is obtained.

[0035] The present application also provides a device for dynamically determining an interaction qualification level, the device for dynamically determining an interaction qualification level comprising:

[0036] an acquisition module, used to acquire the first-stage behavior data of the target user, convert the first-stage behavior data into first graph representation data of a preset data dimension, perform prediction processing on the first graph representation data, and obtain the first-stage behavior prediction result of the first graph representation data;

[0037] A first determination module is used to determine a threshold interval corresponding to the first-stage behavior prediction result, determine a first-stage interaction qualification level of a target user based on the threshold interval, and obtain second-stage behavior data of the target user determined based on the first-stage interaction qualification level;

[0038] The second determination module is used to determine second graph representation data of the second stage behavior data, and determine the second stage interaction qualification level of the target user based on the second graph representation data, wherein the second stage is a time-series continuous stage with the first stage.

[0039] Optionally, the first determining module includes:

[0040] A first determining unit, configured to determine a first-stage interaction qualification level of a first-stage target user based on a numerical value of a first-stage behavior prediction result;

[0041] The second determining unit is configured to determine and parse the first interaction file of the first-stage interaction qualification level to obtain the second-stage behavior data.

[0042] Optionally, the second determining unit includes:

[0043] A first determining subunit, configured to determine a first interaction file of the interaction qualification level at the first stage;

[0044] A second determining subunit is used to determine the channel type of the target user;

[0045] The third determining subunit is used to determine the second-stage behavior data based on the channel type and the first interaction file.

[0046] Optionally, the acquisition module includes:

[0047] A first acquisition unit, configured to acquire first-stage behavior data of a target user and determine first behavior graph data of the first-stage behavior data;

[0048] The third determination unit is used to determine the encoding vector of the first behavior graph data, and perform dimensionality reduction processing on the encoding vector to map it to the preset data dimension to obtain the first graph representation data.

[0049] Optionally, the first acquiring unit includes:

[0050] A generating subunit, configured to generate each first user behavior node of the target user based on the first stage behavior data;

[0051] a fourth determining subunit, configured to determine a first conversion sequence between the first user behavior nodes, so as to connect the first user behavior nodes to obtain initial user behavior graph data;

[0052] The fifth determining subunit is used to determine the first conversion frequency data between the first user behavior nodes, and determine the first behavior graph data based on the first conversion frequency data and the initial user behavior graph data.

[0053] Optionally, the initial user behavior graph data includes a first node connection edge of a first user behavior node;

[0054] The fifth determining subunit is used to implement:

[0055] Determine, based on the first conversion frequency data, a first user behavior conversion frequency corresponding to the first node connection edge;

[0056] Determining a first connection edge weight of the first node connection edge based on the first user behavior conversion frequency and a preset target conversion frequency corresponding to the first node connection edge;

[0057] In the initial user behavior graph data, the first connection edge weight is assigned to the first node connection edge to obtain the first behavior graph data.

[0058] Optionally, the interactive qualification level dynamic determination device further comprises:

[0059] The third determining module is used to determine a second interaction file of the second-stage interaction qualification level, so that the target user can perform various user behaviors based on the second interaction file.

[0060] Optionally, the first determining module further includes:

[0061] A sixth determining subunit, configured to determine a threshold interval corresponding to the first-stage behavior prediction result, and determine a first-stage interaction qualification level of the target user based on the threshold interval;

[0062] A third acquisition unit, configured to acquire the channel type of the target user of the first-stage interaction qualification level;

[0063] A seventh determination subunit, configured to, if the channel type is a public account channel, determine to output a preset selected article to the target user based on the public account channel;

[0064] The fourth acquisition unit is used to acquire the second-stage behavior data of the target user determined based on the preset selected articles.

[0065] The present application also provides a device for dynamically determining an interaction qualification level, which is a physical device. The device for dynamically determining an interaction qualification level comprises: a memory, a processor, and a program of the method for dynamically determining an interaction qualification level stored in the memory and executable on the processor. When the program of the method for dynamically determining an interaction qualification level is executed by the processor, the steps of the method for dynamically determining an interaction qualification level as described above can be implemented.

[0066] The present application also provides a storage medium, on which is stored a program for implementing the above-mentioned method for dynamically determining the interaction qualification level. When the program for the method for dynamically determining the interaction qualification level is executed by a processor, the steps of the above-mentioned method for dynamically determining the interaction qualification level are implemented.

[0067] The present application obtains the first-stage behavior data of the target user, converts the first-stage behavior data into first graph representation data of a preset data dimension, and inputs the first graph representation data into a preset classification model to obtain the first-stage behavior prediction result of the first graph representation data; wherein the preset classification model is obtained by iteratively training a preset model to be trained based on the graph representation data with preset labels; determines a threshold interval corresponding to the first-stage behavior prediction result, determines the first-stage interaction qualification level of the target user based on the threshold interval, and obtains the second-stage behavior data of the target user determined based on the first-stage interaction qualification level; determines second graph representation data of the second-stage behavior data, and determines the second-stage interaction qualification level of the target user based on the second graph representation data, wherein the second stage is a time-series continuous stage with the first stage. Compared with the technical means in the prior art for determining the user's interaction qualification level based on user behavior data with a specific distribution, in the present application, the behavior data at different stages that changes dynamically are converted into the first graph representation data of the dynamically changing feature dimension, and then the user's order rate prediction result is dynamically determined, which can reduce the variability in the order rate prediction process to dynamically determine the user's interaction qualification. This method overcomes the defect in the prior art of determining the interaction qualification level of users at all stages based on user behavior data with a determined feature dimension distribution, improves the accuracy of order rate prediction, and thereby improves the accuracy of determining the user's qualification level. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] 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.

[0069] In order to more clearly illustrate the embodiments of the present application or the technical solutions in 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.

[0070] Figure 1 This is a flow chart of the first embodiment of the method for dynamically determining the interactive qualification level of the present application;

[0071] Figure 2 This is a flowchart of detailed steps of step S20 in the method for dynamically determining the interactive qualification level of this application;

[0072] Figure 3 A schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present application;

[0073] Figure 4 A schematic diagram of a scenario for dynamically determining the interactive qualification level for this application.

[0074] 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

[0075] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0076] The present application embodiment provides a method for dynamically determining an interaction qualification level. In the first embodiment of the method for dynamically determining an interaction qualification level of the present application, refer to Figure 1 The method for dynamically determining the interaction qualification level includes:

[0077] Step S10, obtaining the first-stage behavior data of the target user, converting the first-stage behavior data into first graph representation data of a preset data dimension, and inputting the first graph representation data into a preset classification model to obtain the first-stage behavior prediction result of the first graph representation data;

[0078] The preset classification model is obtained by iteratively training a preset model to be trained based on graph representation data with preset labels;

[0079] Step S20, determining a threshold interval corresponding to the first-stage behavior prediction result, determining a first-stage interaction qualification level of a target user based on the threshold interval, and obtaining second-stage behavior data of the target user determined based on the first-stage interaction qualification level;

[0080] Step S30, determining second graph representation data of the second stage behavior data, and determining the second stage interaction qualification level of the target user based on the second graph representation data, wherein the second stage is a temporally continuous stage with the first stage.

[0081] The specific steps are as follows:

[0082] Step S10, obtaining the first-stage behavior data of the target user, converting the first-stage behavior data into first graph representation data of a preset data dimension, and inputting the first graph representation data into a preset classification model to obtain the first-stage behavior prediction result of the first graph representation data;

[0083] The preset classification model is obtained by iteratively training a preset model to be trained based on graph representation data with preset labels;

[0084] In this embodiment, it should be noted that it can be applied to an interactive qualification level dynamic determination system, which is subordinate to an interactive qualification level dynamic determination device. For the interactive qualification level dynamic determination system, there is a communication connection relationship with an organization such as an enterprise. Therefore, as long as the user has an operation behavior on the enterprise's associated web page, the interactive qualification level dynamic determination system can obtain the user's operation behavior, and predict the user's order possibility or purchase possibility based on the user's operation behavior, and then determine the first-stage interactive qualification level of the target user. Then the enterprise provides feedback service or first feedback information based on the interactive qualification level. The feedback service or first feedback information can be carried out through various channels such as public accounts, WeChat, APP, etc. For enterprises, there are industry characteristics (business scope characteristics), and industry characteristics include business characteristics, and business characteristics include different product characteristics, etc. That is, for enterprises, there are multiple layers of deep characteristics. Therefore, different users' operation behaviors on the enterprise website have different distribution characteristics.

[0085] It should be noted that enterprises and users (user terminals) can interact through multiple entrances or channels, such as through apps, mini-programs, official accounts, or sharing links. After the enterprise has recorded the operation behavior data on the app, mini-programs, official accounts, or sharing links, specifically, after the enterprise obtains the operation behavior data of each user at different depths through the operation data records of each operation button, the operation behavior data of each user at different depths are sent to the interactive qualification level dynamic determination system to predict whether the user will purchase or place an order, so as to obtain the order rate prediction result (user behavior prediction result), and further determine the interactive qualification level of the target user based on the order rate prediction result. In this embodiment, it should be noted that the operation behavior data recorded by the enterprise includes at least one user behavior. For example, the user clicks to read the article, adds the customer service WeChat, and answers the customer service call. Then each user behavior is clicked to read the article, added WeChat, and answered the call, and each user behavior is staged. The staged nature here can refer to: after the initial user behavior, the user receives services based on the initial user behavior feedback from the enterprise, such as the feedback WeChat public account article, and then generates user behavior (such as click to read behavior) based on the feedback service such as the feedback WeChat public account article. The behavior generated afterwards is different from the initial user behavior stage. In this embodiment, the data corresponding to the changed behavior is predicted again. After obtaining the corresponding prediction results, the target user's interactive qualification level is determined. The enterprise can further dynamically adjust the service level and corresponding services based on the target user's interactive qualification level.

[0086] It should be noted that although enterprises can interact with users through different methods such as apps, mini-programs, official accounts or shared links, overall, the interaction between users and enterprises can be determined by behavioral nodes such as enterprises, businesses, portals, pages, etc.

[0087] Acquire the first-stage behavior data of the target user, convert the first-stage behavior data into first graph representation data of a preset data dimension, and input the first graph representation data into a preset classification model to obtain the first-stage behavior prediction result of the first graph representation data. Specifically, the graph representation data is data used to represent the user behavior graph, and the graph representation data includes user graph embedding data. For example, the first graph representation data is input into the preset classification model to obtain the first-stage behavior prediction result of the first graph representation data. It should be noted that, in this embodiment, the graph embedding data is data obtained by converting behavior data with different distributions into the same dimension. Specifically, the behavior data with different distributions are converted into user graph embedding data of a specific dimension, that is, the user behavior of the target user can still be accurately predicted based on the user graph embedding data of the same feature dimension, thereby improving the accuracy of the prediction, and then achieving accurate determination of the level of the order rate prediction result.

[0088] Obtain the first-stage behavior data of the target user, which may refer to the initial behavior data or the behavior data of the current node, determine the first graph representation data of the first-stage behavior data, and determine the first-stage behavior prediction result of the first graph representation data.

[0089] In this embodiment, the first graph representation data is input into a preset classification model to obtain a first-stage behavior prediction result of the first graph representation data;

[0090] The preset classification model is obtained by iteratively training a preset model to be trained based on graph representation data with preset labels;

[0091] Specifically, in this embodiment, a preset classification model is pre-trained, and the preset classification model is obtained by iteratively training a preset model to be trained based on graph representation data with preset labels. For example, the preset classification model is obtained by iteratively training a preset model to be trained based on graph embedding vector data with preset labels. Specifically, the preset label of the graph embedding vector data is compared with the predicted label obtained after input into the preset model to be trained to obtain a comparison result, and the parameters of the preset model to be trained are adjusted based on the comparison result, and the adjusted preset model to be trained is iteratively trained until the iteration reaches a preset number of training times or the preset loss function converges, and the preset classification model can be obtained.

[0092] In this embodiment, the first-stage behavior prediction result of the first-stage behavior prediction result of the first-stage behavior prediction result is obtained by inputting the first-stage graph representation data into a preset classification model; wherein the preset classification model is obtained by iteratively training a preset model to be trained based on the graph representation data with preset labels. Since the preset classification model is accurately trained in this embodiment, the first-stage behavior prediction result is accurately obtained.

[0093] The step of obtaining the first-stage behavior data of the target user and determining the first graph representation data of the first-stage behavior data comprises:

[0094] Step S11, obtaining the first-stage behavior data of the target user, and determining the first behavior graph data of the first-stage behavior data;

[0095] In this embodiment, the first-stage behavior data of the target user is obtained, and the first behavior graph data of the first-stage behavior data is determined. Specifically, the first behavior graph data is determined through the user behavior nodes in the first-stage behavior data, the sequential relationship between each user behavior node, etc.

[0096] Step S12, determining the encoding vector of the first behavior graph data, performing dimensionality reduction processing on the encoding vector to map it to the preset data dimension, and obtaining the first graph representation data.

[0097] Determine the encoding vector of the first behavior graph data, perform dimensionality reduction processing on the encoding vector to map it to the preset data dimension, and obtain the first graph representation data. In the present embodiment, it should be noted that before determining the encoding vector of the first behavior graph data, there is already a set of association graph data of all users. Assume that the set of association graph data includes 5 user behavior subgraphs, that is, there are 5 user behavior subgraphs a, b, c, d, and e. If the first behavior graph data includes user behavior subgraphs a and c, the encoding vector corresponding to the first behavior graph data is (1, 0, 1, 0, 0), wherein 1 indicates that the target user includes the user behavior subgraph corresponding to the bit corresponding to 1, and 0 indicates that the target user does not include the user behavior subgraph corresponding to the bit corresponding to 0.

[0098] Acquiring the first-stage behavior data of the target user, converting the first-stage behavior data into first graph representation data of a preset data dimension, and inputting the first graph representation data into a preset classification model to obtain the first-stage behavior prediction result of the first graph representation data includes: acquiring the first-stage behavior data of the target user, converting the first-stage behavior data into first-stage graph vector embedding data of a preset data dimension, and inputting the first-stage graph vector embedding data into the preset classification model to obtain the first-stage behavior prediction result of the first-stage graph vector embedding data, or acquiring the first-stage behavior data of the target user, converting the first-stage behavior data into first-stage graph matrix embedding data of a preset data dimension, and inputting the first-stage graph matrix embedding data into the preset classification model to obtain the first-stage behavior prediction result of the first-stage graph matrix embedding data.

[0099] Step S20, determining a threshold interval corresponding to the first-stage behavior prediction result, determining a first-stage interaction qualification level of a target user based on the threshold interval, and obtaining second-stage behavior data of the target user determined based on the first-stage interaction qualification level;

[0100] Determine the threshold interval corresponding to the first-stage behavior prediction result, and determine the first-stage interaction qualification level of the target user based on the threshold interval. For example, the first-stage behavior prediction result, i.e., the order rate, is 0.3. At this time, the threshold interval corresponding to the prediction result is the first threshold interval, and the corresponding first feedback information is common questions and answers (pre-stored). After obtaining the first feedback information, obtain the second-stage behavior data of the target user determined based on the first feedback information. Specifically, the user generates a (click or read) behavior based on the first feedback information, then the generated (click or read) behavior and the first-stage behavior data constitute the second-stage behavior data.

[0101] Reference Figure 2 The steps of determining a threshold interval corresponding to the first-stage behavior prediction result, determining a first-stage interaction qualification level of a target user based on the threshold interval, and obtaining second-stage behavior data of the target user determined based on the first-stage interaction qualification level include:

[0102] Step S21, determining a threshold interval corresponding to the first-stage behavior prediction result, and determining the first-stage interaction qualification level of the target user based on the threshold interval;

[0103] Based on the numerical value of the first-stage behavior prediction result (order rate), such as 0.3, the corresponding threshold interval is determined to determine the first-stage interaction qualification level of the first-stage target user. Specifically, Figure 4As shown, if the order rate is divided into three threshold intervals in the interactive qualification level dynamic determination system, where the first threshold interval is (0, 0.33), the second threshold interval is (0.33-0.66), and the third threshold interval is (0.66-1), then the first stage interactive qualification level of the first stage behavior prediction result (order rate prediction result) is the first level.

[0104] Step S22, determining and parsing the first interaction file of the first-stage interaction qualification level to obtain the second-stage behavior data.

[0105] Determine the first interaction file of the first stage interaction qualification level, parse the first interaction file, obtain the first interaction file including the interaction method and the content of the first feedback information, and then determine the target user to obtain the second stage behavior data based on the content of the first feedback information.

[0106] Specifically, for example, the interaction file may be: a preset common marketing article, a pure machine interaction mode, and output common questions and answers after the interaction;

[0107] Or the interaction file can be: preset machine-selected articles, machine-first then human interaction, output popular questions and answers after interaction;

[0108] Or the interaction file can be: preset manually selected articles, pure manual interaction mode, and output manually selected questions and answers after the interaction.

[0109] The step of determining and parsing the first interaction file of the first phase interaction qualification level to obtain the second phase behavior data comprises:

[0110] Step S221, determining a first interaction file of the first stage interaction qualification level;

[0111] Step S222, determining the channel type of the target user;

[0112] In this embodiment, a mapping relationship between interaction qualification levels and interaction files is pre-stored. Therefore, after the first-stage interaction qualification level is determined, the first interaction file of the first-stage interaction qualification level is determined based on the mapping relationship. The first interaction file includes first feedback information, so the first feedback information is determined based on the first interaction file.

[0113] In this embodiment, the channel type of the target user is also determined, and the channel includes a mini program channel, a public account channel, etc. The channel type can be further subdivided.

[0114] Step S223: parsing the first interaction file, and determining the second-stage behavior data based on the channel type and the parsed first interaction file.

[0115] The first interaction file is parsed, and the second-stage behavior data is determined based on the channel type and the parsed first interaction file. Specifically, the target user is interacted with based on the interaction mode of the channel type and the first-stage interaction qualification level, such as interacting with the target user based on a machine mode (click or touch) and the first feedback information (ordinary article), and the second-stage behavior data of the target user determined based on the first feedback information is obtained; interacting with the target user based on a manual mode (manual phone call) and the first feedback information (manually selected article) is obtained, and the second-stage behavior data of the target user determined based on the first feedback information is obtained; interacting with the target user based on a machine-first-manual mode and the first feedback information (machine-selected article) is obtained.

[0116] Step S30, determining second graph representation data of the second stage behavior data, and determining the second stage interaction qualification level of the target user based on the second graph representation data, wherein the second stage is a temporally continuous stage with the first stage.

[0117] In the present embodiment, the second stage and the first stage are stages that are sequential in time, that is, based on the order rate in the previous stage (the result of the user behavior prediction in the previous stage), the first stage interaction qualification level is determined (the corresponding first feedback information or service content is determined), and then the order rate in the subsequent stage is determined based on the first stage interaction qualification level. That is, in the present embodiment, according to the first stage behavior prediction result (the order rate in the previous stage), the service content is graded, and service content corresponding to the level is provided (determined to correspond to the first feedback information or service content). The user generates behavior in the service content (determined to correspond to the first feedback information or service content), and then the corresponding second feedback information is determined. That is, in the present embodiment, the service level is dynamically adjusted.

[0118] Specifically, in the same way as determining the first graph representation data of the first stage behavior data, determining the second graph representation data of the second stage behavior data, and determining the second stage behavior prediction result of the second graph representation data, the second stage behavior prediction result of the second graph representation data can be obtained by inputting the second graph representation data into a preset classification model, and then obtaining the second stage interaction qualification level (qualification level).

[0119] Wherein, after the step of determining the second graph representation data of the second stage behavior data and determining the second stage behavior prediction result of the second graph representation data, the method further comprises:

[0120] Step S40: determining a second interaction file of the second-stage interaction qualification level, so that the target user can perform various user behaviors based on the second interaction file.

[0121] In this embodiment, second feedback information of the second stage behavior prediction result is also determined so that the target user can perform various user behaviors based on the second feedback information, that is, the user's interaction qualification level is continuously adjusted according to the order rate at different stages.

[0122] The present application obtains the first-stage behavior data of the target user, converts the first-stage behavior data into first graph representation data of a preset data dimension, and inputs the first graph representation data into a preset classification model to obtain the first-stage behavior prediction result of the first graph representation data; wherein the preset classification model is obtained by iteratively training a preset model to be trained based on the graph representation data with preset labels; determines a threshold interval corresponding to the first-stage behavior prediction result, determines the first-stage interaction qualification level of the target user based on the threshold interval, and obtains the second-stage behavior data of the target user determined based on the first-stage interaction qualification level; determines second graph representation data of the second-stage behavior data, and determines the second-stage interaction qualification level of the target user based on the second graph representation data, wherein the second stage is a time-series continuous stage with the first stage. Compared with the technical means in the prior art for determining the user's interaction qualification level based on user behavior data with a specific distribution, in the present application, the behavior data at different stages that changes dynamically are converted into the first graph representation data of the dynamically changing feature dimension, and then the user's order rate prediction result is dynamically determined, which can reduce the variability in the order rate prediction process to dynamically determine the user's interaction qualification. This method overcomes the defect in the prior art of determining the interaction qualification level of users at all stages based on user behavior data with a determined feature dimension distribution, improves the accuracy of order rate prediction, and thereby improves the accuracy of determining the user's qualification level.

[0123] Further, based on the first embodiment of the present application, another embodiment of the present application is provided. In this embodiment, the step of determining the first behavior graph data of the first stage behavior data includes:

[0124] Step A1, generating each first user behavior node of the target user based on the first stage behavior data;

[0125] It should be noted that the behavior graph data is a structural graph data that records user behaviors and the order in which user behaviors occur. The behavior graph data specifically includes user behavior nodes, first node connection edges, connection edge weights corresponding to the first node connection edges, and the order relationship between each user behavior node. Specifically, one user behavior node corresponds to one user behavior, such as Figure 4As shown, enterprise, business, portal, page, article, etc. are all names of the user behavior nodes.

[0126] Step A2, determining a first conversion order between the first user behavior nodes to connect the first user behavior nodes to obtain initial user behavior graph data;

[0127] Determine a first conversion sequence between the first user behavior nodes to connect the first user behavior nodes to obtain initial user behavior graph data. Specifically, the first conversion sequence is the order in which the user behaviors occur. For example, assuming that user A communicates by phone first, then adds WeChat, and then decides to buy a movie, then the first conversion sequence is from phone call to adding WeChat, and then adding WeChat to buying a movie.

[0128] Based on the first conversion sequence corresponding to each of the user behaviors, each of the user behavior nodes is connected to obtain initial user behavior graph data.

[0129] Step A3: determine the first conversion frequency data between the first user behavior nodes, and determine the first behavior graph data based on the first conversion frequency data and the initial user behavior graph data.

[0130] In this embodiment, it should be noted that the first conversion frequency data is the conversion number data between each of the first user behavior nodes, or the first conversion frequency is the conversion number between two user behaviors, for example, user A converted from WeChat communication to telephone communication 2 times, that is, the user behavior process is WeChat-telephone-WeChat-telephone, and the first conversion frequency corresponding to the conversion from WeChat to telephone is 2. Based on the first conversion frequency data and the initial user behavior graph data, the first behavior graph data is determined.

[0131] The initial user behavior graph data includes a first node connection edge of a first user behavior node;

[0132] The first node connection edge is a directed curve segment connecting user behavior nodes, which is used to represent the process from one user behavior to another user behavior, and the direction of the user behavior conversion is the direction of the directed curve segment. The first node connection edge has a connection edge weight, and the connection edge weight is used to represent the frequency of the user converting the user behavior from the first node connection edge corresponding to the connection edge weight, wherein the connection edge weight can be represented by the color or path length of the first node connection edge.

[0133] The step of determining the first behavior graph data based on the first conversion frequency data and the initial user behavior graph data comprises:

[0134] Step B1, determining a first user behavior conversion frequency corresponding to the first node connection edge based on the first conversion frequency data;

[0135] Step B2, determining a first connection edge weight of the first node connection edge based on the first user behavior conversion frequency and a preset target conversion frequency corresponding to the first node connection edge;

[0136] Step B3: assigning the first connection edge weight to the first node connection edge in the initial user behavior graph data to obtain the first behavior graph data.

[0137] In this embodiment, it should be noted that the preset target conversion frequency is the maximum number of visits to the first node connection edge, that is, the maximum number of conversions between the user behaviors corresponding to the first node connection edge in all user behavior data. For example, assuming that there are 5 users A, B, C, D, and E locally, and there is a first node connection edge a in the user behavior graph corresponding to each user, and the two user behaviors corresponding to the first node connection edge a are i and j, and the number of conversions of user A between i and j is 1, the number of conversions of user B between i and j is 2, the number of conversions of user C between i and j is 3, the number of conversions of user D between i and j is 4, and the number of conversions of user E between i and j is 5, then the preset target conversion frequency is 5.

[0138] The frequency ratios between each first conversion frequency and the preset target conversion frequency corresponding to each first node connection edge are calculated respectively, and each frequency ratio is used as the connection edge weight corresponding to the corresponding first node connection edge. For example, if the preset target conversion frequency is 5, the connection edge weight corresponding to the first node connection edge in user B is 0.4.

[0139] Step A33: assigning each of the connection edge weights to the corresponding first node connection edge in the initial user behavior graph to obtain the target user behavior graph.

[0140] In this embodiment, in the initial user behavior graph, each connection edge weight is assigned to the corresponding first node connection edge to obtain the target user behavior graph.

[0141] In this embodiment, each first user behavior node of the target user is generated based on the first stage behavior data; the first conversion order between the first user behavior nodes is determined to connect the first user behavior nodes to obtain initial user behavior graph data; the first conversion frequency data between the first user behavior nodes is determined, and the first behavior graph data is determined based on the first conversion frequency data and the initial user behavior graph data. Since the first behavior graph data is accurately determined, the foundation is laid for accurately determining the first graph representation data.

[0142] Further, based on the first embodiment and the second embodiment of the present application, another embodiment of the present application is provided. In this embodiment, the steps of determining the threshold interval corresponding to the first-stage behavior prediction result, determining the first-stage interaction qualification level of the target user based on the threshold interval, and obtaining the second-stage behavior data of the target user determined based on the first-stage interaction qualification level include:

[0143] Step a1, determining a threshold interval corresponding to the first-stage behavior prediction result, and determining the first-stage interaction qualification level of the target user based on the threshold interval;

[0144] Step a2, obtaining the channel type of the target user with the first-stage interaction qualification level;

[0145] Step a3: if the channel type is a public account channel, outputting preset selected articles to the target user based on the public account channel;

[0146] Step a4, obtaining the second-stage behavior data of the target user determined based on the preset selected articles.

[0147] Specifically, for example, the first-stage behavior prediction result, i.e., the order rate, is 0.3, and the channel type of the target user corresponding to the order rate of 0.3 is obtained as the public account channel, and the corresponding first feedback information is outputting a preset selected article to the target user. After obtaining the output of the preset selected article to the target user, the second-stage behavior data of the target user determined based on the preset selected article is obtained. Specifically, the user generates a (click to read, read for half an hour) behavior based on the preset selected article, and the generated (click to read, read for half an hour) behavior and the first-stage behavior data constitute the second-stage behavior data, and then the second feedback information is determined based on the second-stage behavior data. In this embodiment, the second-stage behavior data is accurately determined.

[0148] Reference Figure 3 , Figure 3 It is a schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present application.

[0149] like Figure 3 As shown, the interactive qualification level dynamic determination device may include: a processor 1001, such as a CPU, a memory 1005, and a communication bus 1002. The communication bus 1002 is used to realize the connection and communication between the processor 1001 and the memory 1005. The memory 1005 may be a high-speed RAM memory or a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0150] Optionally, the interactive qualification level dynamic determination device may also include a rectangular user interface, a network interface, a camera, an RF (Radio Frequency) circuit, a sensor, an audio circuit, a WiFi module, etc. The rectangular user interface may include a display screen (Display), an input submodule such as a keyboard (Keyboard), and the optional rectangular user interface may also include a standard wired interface and a wireless interface. The network interface may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0151] Those skilled in the art will understand that Figure 3 The structure of the interactive qualification level dynamic determination device shown in the figure does not constitute a limitation on the interactive qualification level dynamic determination device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0152] like Figure 3 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, and a dynamic determination program for interactive qualification levels. The operating system is a program that manages and controls the hardware and software resources of the interactive qualification level dynamic determination device, and supports the operation of the interactive qualification level dynamic determination program and other software and / or programs. The network communication module is used to realize the communication between the components inside the memory 1005, and the communication with other hardware and software in the interactive qualification level dynamic determination system.

[0153] exist Figure 3 In the interactive qualification level dynamic determination device shown, the processor 1001 is used to execute the interactive qualification level dynamic determination program stored in the memory 1005 to implement the steps of any of the above-mentioned interactive qualification level dynamic determination methods.

[0154] The specific implementation of the interactive qualification level dynamic determination device of the present application is basically the same as the various embodiments of the interactive qualification level dynamic determination method described above, and will not be repeated here.

[0155] The present application also provides a device for dynamically determining an interaction qualification level, the device for dynamically determining an interaction qualification level comprising:

[0156] an acquisition module, used to acquire the first-stage behavior data of the target user, convert the first-stage behavior data into first graph representation data of a preset data dimension, perform prediction processing on the first graph representation data, and obtain the first-stage behavior prediction result of the first graph representation data;

[0157] A first determination module is used to determine a threshold interval corresponding to the first-stage behavior prediction result, determine a first-stage interaction qualification level of a target user based on the threshold interval, and obtain second-stage behavior data of the target user determined based on the first-stage interaction qualification level;

[0158] The second determination module is used to determine second graph representation data of the second stage behavior data, and determine the second stage interaction qualification level of the target user based on the second graph representation data, wherein the second stage is a time-series continuous stage with the first stage.

[0159] Optionally, the first determining module includes:

[0160] A first determining unit, configured to determine a first-stage interaction qualification level of a first-stage target user based on a numerical value of a first-stage behavior prediction result;

[0161] The second determining unit is configured to determine and parse the first interaction file of the first-stage interaction qualification level to obtain the second-stage behavior data.

[0162] Optionally, the second determining unit includes:

[0163] A first determining subunit, configured to determine a first interaction file of the interaction qualification level at the first stage;

[0164] A second determining subunit is used to determine the channel type of the target user;

[0165] The third determining subunit is used to determine the second-stage behavior data based on the channel type and the first interaction file.

[0166] Optionally, the acquisition module includes:

[0167] A first acquisition unit, configured to acquire first-stage behavior data of a target user and determine first behavior graph data of the first-stage behavior data;

[0168] The third determination unit is used to determine the encoding vector of the first behavior graph data, and perform dimensionality reduction processing on the encoding vector to map it to the preset data dimension to obtain the first graph representation data.

[0169] Optionally, the first acquiring unit includes:

[0170] A generating subunit, configured to generate each first user behavior node of the target user based on the first stage behavior data;

[0171] a fourth determining subunit, configured to determine a first conversion sequence between the first user behavior nodes, so as to connect the first user behavior nodes to obtain initial user behavior graph data;

[0172] The fifth determining subunit is used to determine the first conversion frequency data between the first user behavior nodes, and determine the first behavior graph data based on the first conversion frequency data and the initial user behavior graph data.

[0173] Optionally, the initial user behavior graph data includes a first node connection edge of a first user behavior node;

[0174] The fifth determining subunit is used to implement:

[0175] Determine, based on the first conversion frequency data, a first user behavior conversion frequency corresponding to the first node connection edge;

[0176] Determining a first connection edge weight of the first node connection edge based on the first user behavior conversion frequency and a preset target conversion frequency corresponding to the first node connection edge;

[0177] In the initial user behavior graph data, the first connection edge weight is assigned to the first node connection edge to obtain the first behavior graph data.

[0178] Optionally, the interactive qualification level dynamic determination device further comprises:

[0179] The third determining module is used to determine a second interaction file of the second-stage interaction qualification level, so that the target user can perform various user behaviors based on the second interaction file.

[0180] Optionally, the first determining module further includes:

[0181] A sixth determining subunit, configured to determine a threshold interval corresponding to the first-stage behavior prediction result, and determine a first-stage interaction qualification level of the target user based on the threshold interval;

[0182] A third acquisition unit, configured to acquire the channel type of the target user of the first-stage interaction qualification level;

[0183] A seventh determination subunit, configured to, if the channel type is a public account channel, determine to output a preset selected article to the target user based on the public account channel;

[0184] The fourth acquisition unit is used to acquire the second-stage behavior data of the target user determined based on the preset selected articles.

[0185] The specific implementation of the interactive qualification level dynamic determination device of the present application is basically the same as the various embodiments of the interactive qualification level dynamic determination method described above, and will not be repeated here.

[0186] An embodiment of the present application provides a storage medium, and the storage medium stores one or more programs, and the one or more programs can also be executed by one or more processors to implement the steps of any of the above-mentioned methods for dynamically determining the interaction qualification level.

[0187] The specific implementation of the storage medium of the present application is basically the same as the above-mentioned embodiments of the method for dynamically determining the interaction qualification level, and will not be repeated here.

[0188] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0189] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0190] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0191] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for dynamically determining an interaction qualification level, characterized in that: The method for dynamically determining the interaction qualification level includes: Acquire the first-stage behavior data of the target user, convert the first-stage behavior data into first graph representation data of a preset data dimension, perform prediction processing on the first graph representation data, and obtain the first-stage behavior prediction result of the first graph representation data; Determine the first-stage interaction qualification level of the target user based on the threshold interval corresponding to the first-stage behavior prediction result, so as to determine the second-stage behavior data of the target user; Determine the second graph representation data of the second-stage behavior data, and determine the second-stage interaction qualification level of the target user based on the second graph representation data, wherein the second stage is a time-series continuous stage with the first stage; wherein, in the same manner as determining the first graph representation data of the first-stage behavior data, determine the second graph representation data of the second-stage behavior data, and determine the second-stage behavior prediction result of the second graph representation data, so as to obtain the second-stage behavior prediction result of the second graph representation data, and further obtain the second-stage interaction qualification level; The step of determining the first-stage interaction qualification level of the target user based on the threshold interval corresponding to the first-stage behavior prediction result to determine the second-stage behavior data of the target user includes: Determine a threshold interval corresponding to the first-stage behavior prediction result, and determine the first-stage interaction qualification level of the target user based on the threshold interval; Determine and analyze the first interaction file of the first stage interaction qualification level to obtain the second stage behavior data; wherein the first interaction file includes first feedback information; The step of determining and parsing the first interaction file of the first phase interaction qualification level to obtain the second phase behavior data comprises: A first interaction document for determining the first stage interaction qualification level; Determine the channel type of the target user; The first interaction file is parsed to obtain first feedback information, and second-stage behavior data is determined based on the channel type and the first feedback information.

2. The method for dynamically determining the interaction qualification level according to claim 1, characterized in that: The step of obtaining the first-stage behavior data of the target user and converting the first-stage behavior data into a first graph representation data of a preset data dimension comprises: Acquire the first-stage behavior data of the target user, and determine the first behavior graph data of the first-stage behavior data; Determine the encoding vector of the first behavior graph data, perform dimensionality reduction processing on the encoding vector to map it to the preset data dimension, and obtain first graph representation data.

3. The method for dynamically determining the interactive qualification level according to claim 2, characterized in that: The step of determining the first behavior graph data of the first stage behavior data comprises: Generate each first user behavior node of the target user based on the first stage behavior data; Determining a first conversion order between the first user behavior nodes to connect the first user behavior nodes to obtain initial user behavior graph data; Determine first conversion frequency data between the first user behavior nodes, and determine the first behavior graph data based on the first conversion frequency data and the initial user behavior graph data.

4. The method for dynamically determining the interactive qualification level according to claim 3, characterized in that: The initial user behavior graph data includes a first node connection edge of a first user behavior node; The step of determining the first behavior graph data based on the first conversion frequency data and the initial user behavior graph data comprises: Determine, based on the first conversion frequency data, a first user behavior conversion frequency corresponding to the first node connection edge; Determining a first connection edge weight of the first node connection edge based on the first user behavior conversion frequency and a preset target conversion frequency corresponding to the first node connection edge; In the initial user behavior graph data, the first connection edge weight is assigned to the first node connection edge to obtain the first behavior graph data.

5. The method for dynamically determining the interaction qualification level according to claim 1, characterized in that: After the step of determining the second graph representation data of the second-stage behavior data and determining the second-stage interaction qualification level of the target user based on the second graph representation data, the method further comprises: A second interaction file of the second-stage interaction qualification level is determined, so that the target user can perform various user behaviors based on the second interaction file.

6. The method for dynamically determining the interaction qualification level according to claim 1, characterized in that: The steps of determining a threshold interval corresponding to the first-stage behavior prediction result, determining a first-stage interaction qualification level of a target user based on the threshold interval, and obtaining second-stage behavior data of the target user determined based on the first-stage interaction qualification level include: Determine a threshold interval corresponding to the first-stage behavior prediction result, and determine the first-stage interaction qualification level of the target user based on the threshold interval; Obtaining the channel type of the target user with the first-stage interaction qualification level; If the channel type is a public account channel, outputting preset selected articles to the target user based on the public account channel; The second-stage behavior data of the target user determined based on the preset selected articles is obtained.

7. A device for dynamically determining an interaction qualification level, characterized in that: The interactive qualification level dynamic determination device comprises: An acquisition module, used to acquire the first-stage behavior data of the target user, convert the first-stage behavior data into first graph representation data of a preset data dimension, and input the first graph representation data into a preset classification model to obtain the first-stage behavior prediction result of the first graph representation data; The preset classification model is obtained by iteratively training a preset model to be trained based on graph representation data with preset labels; A first determination module is used to determine a threshold interval corresponding to the first-stage behavior prediction result, determine a first-stage interaction qualification level of a target user based on the threshold interval, and obtain second-stage behavior data of the target user determined based on the first-stage interaction qualification level; A second determination module is used to determine the second graph representation data of the second-stage behavior data, and determine the second-stage interaction qualification level of the target user based on the second graph representation data, wherein the second stage is a time-series continuous stage with the first stage; wherein, in the same manner as determining the first graph representation data of the first-stage behavior data, the second graph representation data of the second-stage behavior data is determined, and the second-stage behavior prediction result of the second graph representation data is determined to obtain the second-stage behavior prediction result of the second graph representation data, and then obtain the second-stage interaction qualification level; The first determining module is used to implement: Determine a threshold interval corresponding to the first-stage behavior prediction result, and determine the first-stage interaction qualification level of the target user based on the threshold interval; Determine and analyze the first interaction file of the first stage interaction qualification level to obtain the second stage behavior data; wherein the first interaction file includes first feedback information; The first determining module is further used to implement: A first interaction document for determining the first stage interaction qualification level; Determine the channel type of the target user; The first interaction file is parsed to obtain first feedback information, and second-stage behavior data is determined based on the channel type and the first feedback information.

8. A device for dynamically determining an interaction qualification level, characterized in that: The interactive qualification level dynamic determination device comprises: a memory, a processor, and a program stored in the memory for implementing the interactive qualification level dynamic determination method. The memory is used to store a program for implementing a method for dynamically determining an interaction qualification level; The processor is used to execute a program for implementing the method for dynamically determining an interaction qualification level, so as to implement the steps of the method for dynamically determining an interaction qualification level according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium stores a program for implementing a method for dynamically determining an interaction qualification level, and the program for implementing a method for dynamically determining an interaction qualification level is executed by a processor to implement the steps of the method for dynamically determining an interaction qualification level as claimed in any one of claims 1 to 6.

Images