Form update methods, devices, computer equipment, readable storage media, and program products

By acquiring and analyzing user behavior data, the form interaction path can be dynamically adjusted, solving the problem of rigid interaction in traditional form systems and improving the flexibility and adaptability of form updates.

CN123114503APending Publication Date: 2026-11-17BEIJING & AIWEIWU TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610424793.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-04-01
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Traditional form systems cannot adjust information collection strategies in real time based on users' hesitation and modification behavior during the form filling process, resulting in rigid interaction.

Method used

By acquiring the target account's behavioral data in the form interface, a current state vector is constructed. Based on the state vector, interaction path planning is performed, interaction action codes are generated, and form update data is generated according to the codes. Finally, the data is rendered and displayed in the form interface.

Benefits of technology

It enables flexible form updates, reduces the difficulty of secondary development and maintenance costs, and is suitable for more personalized business scenarios.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123114503A_ABST
    Figure CN123114503A_ABST
Patent Text Reader

Abstract

This application relates to a form update method, apparatus, computer device, readable storage medium, and program product. The method includes: acquiring behavioral data of a target account in a form interface; constructing a current state vector of the target account using the behavioral data; planning an interaction path based on the current state vector to generate corresponding interaction action codes; generating form update data based on the interaction action codes; and sending the form update data to the form interface for rendering and display. This method can improve interaction flexibility.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a form updating method, apparatus, computer device, computer-readable storage medium, and computer program product. Background Technology

[0002] Forms are interactive components used for information collection, relying on data structures, front-end rendering, and back-end storage. Traditional form-based systems store form configurations in a database, use front-end components to render pages based on these configurations for user interaction, and save user-submitted data to a fixed-structure database table. However, because the navigation paths between options and questions are hard-coded and fixed during the form development phase, traditional forms cannot adjust their information collection strategies in real time based on user hesitation or modifications during the form-filling process, resulting in a rigid interaction mechanism. Summary of the Invention

[0003] Therefore, it is necessary to provide a form update method, apparatus, computer device, computer-readable storage medium, and computer program product that can improve the flexibility of interaction in response to the above-mentioned technical problems.

[0004] Firstly, this application provides a form updating method, including:

[0005] Obtain behavioral data of the target account in the form interface;

[0006] Construct the current state vector of the target account using the behavioral data;

[0007] Based on the current state vector, perform interactive path planning and generate corresponding interactive action codes;

[0008] Generate form update data based on the interaction action encoding;

[0009] The updated form data is sent to the form interface for rendering and display.

[0010] In one embodiment, the step of planning the interaction path based on the current state vector and generating the corresponding interaction action code includes:

[0011] Based on each candidate question in the candidate question pool and the current state vector, predict the future state vector corresponding to each candidate question;

[0012] Based on the future state vector and the current state vector, calculate the information gain parameter for each of the questions to be selected;

[0013] The target question is determined from the candidate question pool based on the information gain parameter.

[0014] The action of selecting the target problem in the next step is encoded to obtain the interaction action code.

[0015] In one embodiment, calculating the information gain parameter for each of the candidate problems based on the future state vector and the current state vector includes:

[0016] Based on the current state vector, calculate the account profile information entropy of the target account;

[0017] Based on the future state vector corresponding to each of the candidate questions, calculate the dynamic profile information entropy corresponding to each of the candidate questions.

[0018] Determine the amount of entropy decrease between the account profile information entropy and the dynamic profile information entropy, and use the amount of entropy decrease as the information gain parameter corresponding to the question to be selected.

[0019] In one embodiment, constructing the current state vector of the target account using the behavioral data includes:

[0020] Explicit feature data and implicit behavioral feature data are extracted from the behavioral data respectively;

[0021] The explicit feature data and the implicit behavioral feature data are vectorized and encoded respectively, and then combined to obtain the current state vector.

[0022] In one embodiment, generating form update data based on the interaction action encoding includes:

[0023] Based on the current state vector, the corresponding form interaction data is retrieved from the preset knowledge base;

[0024] The interactive action is encoded into corresponding interactive action instruction data, and the interactive action instruction data is combined with the form interaction data to obtain context prompt words.

[0025] The form update data is generated based on the contextual prompts.

[0026] In one embodiment, before sending the form update data to the form interface for rendering and display, the method further includes:

[0027] The updated form data is validated based on preset rules to obtain the validation result.

[0028] If the verification result is successful, the operation of sending the updated form data to the form interface for rendering and display is performed.

[0029] Secondly, this application also provides a form updating device, comprising:

[0030] The data acquisition module is used to acquire behavioral data of the target account in the form interface;

[0031] A state construction module is used to construct the current state vector of the target account using the behavioral data;

[0032] The action coding module is used to perform interaction path planning based on the current state vector and generate corresponding interaction action codes.

[0033] The data generation module is used to generate form update data based on the interaction action encoding;

[0034] The form update module is used to send the form update data to the form interface for rendering and display.

[0035] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the form update method described in any of the embodiments of the first aspect.

[0036] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the form update method described in any of the embodiments of the first aspect.

[0037] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the form update method described in any of the embodiments of the first aspect.

[0038] The aforementioned form update method, apparatus, computer equipment, computer-readable storage medium, and computer program product acquire behavioral data of the target account in the form interface; construct the current state vector of the target account using the behavioral data; plan the interaction path based on the current state vector to generate corresponding interaction action codes; generate form update data based on the interaction action codes; and send the form update data to the form interface for rendering and display. This allows for dynamic adjustment of the interaction path based on the real-time behavior of the target account in the form interface, improving the flexibility of form updates. Furthermore, by decoupling the form update business logic from the form presentation layer at the software logic level, the aforementioned form update method allows for the integration of newly added form content without redeveloping jump path code, reducing the difficulty and maintenance cost of secondary form development and making it suitable for more personalized business scenarios. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a diagram illustrating the application environment of a form update method in one embodiment.

[0041] Figure 2 This is a flowchart illustrating a form update method in one embodiment;

[0042] Figure 3 This is a flowchart illustrating the interactive action encoding generation steps in one embodiment;

[0043] Figure 4 This is a flowchart illustrating the information gain parameter calculation steps in one embodiment;

[0044] Figure 5 This is a flowchart illustrating the form update data generation steps in one embodiment;

[0045] Figure 6 This is a system architecture diagram of a distributed multi-agent system 600 in one embodiment;

[0046] Figure 7 This is a structural block diagram of a form updating device 700 in one embodiment;

[0047] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0049] It should be noted that the terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" as used in this application refers to two or more. The term "and / or" as used in this application refers to one of the solutions, or any combination of multiple solutions. The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of related data must comply with relevant regulations.

[0050] The form update method provided in this application embodiment can be applied to, for example, Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or placed on a cloud or other network server. Specifically, terminal 102 can display a form interface and collect behavioral data of the target account within the form interface. Server 104 can receive the behavioral data sent by terminal 102, construct the current state vector of the target account using the behavioral data, perform interaction path planning based on the current state vector, generate corresponding interaction action codes, and generate form update data based on the interaction action codes. Server 104 can send the form update data to terminal 102, enabling terminal 102 to render and display the form update data on the form interface, completing the form interface update operation.

[0051] The terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart TVs, smart in-vehicle devices, and projection devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted displays. Head-mounted displays can be virtual reality (VR) devices, augmented reality (AR) devices, and smart glasses. The server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0052] In one exemplary embodiment, such as Figure 2 As shown, a form update method is provided, which is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps S202 to S210. Wherein:

[0053] Step S202: Obtain the target account's behavioral data in the form interface.

[0054] Behavioral data may include, but is not limited to, the target account's behavior of lingering on form options, modifying form options, clicking form options, the time spent entering answers to form questions, the content entered, and the content selected.

[0055] For example, the server can respond to the answer submission operation triggered by the target account in the form interface forwarded by the terminal, and receive the behavioral data of the target account in the form interface reported by the terminal. Alternatively, the server can also actively read the behavioral data of the target account in the form interface from the terminal at preset intervals.

[0056] Step S204: Construct the current state vector of the target account using behavioral data.

[0057] For example, the server can extract features from the behavioral data to obtain feature data of the target account. The feature data is then vectorized to obtain the current state vector of the target account. This feature data may include, but is not limited to, the dwell time of the behavior, the number of times form options were modified, the duration of clicking form options, the identifier of the selected form option, and the value of the answers to the entered form questions.

[0058] Step S206: Based on the current state vector, perform interactive path planning and generate corresponding interactive action codes.

[0059] For example, the server can input the current state vector into a reinforcement learning model. Based on a pre-defined Markov decision policy, the reinforcement learning model predicts the probability of executing different form update strategies under the current state vector, thereby determining the optimal interaction action for the next step. This optimal interaction action is then mapped to a corresponding interaction action code and output. Alternatively, the server can construct a cue word using the current state vector, input the cue word into a larger model, and use the larger model to obtain the optimal interaction action for the next step based on zero-shot inference, thereby outputting the corresponding interaction action code.

[0060] Step S208: Generate form update data based on the interaction action code.

[0061] For example, the server can map interactive actions to corresponding form update data based on a preset mapping relationship between encoding and data. This form update data can be used to update form questions, form options, or the question guidance text on the form interface.

[0062] Step S210: Send the updated form data to the form interface for rendering and display.

[0063] For example, the server can send form update data to the terminal to instruct the terminal to render and display the form update data in the form interface, thereby updating the form interface display. Subsequently, the server can continue to receive the target account's behavior data in the form interface after it has been updated with the form update data, and repeat steps S202 to S210 to achieve a new round of form updates.

[0064] The above form update method involves acquiring the target account's behavioral data within the form interface; constructing the target account's current state vector using this data; planning interaction paths based on the current state vector to generate corresponding interaction action codes; generating form update data based on these codes; and sending the updated data to the form interface for rendering and display. This approach dynamically adjusts the interaction path based on the target account's real-time behavior within the form interface, improving the flexibility of form updates. Furthermore, by decoupling the form update business logic from the form presentation layer at the software logic level, this method allows for the integration of newly added form content without redeveloping redirection path code, reducing the difficulty and maintenance cost of secondary form development and making it suitable for more personalized business scenarios.

[0065] In one exemplary embodiment, such as Figure 3 As shown, step S206 includes steps S302 to S308. Wherein:

[0066] Step S302: Based on each candidate question in the candidate question pool and the current state vector, predict the future state vector corresponding to each candidate question.

[0067] For example, the server can store a pool of candidate questions for a form. A state transition function is constructed using a Markov decision strategy. For each candidate question in the pool, the following operations are performed: the state transition function is used to process the current state vector and the candidate question to predict and generate the next future state vector of the target account after selecting that candidate question in the current state vector.

[0068] Step S304: Based on the future state vector and the current state vector, calculate the information gain parameter corresponding to each problem to be selected.

[0069] For example, the server can calculate the difference between the future state vector and the current state vector, using this difference as the information gain parameter for the question to be chosen corresponding to the future state vector. Alternatively, the server can process the future state vector and the current state vector based on a preset reward function to obtain the information gain parameter for the question to be chosen corresponding to the future state vector. Alternatively, the server can also use the future state vector and the current state vector to measure the change in information entropy of the target account after selecting a question, and use this change in information entropy as the information gain parameter for the question to be chosen corresponding to the future state vector.

[0070] Step S306: Determine the target problem from the candidate problem pool based on the information gain parameter.

[0071] For example, the server can select the candidate question with the highest information gain parameter from each candidate question in the candidate question pool as the target question. Alternatively, it can select candidate questions with an information gain parameter greater than a preset threshold as the target question. Alternatively, it can select candidate questions with information gain parameters in the top 30% as the target questions and determine the display order of each target question according to the magnitude of the information gain parameter.

[0072] Step S308: Encode the action of selecting the target problem in the next step to obtain the interaction action code.

[0073] For example, the server can use the action template for selecting a new form question in the next step, combined with the identifier of the target question, its storage address, or the answer content value, to generate the corresponding interactive action code based on preset action coding rules.

[0074] In this embodiment, by using Markov decision prediction to select the future state vector of the target account after each candidate problem is selected, the information gain parameter for selecting each candidate problem in the next step is evaluated based on the future state vector and the current state vector, and the optimal target problem is determined according to the information gain parameter, and the corresponding interactive action code is generated. This can reduce the state modeling complexity and computational overhead of the future state vector, quickly and accurately determine the optimal target problem, and thus improve the accuracy of the interactive action code.

[0075] In one exemplary embodiment, such as Figure 4 As shown, step S304 includes steps S402 to S406. Wherein:

[0076] Step S402: Calculate the account profile information entropy of the target account based on the current state vector.

[0077] Optionally, in some implementations, the server can construct a probability distribution of the target account's account profile using the current state vector. The probability distribution of the account profile is then processed using the following formula to obtain the account profile information entropy of the target account:

[0078] ,in, It is the entropy of account profile information. It is the current state vector. This represents the i-th case in the probability distribution of the account profile, where n represents all possible cases of the probability distribution of the account profile. It represents the probability that the account profile belongs to the i-th type.

[0079] Step S404: Based on the future state vector corresponding to each question to be selected, calculate the dynamic profile information entropy corresponding to each question to be selected.

[0080] For example, the server can perform the following operations on the future state vector corresponding to each question to be selected: referring to the above calculation method of account profile information entropy, construct the probability distribution of the new account profile of the target account using the future state vector, and calculate the corresponding dynamic profile information entropy using the probability distribution of the new account profile and referring to the above calculation formula of account profile information entropy.

[0081] Step S406: Determine the entropy decrease between the account profile information entropy and the dynamic profile information entropy, and use the entropy decrease as the information gain parameter corresponding to the question to be selected.

[0082] For example, the server can subtract the dynamic profile information entropy from the account profile information entropy to obtain the entropy reduction. This entropy reduction is then used as the information gain parameter corresponding to the question to be chosen. Optionally, in some implementations, the Markov decision strategy can also be expressed as follows:

[0083] , ,in, For information gain parameters, Information entropy for account profiling The dynamic profile information entropy corresponding to the question q to be selected. This represents a Markov decision strategy. Therefore, this Markov decision strategy is used to select the candidate problem that causes the fastest decrease in entropy as the optimal objective problem.

[0084] In this embodiment, by using state vectors to model the account profile of the target account, the decrease in information entropy of the account profile between the current state vector and the future state vector is calculated as the information gain parameter. This makes it easier to select the candidate problem that causes the fastest decrease in entropy as the optimal target problem, so as to build the most accurate account profile of the target account in the fewest form interaction rounds.

[0085] In an exemplary embodiment, step S204 includes: extracting explicit feature data and implicit behavioral feature data from the behavioral data respectively; vectorizing and encoding the explicit feature data and implicit behavioral feature data respectively, and combining them to obtain the current state vector.

[0086] For example, the server extracts the selected form option identifier and the answer value of the input form question from the raw data stream of behavioral data as explicit feature data, and extracts the dwell time, the number of times the form option was modified, and the duration of clicking the form option as implicit behavioral feature data. The explicit and implicit behavioral feature data are then vectorized to obtain explicit feature vectors and implicit behavioral feature vectors. The explicit and implicit behavioral feature vectors are combined in a preset order to obtain the current state vector.

[0087] In this embodiment, by extracting explicit feature data and implicit behavioral feature data from behavioral data to construct the current state vector, it is possible to comprehensively cover observable interaction information and potential user intentions, thereby improving the completeness and accuracy of the current state vector.

[0088] In one exemplary embodiment, such as Figure 5 Step S208 includes steps S502 to S506. Wherein:

[0089] Step S502: Based on the current state vector, retrieve the corresponding form interaction data from the preset knowledge base.

[0090] Step S504: Convert the interactive action code into corresponding interactive action instruction data, and combine the interactive action instruction data with the form interactive data to obtain context prompt words.

[0091] Step S506: Generate form update data based on contextual prompts.

[0092] For example, the server can use retrieval-augmented generation to perform similarity matching between the current state vector and historical form interaction records stored in a preset knowledge base, obtaining form interaction data corresponding to the current state vector. Based on a preset mapping relationship between encoding and data, the interaction action encoding is mapped to corresponding interaction action instruction data. The interaction action instruction data is combined with the form interaction data according to a preset prompt word template to obtain context prompt words. The context prompt words are then input into the generative model to obtain the corresponding form update data.

[0093] In this embodiment, by retrieving form interaction data from a preset knowledge base, constructing contextual prompts using the form interaction data and interaction action instruction data, and generating form update data, the generation process of form update data can be enhanced with the context of an external knowledge base, thereby improving the accuracy of the form update data.

[0094] In an exemplary embodiment, before step S210, the method further includes: validating the form update data based on preset rules to obtain a validation result; and if the validation result is successful, performing the operation of sending the form update data to the form interface for rendering and display.

[0095] For example, the server may store preset rules for detecting the compliance of the content displayed on the form interface. The form update data is validated using the preset rules to obtain a validation result indicating whether the form update data passes the compliance check. If the validation result is successful, the operation of sending the form update data to the form interface for rendering and display is performed.

[0096] Optionally, in some implementations, if the validation result is unsuccessful, the validation result and the updated form data are re-entered into the generative model to regenerate new updated form data. The new updated form data is then validated using preset rules until the validation result is successful.

[0097] In this embodiment, by performing compliance verification using preset rules before sending form update data, the problem of non-compliant content in the subsequently updated form interface can be avoided.

[0098] In one exemplary embodiment, such as Figure 6 As shown, a distributed multi-agent system 600 is provided, including: a perceptual agent 602, a decision-making agent 604, a generating agent 606, and a verification agent 608.

[0099] Among them, the perceptual agent 602 can be used to acquire multimodal behavioral data of the target account on the form interface, such as text flow, click flow and dwell time, and input the behavioral data into the decision agent 604.

[0100] Decision agent 604 can extract explicit and implicit behavioral feature data from behavioral data. It vectorizes and encodes both explicit and implicit behavioral feature data, combining them to obtain the current state vector. Using a Markov decision model, it calculates the account profile information entropy of the target account based on the current state vector. Based on each candidate question in the pool and the current state vector, it predicts the future state vector for each candidate question. Based on the future state vector for each candidate question, it calculates the dynamic profile information entropy for each candidate question. It determines the entropy decrease between the account profile information entropy and the dynamic profile information entropy, using this decrease as the information gain parameter for the candidate question. It selects the candidate question with the largest entropy decrease as the target question. It maps the form update strategy for selecting the target question to the corresponding interaction action code. The interaction action code is then input into agent 606.

[0101] The generating agent 606 can interact with an external knowledge base through a retrieval engine. It matches the current state vector with a vector database of historical form interaction records in the external knowledge base to retrieve similar vector data corresponding to the current state vector. It then obtains interaction template data from a business template library and combines the similar vector data with the interaction template data to obtain form interaction data. The interaction actions are encoded and converted into corresponding interaction action command data, which is then combined with the form interaction data to obtain context prompts. Based on the context prompts, form update data is generated. The form update data is rendered to obtain the form content to be published. Finally, the form content to be published is sent to the verification agent 608.

[0102] The validation agent 608 validates the content of the form to be published based on preset rules and obtains the validation result. If the validation result is successful, the form content to be published is sent to the form interface for rendering and display. If the validation result is unsuccessful, the form content to be published and the validation result are sent to the generation agent 606 to instruct the generation agent 606 to regenerate new form update data.

[0103] In this embodiment, the form is transformed from a static questionnaire into a real-time game interaction system through the aforementioned distributed multi-agent architecture. This system can synchronize the physical behavioral characteristics and semantic intent of the target account on the form interface, achieving dual-path perception of behavior and semantics. By adopting an information gain-based question selection strategy, the system can dynamically calculate and push the target questions that best reduce the uncertainty of the account profile. This allows for dynamic adjustment of the difficulty and order of questions on the form interface, reducing redundancy and improving the efficiency of form data collection.

[0104] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0105] Based on the same inventive concept, this application also provides a form updating apparatus for implementing the form updating method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more form updating apparatus embodiments provided below can be found in the limitations of the form updating method described above, and will not be repeated here.

[0106] In one exemplary embodiment, such as Figure 7 As shown, a form update device 700 is provided, including: a data acquisition module 702, a state construction module 704, an action encoding module 706, a data generation module 708, and a form update module 710, wherein:

[0107] The data acquisition module 702 is used to acquire the behavioral data of the target account in the form interface.

[0108] State construction module 704 is used to construct the current state vector of the target account using behavioral data.

[0109] The action coding module 706 is used to plan the interaction path based on the current state vector and generate the corresponding interaction action code.

[0110] The data generation module 708 is used to generate form update data based on the interaction action encoding.

[0111] The form update module 710 is used to send the updated form data to the form interface for rendering and display.

[0112] In an exemplary embodiment, the action encoding module 706 includes: a state prediction unit, configured to predict the future state vector corresponding to each candidate question based on each candidate question in the candidate question pool and the current state vector; a gain calculation unit, configured to calculate the information gain parameter corresponding to each candidate question based on the future state vector and the current state vector; and a question selection unit, configured to determine the target question from the candidate question pool based on the information gain parameter. The action of selecting the target question in the next step is encoded to obtain the interaction action code.

[0113] In an exemplary embodiment, the gain calculation unit is further configured to calculate the account profile information entropy of the target account based on the current state vector. Based on the future state vector corresponding to each candidate question, the dynamic profile information entropy corresponding to each candidate question is calculated separately. The entropy decrease between the account profile information entropy and the dynamic profile information entropy is determined, and the entropy decrease is used as the information gain parameter corresponding to the candidate question.

[0114] In an exemplary embodiment, the state construction module 704 is further configured to extract explicit feature data and implicit behavioral feature data from the behavioral data, respectively. The explicit feature data and implicit behavioral feature data are then vectorized and encoded, and combined to obtain the current state vector.

[0115] In an exemplary embodiment, the data generation module 708 is further configured to retrieve corresponding form interaction data from a preset knowledge base based on the current state vector. The interaction action is encoded and converted into corresponding interaction action instruction data, and the interaction action instruction data is combined with the form interaction data to obtain context prompts. Based on the context prompts, form update data is generated.

[0116] In one exemplary embodiment, the form update module 710 is further configured to validate the form update data based on preset rules and obtain a validation result. If the validation result is successful, the form update data is sent to the form interface for rendering and display.

[0117] Each module in the aforementioned form updating device 700 can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0118] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 8 As shown, this computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores behavioral data, current state vectors, interactive action codes, form update data, etc. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When executed by the processor, the computer program implements a form update method.

[0119] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0120] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0121] In one exemplary embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above-described method embodiments.

[0122] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0123] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0124] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0125] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A form update method, characterized in that, The method includes: Obtain behavioral data of the target account in the form interface; Construct the current state vector of the target account using the behavioral data; Based on the current state vector, perform interactive path planning and generate corresponding interactive action codes; Generate form update data based on the interaction action encoding; The updated form data is sent to the form interface for rendering and display.

2. The method according to claim 1, characterized in that, The step of planning the interaction path based on the current state vector and generating the corresponding interaction action code includes: Based on each candidate question in the candidate question pool and the current state vector, predict the future state vector corresponding to each candidate question; Based on the future state vector and the current state vector, calculate the information gain parameter for each of the questions to be selected; The target question is determined from the candidate question pool based on the information gain parameter. The action of selecting the target problem in the next step is encoded to obtain the interaction action code.

3. The method according to claim 2, characterized in that, The step of calculating the information gain parameter for each of the candidate problems based on the future state vector and the current state vector includes: Based on the current state vector, calculate the account profile information entropy of the target account; Based on the future state vector corresponding to each of the candidate questions, calculate the dynamic profile information entropy corresponding to each of the candidate questions. Determine the amount of entropy decrease between the account profile information entropy and the dynamic profile information entropy, and use the amount of entropy decrease as the information gain parameter corresponding to the question to be selected.

4. The method according to claim 1, characterized in that, The step of constructing the current state vector of the target account using the behavioral data includes: Explicit feature data and implicit behavioral feature data are extracted from the behavioral data respectively; The explicit feature data and the implicit behavioral feature data are vectorized and encoded respectively, and then combined to obtain the current state vector.

5. The method according to claim 1, characterized in that, The step of generating form update data based on the interaction action encoding includes: Based on the current state vector, the corresponding form interaction data is retrieved from the preset knowledge base; The interactive action is encoded into corresponding interactive action instruction data, and the interactive action instruction data is combined with the form interaction data to obtain context prompt words. The form update data is generated based on the contextual prompts.

6. The method according to claim 1, characterized in that, Before sending the form update data to the form interface for rendering and display, the method further includes: The updated form data is validated based on preset rules to obtain the validation result. If the verification result is successful, the operation of sending the updated form data to the form interface for rendering and display is performed.

7. A form updating device, characterized in that, The device includes: The data acquisition module is used to acquire behavioral data of the target account in the form interface; A state construction module is used to construct the current state vector of the target account using the behavioral data; The action coding module is used to perform interaction path planning based on the current state vector and generate corresponding interaction action codes. The data generation module is used to generate form update data based on the interaction action encoding; The form update module is used to send the form update data to the form interface for rendering and display.

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 method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.