Form filling method and device, electronic equipment, storage medium and program product

Abstract

The embodiment of the invention discloses a form filling method and device, electronic equipment, a storage medium and a program product. The method comprises the steps that a form filling interface is displayed, the form filling interface comprises a to-be-filled form, and the to-be-filled form comprises to-be-filled form items; in response to an automatic filling instruction of a to-be-filled form, receiving source data input through the form filling interface, and obtaining form information of the to-be-filled form; based on first content data recommended by an artificial intelligence AI component for the form item, second content data used for filling in the form item is obtained, the second content data is filled in the form item, and the first content data is obtained by the AI component from the source data based on the form information. In this way, the AI technology can be used for performing form filling on the semantic comprehension ability and the information extraction ability of the natural language, and therefore the problems that a traditional form filling mode is low in input efficiency, high in error rate, poor in experience and the like are solved.

Description

Technical Field

[0001] This document relates to the field of artificial intelligence technology, and in particular to a form filling method, device, electronic device, storage medium and program product. Background Art

[0002] aPaaS (Application Platform as a Service) is a cloud computing service model that allows users to quickly build an application that meets business needs with zero code / no code (No-code) or low code (Low-code), thereby improving team development efficiency and enhancing enterprise effectiveness.

[0003] In some aPaaS-based applications, form filling is often required. Traditional form filling methods require users to manually enter the content of each form item, which has problems such as low input efficiency, high error rate, and poor user experience. Summary of the Invention

[0004] The purpose of the embodiments of this specification is to provide a form filling method, device, electronic device, storage medium and program product, which combine AI (Artificial Intelligence) technology with forms in applications built on aPaaS, and use AI technology's semantic understanding and information extraction capabilities of natural language to fill in forms, thereby solving the problems of low input efficiency, high error rate and poor experience in traditional form filling methods.

[0005] In order to achieve the above objectives, the embodiments of this specification adopt the following technical solutions: In a first aspect, a form filling method is provided, comprising: Displaying a form filling interface, wherein the form filling interface includes a form to be filled in, and the form to be filled in includes form items to be filled in; In response to an automatic filling instruction for the form to be filled in, receiving source data input through the form filling interface, and acquiring form information of the form to be filled in, the form information including form structure information; Based on the first content data recommended by the artificial intelligence AI component for the form item, second content data for filling in the form item is obtained, and the second content data is filled into the form item, wherein the first content data is obtained by the AI ​​component from the source data based on the form information.

[0006] In a second aspect, a form filling method is provided, comprising: Receive a form filling request sent by the AI ​​component, wherein the form filling request carries form information and source data of a form to be filled in, wherein the form to be filled in includes form items to be filled in, and the form information includes form structure information; Parsing the source data based on the form information using a large model to obtain first content data of the form item; The first content data is returned to the AI ​​component, where the first content data is used to fill in the form item.

[0007] In a third aspect, a form filling device is provided, comprising: A first display module is used to display a form filling interface, wherein the form filling interface includes a form to be filled in, and the form to be filled in includes form items to be filled in; a first acquisition module, configured to receive source data input through the form filling interface in response to an automatic filling instruction for the form to be filled in, and acquire form information of the form to be filled in, the form information including form structure information; A filling module is used to obtain second content data for filling in the form item based on the first content data recommended by the artificial intelligence AI component for the form item, and fill the second content data into the form item, wherein the first content data is obtained by the AI ​​component from the source data based on the form information.

[0008] In a fourth aspect, a form filling device is provided, comprising: A first receiving module is configured to receive a form filling request sent by an AI component, wherein the form filling request carries form information and source data of a form to be filled in, wherein the form to be filled in includes form items to be filled in, and the form information includes form structure information; a parsing module, configured to parse the source data based on the form information using a large model to obtain first content data of the form item; The first sending module is used to return the first content data to the AI ​​component, where the first content data is used to fill in the form item.

[0009] According to a fifth aspect, an electronic device is provided, including: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the form filling method provided in the first aspect, or to implement the form filling method provided in the second aspect.

[0010] In a fourth aspect, a computer-readable storage medium is provided. When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the form filling method provided in the first aspect, or execute the form filling method provided in the second aspect.

[0011] In a fifth aspect, a computer program product is provided, comprising a non-transitory computer-readable storage medium storing a computer program, wherein the computer program is operable to cause a computer to execute some or all of the steps in the form filling method provided in the first aspect, or to execute some or all of the steps in the form filling method provided in the second aspect.

[0012] The solution of the embodiment of this specification is that the client displays a form filling interface, responds to the automatic filling instruction of the form to be filled in the form filling interface, receives the source data input through the form filling interface, and obtains form information such as the form structure information of the form to be filled in; further considering that AI technology has excellent effects in the semantic understanding and information extraction of natural language, and is good at identifying content in different formats, therefore, an AI component is configured on the client, and through the AI ​​component, the content data of the form items of the form to be filled in is obtained from the source data based on the form information, and the form items are filled in accordingly. In this way, AI technology is combined with forms, and the semantic understanding and information extraction capabilities of AI technology for natural language are used to fill in forms, thereby solving the problems of low input efficiency, high error rate and poor experience existing in the traditional form filling method. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The drawings described herein are used to provide a further understanding of this specification and constitute a part of this specification. The exemplary embodiments and descriptions of this specification are used to explain this specification and do not constitute an improper limitation of this specification. In the drawings: Figure 1 A schematic diagram of an implementation environment provided for one embodiment of this specification; Figure 2 A schematic diagram of an implementation environment provided for another embodiment of this specification; Figure 3 A schematic diagram of an implementation environment provided for yet another embodiment of this specification; Figure 4 A flowchart of a form filling method provided in one embodiment of this specification; Figure 5A One of the schematic diagrams of a form filling interface provided in one embodiment of this specification; Figure 5B This is a second schematic diagram of a form filling interface provided in one embodiment of this specification; Figure 5C A third schematic diagram of a form filling interface provided in accordance with an embodiment of this specification; Figure 5D A fourth schematic diagram of a form filling interface provided in accordance with an embodiment of this specification; Figure 5E A fifth schematic diagram of a form filling interface provided in accordance with an embodiment of this specification; Figure 6 A flowchart of a form filling method provided in accordance with another embodiment of the present disclosure; Figure 7 A flowchart of a form filling method provided in yet another embodiment of this specification; Figure 8 A flow chart of a streaming speech recognition method provided in accordance with an embodiment of the present disclosure; Figure 9 A schematic structural diagram of a form filling device provided in one embodiment of this specification; Figure 10 A schematic structural diagram of a form filling device provided in another embodiment of this specification; Figure 11 The present invention provides a schematic structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0014] To make the purpose, technical solutions, and advantages of this specification more clear, the technical solutions of this specification will be clearly and completely described below in conjunction with the specific embodiments of this specification and the corresponding drawings. Obviously, the embodiments described are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this document.

[0015] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description. The term "responsive to" is used to indicate the conditions or states on which the operations performed depend. When the dependent conditions or states are met, the one or more operations performed may be in real time or with a set delay. Unless otherwise specified, there is no restriction on the order of the multiple operations performed.

[0016] It should be noted that the concepts of "first" and "second" mentioned in this document are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0017] It should be noted that the modifiers such as "one" and "multiple" mentioned in this document are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly stated in the context, they should be understood as "one or more".

[0018] The names of the messages or information exchanged between multiple devices in the embodiments of this document are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0019] As mentioned above, the traditional form filling method requires users to manually enter the content of each form item, which has the following problems: (1) Low input efficiency: For example, in scenarios such as procurement and delivery, offline document formats vary, resulting in low efficiency in manually filling out forms; (2) High error rate: For information such as ID card number and bank card number, the error rate is high when entering it manually; (3) Poor user experience: Form items such as travel application forms and vacation application forms can often be explained clearly in one sentence, but manual operation does not provide a good user experience.

[0020] In view of this, the embodiments of this specification propose a form filling method. Taking into account that AI technology has excellent effects in semantic understanding and information extraction of natural language, it is good at identifying content in different formats. To this end, by developing an AI component that integrates AI technology, the AI ​​component obtains the content data of the form items to be filled in the form from the source data provided by the user, and fills in the form items accordingly. In this way, AI technology is combined with the forms in the application built on aPaaS, and the semantic understanding and information extraction capabilities of AI technology for natural language are used to fill in the form, thereby solving the problems of low input efficiency, high error rate and poor experience existing in the traditional form filling method.

[0021] Before introducing the form filling method provided by the embodiment of this specification in detail, the implementation environment involved in the embodiment of this specification is introduced first. Figure 1 , is a schematic diagram of an implementation environment involved in an embodiment of this specification, which includes a client 10 and a server 20.

[0022] Client 10 can be a client of an aPaaS system. Server 20 can include at least one of a single server, multiple servers, a cloud computing server, and a virtualization center. For example, server 20 can include a server (i.e., a first server) 21 of the aPaaS system and a server (i.e., a second server) 22 for providing AI services. Client 10 is connected to server 20 via a wireless network or a wired network.

[0023] The client 10 is configured with an AI component for automatically filling in business components such as forms, tables, and charts. Figure 2 As shown, the AI ​​component can interact with the business component to obtain component information (such as structural information) of the business component and fill the business component with corresponding content data. The AI ​​component can also call the server 20 to request the server 20 to provide content data to fill in the business component using the large model.

[0024] In one implementation, the AI ​​component may include an AI button and an AI agent. The AI ​​button is used to trigger the automatic filling function of the business component, and the AI ​​agent is used to implement the automatic filling function of the business component. The client 10 can display the business component to be filled in (such as a form to be filled in) and the AI ​​button. In response to the triggering operation of the AI ​​button, a first control is displayed for the user to input the source data required for the business component and obtain the component information of the business component (such as the structural information of the business component, the historical content data filled in the business component, etc.); then, the AI ​​agent calls the server 20 based on the source data and component information to request the server 20 to obtain the content data to be filled in the business component from the source data based on the component information through the large model; further, the AI ​​agent fills in the business component based on the content data.

[0025] Considering that the mutual dependence between AI components and business components may cause some problems, such as excessive coupling between the two, difficulty in maintenance, and difficulty in expansion, in order to avoid these problems, such as Figure 2 As shown in the figure, an AI interface layer is designed between the AI ​​component and the business component.

[0026] The AI ​​interface layer serves as a bridge between AI components and business components, eliminating direct dependencies between them. The AI ​​interface layer defines a standard interface (called the first interface), which encapsulates the methods required for interaction between AI and business components. Examples include, but are not limited to: generating a pattern or structure based on configuration, callbacks during the generation process, callbacks for accepting and rejecting operations, AI return information, data updates, and callbacks for errors. AI components rely on the AI ​​interface layer, while business components implement the AI ​​interface layer's interface (called the second interface). This allows AI components to associate with any business component, and business components to be associated with any AI component.

[0027] In another implementation, Figure 3 As shown, client 10 also adds the ability for form classes to interact with AI components. The form class implements the first interface in the AI ​​interface layer. This allows any business component, such as a form that inherits from the form class, to directly interact with AI components, such as process forms and future third-party forms.

[0028] Specifically, the form class is a class for managing form-related logic, and provides some methods, such as a first method for returning an object that implements the first interface, and a second method for returning structural information of the form.

[0029] The form class can include the field detail class and the field class. The field detail class is used to manage the field details of the form and provides methods for returning field details. The field class is used to manage fields and provides methods for returning the structure definition of the field.

[0030] In yet another implementation, Figure 3 As shown, a draft capability interface is also defined in the AI ​​interface layer, which is used to maintain the content data of the form items filled in by the AI ​​component before adoption, such as maintaining the draft value status of the form to be filled in, submitting the content data filled in by the AI ​​component as real data, clearing the content data filled in by the AI ​​component, etc., so as to achieve a more flexible automatic filling function.

[0031] Based on the implementation environment introduced above, the form filling method provided in the embodiment of this specification is described in detail with reference to the accompanying drawings.

[0032] Please refer to Figure 4 , is a flow chart of a form filling method provided in one embodiment of this specification, the method is applied to the client 10, and the method may include the following steps: S402, displaying a form filling interface.

[0033] The form filling interface includes a form to be filled in, and the form to be filled in includes form items to be filled in. The form to be filled in can be a form to be filled in for the first time or a form to be filled in for a second time, and this specification does not limit this. The number of form items can be one or more, and this embodiment of the specification does not limit this.

[0034] For example, Figure 5A The form filling interface 50 shown includes a form 51 to be filled out. Form 51 is used to collect supplier information. Form 51 includes multiple form items 511 for company name, unified social credit code, business scope, legal representative, etc. Each form item 511 includes a corresponding control 512, which is used to enter the content data of form item 511.

[0035] In actual applications, the type of control 512 corresponding to the form item 511 can be selected according to actual needs, such as a text box, radio button, check box, drop-down selection box, date selector, etc., which is not limited in this embodiment of the present specification.

[0036] S404 , in response to an automatic filling instruction for a form to be filled in, receiving source data input through a form filling interface, and acquiring form information of the form to be filled in.

[0037] The automatic filling instruction for the form to be filled in can be triggered in various appropriate ways. In one implementation, in response to a first triggering operation on the form filling interface, it is determined that the automatic filling instruction for the form to be filled in is received.

[0038] In one example, the form filling interface includes a second control, and the first trigger operation may include a trigger operation on the second control.

[0039] For example, the table Figure 5A The form filling interface 50 shown also includes a second control 52 named "Smart Form Filling". The user only needs to click the second control 52 to trigger the automatic filling instruction of the form to be filled 51, which is convenient to operate.

[0040] In another implementation, the auto-fill instruction can also be triggered by voice. For example, after receiving the user's voice input and converting the voice into text, if the client detects that the text contains preset keywords such as "auto-fill form" or "auto-fill", it determines that the auto-fill instruction to fill in the form has been received.

[0041] In response to an automatic fill instruction for a form to be filled, the client may display a first control on the form filling interface. The first control is used to input source data required to fill the form to be filled. In one implementation, there are multiple first controls, each of which is used to input different types of source data. The different types of source data may include, but are not limited to, voice, text, and images. In other words, the source data input through the form filling interface may include, but are not limited to, at least one of the following types of data: voice, text, and images.

[0042] For example, Figure 5B As shown, in response to the user triggering an auto-fill instruction via second control 52, the client displays multiple first controls 53 on form filling interface 50, namely, a first control for inputting text or voice, a first control for uploading an image, and a first control for uploading a file. In this scenario, the user can enter various types of source data, such as uploading an image of a supplier's business license as source data, to automatically fill in the form to be filled 51.

[0043] Form information refers to the information used to describe the form to be filled out, and can include form structure information. Form structure information, also known as the form schema, is a standardized definition of the form structure and data rules. It describes the fields to be filled out, the data types supported by each field, the format and constraints of each field, and other information to ensure the accuracy and consistency of form data.

[0044] If the form to be filled in is not a form filled in for the first time, the form information may further include historical content data of the form items, that is, content data previously filled in the form items.

[0045] In another embodiment, after receiving the source data input through the form filling interface, the client starts the automatic filling function and displays the second prompt information on the form filling interface, where the second prompt information is used to prompt the filling status of the form to be filled.

[0046] For example, Figure 5C As shown, after the user enters source data through the first control 53, the form filling interface 50 also displays a prompt box 54, which includes a second prompt message "Filling the form." In addition, the prompt box 54 may also include a stop button 541, which is used to stop the automatic filling function of the form to be filled.

[0047] S406 , based on the first content data recommended by the artificial intelligence AI component for the form item, obtain second content data for filling in the form item, and fill the second content data into the form item.

[0048] The first content data is obtained by the AI ​​component from the source data based on the form information. The AI ​​component can obtain the first content data recommended for the form item in various appropriate ways, which are not limited in this embodiment of the specification.

[0049] In one implementation, the AI ​​component may send a form filling request to the server, which carries the form information and source data of the form to be filled out. The form filling request requests the server to utilize the large model to obtain content data from the source data for filling in the form items based on the form information. The AI ​​component uses the content data returned by the server as the first content data recommended for the form item.

[0050] In another implementation, when the hardware resources of the client's terminal device meet the requirements for running the large model, the large model can also be deployed on the terminal device. In this case, the AI ​​component can use the large model deployed on the terminal device to parse the source data based on the form information to obtain the first content data recommended for the form item.

[0051] After obtaining the first content data recommended for the form item, the second content data for filling in the form item can be obtained through various appropriate methods, which are not limited in the embodiments of this specification.

[0052] In one implementation, the first content data recommended by the AI ​​component for the form item can be used as the second content data for filling in the form item, and the second content data can be filled in the form item, thereby achieving automatic filling of the form to be filled in.

[0053] In another implementation, the first content data recommended by the AI ​​component for the form item is obtained; if the first content data of the form item meets the preset correction condition, the first content data of the form item is corrected based on the preset correction strategy to obtain the second content data for filling in the form item.

[0054] Among them, the preset correction conditions can be set according to actual needs, and the embodiments of this specification do not limit this. For example, considering that the AI ​​component may not be able to identify the first content data of the form item from some source data due to recognition accuracy problems, or the source data does not contain data related to some form items, these will cause the first content data to be empty. In addition, the type of the first content data identified by the AI ​​component for some form items may not match the type supported by the form item, and so on. These will affect the final filling effect of the form to be filled in. To this end, the preset correction conditions can include the first content data being empty, or the data type of the first content data does not belong to the data type supported by the form item in the form structure information, etc.

[0055] Specifically, if the first content data is empty, a first prompt message is displayed on the form filling interface, and the first prompt message is used to prompt the user to fill in the form item. In response to the filling instruction for the form item, the user receives the second content data for filling in the form item input through the form filling interface.

[0056] In addition, in some other implementations, after receiving the second content data input through the form filling interface, the first prompt information can be canceled on the form filling interface.

[0057] The filling instruction can be triggered by various appropriate methods, such as clicking a form item, clicking the first prompt information, etc., and the embodiments of this specification do not limit this.

[0058] For example, Figure 5D As shown in FIG, the first content data recommended by the AI ​​component for the form item “Legal Representative” is empty, and the first prompt information “manual selection required” 55 is displayed on the right side of the form item. Figure 5E As shown, in response to a click on the form item, a third control 56 is displayed. Thus, the user can use the third control 56 to select a legal representative from the list of personnel, such as selecting Person 1 as the second content data in the form item "Legal Representative." Furthermore, the first prompt message 55 is removed from the form filling interface 50.

[0059] It can be seen that this implementation method combines the automatic filling of AI components and the manual filling of users to provide a more flexible and reliable solution for form filling. When the AI ​​component cannot obtain the content data of certain form items based on the source data, it can also be manually supplemented by the user, thereby improving the filling efficiency while ensuring the accuracy and completeness of the filled content.

[0060] If the data type of the first content data does not belong to the data type supported by the form item in the form structure information, the first content data is type-converted to obtain second content data for filling in the form item.

[0061] For example, assuming that the quantity type of the first content data of the form item "Uniform Social Credit Code" is a number, and the data type supported by the form item in the form information is a string, the first content data is converted from a number to a string to obtain the second content data of the form item.

[0062] In actual application scenarios, some form items may be stored using data types that do not conform to the form item's design specifications due to differences in data sources or processing flows. By performing type conversion on such content data, the above implementation eliminates data type mismatches, ensuring that the resulting content data can be successfully entered into the form item, avoiding data loss or recognition issues caused by type conflicts.

[0063] In another embodiment, after S406, the following steps may be further included: in response to a second trigger operation on the form filling interface, generating a target form based on the form information and the second content data; and in response to a third trigger operation on the form filling interface, discarding the second content data filled in the form item. This allows users to independently decide whether to accept the form data filled in by the AI ​​component, making form filling more flexible and helping to ensure that the data in the final generated target form is accurate, providing reliable data support for subsequent business processing.

[0064] In some examples, the form filling interface further includes a fourth control and a fifth control, the fourth control being used to indicate acceptance of the second content data, and the fifth control being used to indicate rejection of the second content data. The second triggering operation may include a triggering operation on the fourth control, and the third operation may include a triggering operation on the fifth control.

[0065] For example, Figure 5D As shown, after the second content data of each form item is filled into the corresponding form item, a prompt box 57 is also displayed on the form filling interface 50. The prompt box 57 includes a third prompt message "AI has filled in the form for you, please confirm", an acceptance button 571 (i.e., the fourth control) and a rejection button 572 (i.e., the fifth control).

[0066] The user can click the adoption button 571 to adopt the second content data of each form item. In response to the triggering operation of the adoption button 571, the client generates a supplier information table based on the form information of the form 51 to be filled in and the second content data of each form item 511 for subsequent business use.

[0067] The user can also click the reject button 572 to reject the second content data of each form item. In response to the triggering operation of the reject button 572, the client clears the second content data of each form item of the form 51 to be filled in. In this way, the form filling interface 50 returns to the default state. Figure 5A The status shown.

[0068] The form filling methods provided in one or more of the above embodiments involve interaction between AI components and business components such as forms. If these components are interdependent, this can lead to problems such as excessive coupling, difficult maintenance, and difficulty in scalability. To mitigate these issues, an AI interface layer is designed between the AI ​​components and the forms.

[0069] In this case, in S404, obtaining the form information of the form to be filled in includes: calling the interface of the form to be filled in through the AI ​​interface layer that the AI ​​component relies on to query the form information of the form to be filled in; and receiving the form information of the form to be filled in returned by the interface of the form to be filled in through the AI ​​interface layer. Furthermore, in S406, filling the form item with the second content data includes: calling the interface of the form to be filled in through the AI ​​interface layer to fill the form item with the second content data.

[0070] Among them, the form to be filled in inherits from the pre-defined form class, and the AI ​​interface layer is implemented based on the form class.

[0071] The form filling method provided in the embodiment of this specification is that the client displays a form filling interface, responds to the automatic filling instruction of the form to be filled in the form filling interface, receives the source data input through the form filling interface, and obtains form information such as the form structure information of the form to be filled; further considering that AI technology has excellent effects in the semantic understanding and information extraction of natural language, and is good at identifying content in different formats, therefore, an AI component is configured on the client, and through the AI ​​component, the content data of the form items of the form to be filled in is obtained from the source data based on the form information, and the form items are filled in accordingly. In this way, AI technology is combined with forms, and the semantic understanding and information extraction capabilities of AI technology for natural language are used to fill in forms, thereby solving the problems of low input efficiency, high error rate and poor experience in traditional form filling methods. Please refer to Figure 6 , is a flowchart of a form filling method provided in another embodiment of this specification, the method is applied to the server 20, and the method may include the following steps: S602: Receive a form filling request sent by the AI ​​component.

[0072] A form fill request carries the form information and source data for the form to be filled out. The form to be filled out includes the form items to be filled out. The form information includes form structure information. The form fill request is used to request the content data to be filled in the form items based on the form information and source data.

[0073] S604: Parse the source data based on the form information using the large model to obtain first content data of the form item.

[0074] The large model has powerful capabilities for identifying diverse information and understanding semantics. It combines form information to identify and semantically understand the source data and obtain the first content data of the form item.

[0075] In one implementation, the server can deploy multiple large models, each suitable for processing different types of data. In this case, S604 may include the following steps: based on the type of source data, determining a target large model from the deployed large models that is suitable for the source data; and parsing the source data using the target large model to obtain the first content data of the form item. This allows the specialized capabilities of the large models to be fully utilized, more accurately extracting and parsing the source data, obtaining high-quality first content data, and thus improving form filling accuracy.

[0076] For example, deployed large models include, but are not limited to, large language models and multimodal models. If the source data type is text, the large language model is determined as the target large model. If the source data type is image, the multimodal model is determined as the target large model. If the source data type is speech, streaming speech recognition is performed on the source data to obtain text, and the language model is determined as the target large model.

[0077] In another implementation, the above S604 includes the following steps: obtaining reference form structure information of the form to be filled in; performing consistency verification on the reference form structure information and the form structure information in the form information; if the verification passes, parsing the source data based on the form information through the big model to obtain the first content data of the form item.

[0078] Reference form structure information refers to the standard structure associated with the form to be filled out. A consistency check is performed between the reference form structure information and the form structure information in the form information. This can include consistency checks on form item settings, data types, format requirements, and other information. If the check passes, subsequent source data parsing and processing proceeds. This ensures the standardization and accuracy of the form filling process and avoids data processing errors caused by inconsistent form structures.

[0079] S606: Return the first content data to the AI ​​component.

[0080] The first content data is used to fill in the form items.

[0081] In the form filling method provided in the embodiments of this specification, the client's AI component sends the acquired source data and form information of the form to be filled out to the server. The server uses a large model to identify and parse the source data based on the form information, obtains the first content data to be filled in the form item of the form to be filled out, and returns it to the AI ​​component. The AI ​​component automatically fills out the form to be filled out based on the first content data. In this way, AI technology is combined with forms, and its ability to understand the semantics of natural language and extract information is utilized to fill out forms, thus solving the problems of low input efficiency, high error rate, and poor user experience in traditional form filling methods.

[0082] In order to facilitate the understanding of the above form filling method, the following Figure 7 and Figure 8 Provide explanation.

[0083] like Figure 7 As shown, the server includes a first server and a second server, and the second server is deployed with multiple large models.

[0084] The client displays a form filling interface, which includes a form to be filled in and a second control, wherein the form to be filled in includes form items to be filled in. The user can click the second control to trigger an automatic filling instruction for the form to be filled in.

[0085] In response to the automatic filling instruction, the client displays a first control on the form filling interface. The user can input various types of source data, such as text, voice, and pictures, through the first control for the client to automatically fill in the form to be filled.

[0086] The client obtains form information such as form structure information of the form to be filled in. If the form to be filled in is not the first time to be filled in, the client also obtains historical content data of the form items. The client transmits the source data, form information, and historical content data of the form items to the first server.

[0087] In addition, if the source data is of voice type, the client also calls the voice service to perform streaming voice recognition on the source data before passing the source data to the first server. The voice service can be encapsulated by a streaming voice recognition model or a streaming voice recognition tool. Specifically, Figure 8 As shown, the client sends a request to the server for a JWT token. The server responds to the request and returns the JWT token. Subsequently, upon detecting that the user has initiated reading or voice input, the client establishes a webSocket connection with the voice service and passes the JWT token to the voice service. The voice service returns a successful connection response. Furthermore, the client and the voice service exchange data for an extended period of time. That is, the user continuously inputs voice, the client continuously transmits the user's input to the voice service, and the voice service returns the recognized text in real time to the client. In response to the user ending voice input, the client initiates a disconnect request to the voice service. The voice service returns the last packet of data (i.e., the last recognized text) to the client and disconnects from the client.

[0088] Please continue to see Figure 7 The first server converts the data transmitted by the client into a format that can be processed by the second server and sends it to the second server.

[0089] The second server performs a consistency check between the form structure information in the form information and the reference form structure information of the form to be filled out. If the check passes, it assembles a prompt based on the source data and the form information, indicating the need to provide content data for the form item. Based on the type of the source data, the second server also determines a target big model from among the deployed big models that is appropriate for the source data and sends a request to the target big model, including the prompt. The big model then parses the source data based on the prompt to obtain the first content data for the form item.

[0090] The second server returns the first content data to the first server in a transparent response manner. The first server sends the first content data to the client.

[0091] If the data type of the first content data of a certain form item(s) does not belong to the data type supported by the form item in the form structure information, the first content data is converted to obtain the second content data to be filled in the form item(s). If the first content data of a certain form item(s) is empty, the client processes the form item(s) as a special component and modifies the form state, including: displaying a first prompt message on the form item filling interface to prompt the user to fill in the form item; displaying a third control corresponding to the form item in response to a fill instruction for the form item; receiving the second content data for filling in the form item input through the third control; and canceling the display of the first prompt message on the form filling interface. Furthermore, the second content data is filled in as draft data in the form item and the completed form to be filled in (i.e., a pre-filled form) is displayed.

[0092] The user can make a second modification to the second content data entered into the form item. The client updates the draft data based on the user's second modification operation. The client also displays a fourth control and a fifth control on the form filling interface. The fourth control is used to indicate whether to accept the current draft data, and the fifth control is used to indicate whether to reject the current draft data.

[0093] In response to the triggering operation on the fourth control, the client uses the draft data (ie, the content data currently filled in the form item) as the formal content data, and generates a target form accordingly for subsequent business processing.

[0094] In response to the triggering operation on the fifth control, the client discards the draft data (ie, the content data currently filled in the form item).

[0095] In addition, with the above Figure 4 Corresponding to the form filling method shown, the embodiment of this specification also provides a form filling device. Figure 91 is a structural diagram of a form filling device 900 provided in an embodiment of this specification. The device 900 is applied to a client 10 and includes: a first display module 910 , a first acquisition module 920 and a filling module 930 .

[0096] The first display module 910 is used to display a form filling interface, where the form filling interface includes a form to be filled in, and the form to be filled in includes form items to be filled in.

[0097] The first acquisition module 920 is configured to receive source data input through the form filling interface in response to an automatic filling instruction for the form to be filled in, and to acquire form information of the form to be filled in, wherein the form information includes form structure information.

[0098] The filling module 930 is used to obtain second content data for filling in the form item based on the first content data recommended by the artificial intelligence AI component for the form item, and fill the second content data into the form item, wherein the first content data is obtained by the AI ​​component from the source data based on the form information.

[0099] The form filling device provided in the embodiment of this specification displays a form filling interface on the client side, responds to the automatic filling instruction of the form to be filled in the form filling interface, receives the source data input through the form filling interface, and obtains form information such as the form structure information of the form to be filled in; further considering that AI technology has excellent effects in the semantic understanding and information extraction of natural language, and is good at identifying content in different formats, therefore, an AI component is configured on the client side, and through the AI ​​component, the content data of the form items of the form to be filled in is obtained from the source data based on the form information, and the form items are filled in accordingly. In this way, AI technology is combined with forms, and the semantic understanding and information extraction capabilities of AI technology for natural language are used to fill in forms, thereby solving the problems of low input efficiency, high error rate and poor experience in traditional form filling methods. In another embodiment, the first acquisition module 920 is further configured to determine, in response to a first triggering operation on the form filling interface, that an automatic filling instruction for the form to be filled in is received.

[0100] In another embodiment, the filling module 930 includes: A first acquisition submodule, configured to acquire first content data recommended by the AI ​​component for the form item; The correction submodule is configured to correct the first content data of the form item based on a preset correction strategy if the first content data of the form item meets a preset correction condition, so as to obtain second content data for filling in the form item.

[0101] In another embodiment, the preset modification condition includes that the first content data is empty; The correction submodule is used to: If the first content data is empty, displaying first prompt information on the form filling interface, the first prompt information is used to prompt the user to fill in the form item; In response to a filling instruction for the form item, second content data for filling the form item input through the form filling interface is received.

[0102] In another embodiment, the preset modification condition includes that the data type of the first content data does not belong to the data type supported by the form item in the form information; The correction submodule is used to: If the data type of the first content data does not belong to the data type supported by the form item in the form structure information, type conversion is performed on the first content data to obtain second content data for filling in the form item.

[0103] In another embodiment, the first acquisition module 920 includes: A first calling submodule is configured to call the interface of the form to be filled in through the AI ​​interface layer on which the AI ​​component relies, so as to query form information of the form to be filled in; A first receiving submodule is configured to receive, through the AI ​​interface layer, form information of the form to be filled in returned by the interface of the form to be filled in; The filling module 930 includes: The second calling submodule is used to call the interface of the form to be filled in through the AI ​​interface layer, and fill the second content data into the form item.

[0104] In another embodiment, the form to be filled in is inherited from a predefined form class, and the AI ​​interface layer is implemented based on the form class.

[0105] In another embodiment, the form filling device further includes: a generating module, configured to generate a target form based on the form information and the second content data in response to a second triggering operation on the form filling interface; The discarding module is configured to discard the second content data filled in the form item in response to a third triggering operation on the form filling interface.

[0106] In another embodiment, the source data includes at least one type of data: voice, text, and picture.

[0107] In another embodiment, if the form to be filled in is not a form to be filled in for the first time, the form information further includes historical content data of the form items.

[0108] Obviously, the form filling device in the embodiment of this specification can be used as the above Figure 4 The execution body of the form filling method shown in Figure 4 Since the principle is the same, the functions realized are not described here.

[0109] In addition, with the above Figure 6 Corresponding to the form filling method shown, the embodiment of this specification also provides a form filling device. Figure 10 1 is a structural diagram of a form filling device 1000 provided in an embodiment of this specification. The device 1000 is applied to the server 20 and includes: a first receiving module 1010 , a parsing module 1020 and a first sending module 1030 .

[0110] The first receiving module 1010 is used to receive a form filling request sent by the AI ​​component, wherein the form filling request carries form information and source data of a form to be filled in, wherein the form to be filled in includes form items to be filled in, and the form information includes form structure information.

[0111] The parsing module 1020 is configured to parse the source data based on the form information using a large model to obtain the first content data of the form item.

[0112] The first sending module 1030 is used to return the first content data to the AI ​​component, where the first content data is used to fill in the form item.

[0113] In the form-filling device provided in the embodiments of this specification, the client's AI component sends the acquired source data and form information of the form to be filled out to the server. The server uses a large model to identify and parse the source data based on the form information, obtains the first content data to be filled in the form item of the form to be filled out, and returns it to the AI ​​component. The AI ​​component then automatically fills out the form to be filled out based on the first content data. Thus, AI technology is combined with forms, and its ability to understand the semantics of natural language and extract information is utilized to fill out forms, thereby solving the problems of low input efficiency, high error rate, and poor user experience that exist in traditional form-filling methods.

[0114] In another embodiment, the parsing module 1020 includes: A first determining submodule is configured to determine, based on the type of the source data, a target large model suitable for the source data from among the deployed large models; The first parsing submodule is used to parse the source data through the target large model to obtain the first content data of the form item.

[0115] In another embodiment, the parsing module 1020 includes: The second acquisition submodule is used to obtain the reference form structure information of the form to be filled in; A verification submodule, configured to perform consistency verification on the reference form structure information and the form structure information in the form information; The second parsing submodule is configured to parse the source data based on the form information through a large model to obtain first content data of the form item if the verification passes.

[0116] Obviously, the form filling device in the embodiment of this specification can be used as the above Figure 6 The execution body of the form filling method shown in Figure 6 Since the principle is the same, the functions realized are not described here.

[0117] Figure 11 This is a schematic diagram of the structure of an electronic device provided by an embodiment of this specification. Figure 11 At the hardware level, the electronic device includes a processor and, optionally, an internal bus, a network interface, and memory. The memory may include internal memory, such as high-speed random-access memory (RAM), or non-volatile memory, such as at least one disk drive. Of course, the electronic device may also include other hardware required for its services.

[0118] The processor, network interface, and memory can be interconnected via an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, Figure 11 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0119] The memory is used to store programs. Specifically, the program may include program code, which includes computer operating instructions. The memory may include internal memory and non-volatile memory, and provides instructions and data to the processor.

[0120] The processor reads the corresponding computer program from the non-volatile memory into the internal memory and then runs it, forming a form filling device at the logical level. The processor executes the program stored in the memory and is specifically used to perform the following operations: Displaying a form filling interface, wherein the form filling interface includes a form to be filled in, and the form to be filled in includes form items to be filled in; In response to an automatic filling instruction for the form to be filled in, receiving source data input through the form filling interface, and acquiring form information of the form to be filled in, the form information including form structure information; Based on the first content data recommended by the artificial intelligence AI component for the form item, second content data for filling in the form item is obtained, and the second content data is filled into the form item, wherein the first content data is obtained by the AI ​​component from the source data based on the form information.

[0121] Alternatively, the processor executes the program stored in the memory and is specifically configured to perform the following operations: Receive a form filling request sent by the AI ​​component, wherein the form filling request carries form information and source data of a form to be filled in, wherein the form to be filled in includes form items to be filled in, and the form information includes form structure information; Parsing the source data based on the form information using a large model to obtain first content data of the form item; The first content data is returned to the AI ​​component, where the first content data is used to fill in the form item.

[0122] The above is as in this manual Figure 4 The method performed by the form filling device disclosed in the embodiment shown, or the method described in the specification Figure 6The methods performed by the form filling device disclosed in the illustrated embodiments can be applied to or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be performed by hardware integrated logic circuits within the processor or by software instructions. The processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The methods, steps, and logic block diagrams disclosed in the embodiments of this specification can be implemented or executed. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in conjunction with the embodiments of this specification can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules within the decoding processor. The software module can be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above method.

[0123] It should be understood that the electronic device of the embodiment of this specification can realize the form filling device in Figure 4 The functions of the embodiment shown in the figure, or the electronic device of the embodiment of this specification can realize the form filling device in Figure 6 Since the principles are the same, the embodiments of this specification will not be described in detail here.

[0124] Of course, in addition to software implementation, the electronic device in this specification does not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc. That is to say, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0125] The embodiment of this specification also proposes a computer-readable storage medium, which stores one or more programs, and the one or more programs include instructions. When the instructions are executed by an electronic device including multiple application programs, the electronic device can execute Figure 4 The method of the embodiment shown is specifically used to perform the following operations: Displaying a form filling interface, wherein the form filling interface includes a form to be filled in, and the form to be filled in includes form items to be filled in; In response to an automatic filling instruction for the form to be filled in, receiving source data input through the form filling interface, and acquiring form information of the form to be filled in, the form information including form structure information; Based on the first content data recommended by the artificial intelligence AI component for the form item, second content data for filling in the form item is obtained, and the second content data is filled into the form item, wherein the first content data is obtained by the AI ​​component from the source data based on the form information.

[0126] When the instruction is executed by an electronic device including multiple applications, the electronic device can execute Figure 6 The method of the embodiment shown is specifically used to perform the following operations: Receive a form filling request sent by the AI ​​component, wherein the form filling request carries form information and source data of a form to be filled in, wherein the form to be filled in includes form items to be filled in, and the form information includes form structure information; Parsing the source data based on the form information using a large model to obtain first content data of the form item; The first content data is returned to the AI ​​component, where the first content data is used to fill in the form item.

[0127] An embodiment of this specification also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps in the form filling method provided in the embodiment of this specification.

[0128] The foregoing description of this specification describes specific embodiments. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0129] In short, the above description is only a preferred embodiment of this specification and is not intended to limit the scope of protection of this specification. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of this specification shall be included in the scope of protection of this specification.

[0130] The systems, devices, modules, or units described in the above embodiments may be implemented by computer chips or entities, or by products having certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0131] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0132] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0133] The various embodiments in this specification are described in a progressive manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiments are generally similar to the method embodiments, so the description is relatively simple. For relevant parts, refer to the description of the method embodiments.

Claims

1. A form filling method, characterized in that: include: Displaying a form filling interface, wherein the form filling interface includes a form to be filled in, and the form to be filled in includes form items to be filled in; In response to an automatic filling instruction for the form to be filled in, receiving source data input through the form filling interface, and acquiring form information of the form to be filled in, the form information including form structure information; Based on the first content data recommended by the artificial intelligence AI component for the form item, second content data for filling in the form item is obtained, and the second content data is filled into the form item, wherein the first content data is obtained by the AI ​​component from the source data based on the form information.

2. The method according to claim 1, characterized in that Before receiving source data input through the form filling interface in response to the automatic filling instruction for the form to be filled in, the method further includes: In response to a first triggering operation on the form filling interface, it is determined that an automatic filling instruction for the form to be filled in is received.

3. The method according to claim 1, characterized in that The obtaining of the second content data for filling in the form item based on the first content data recommended by the artificial intelligence (AI) component for the form item includes: Obtaining first content data recommended by the AI ​​component for the form item; If the first content data of the form item meets the preset correction condition, the first content data of the form item is corrected based on a preset correction strategy to obtain second content data for filling in the form item.

4. The method according to claim 3, characterized in that The preset correction condition includes that the first content data is empty; If the first content data of the form item meets a preset correction condition, the first content data of the form item is corrected based on a preset correction strategy to obtain second content data for filling in the form item, including: If the first content data is empty, displaying first prompt information on the form filling interface, the first prompt information is used to prompt the user to fill in the form item; In response to a filling instruction for the form item, second content data input through the form filling interface for filling in the form item is received.

5. The method according to claim 3, characterized in that The preset correction condition includes that the data type of the first content data does not belong to the data type supported by the form item in the form information; If the first content data of the form item meets a preset correction condition, the first content data of the form item is corrected based on a preset correction strategy to obtain second content data for filling in the form item, including: If the data type of the first content data does not belong to the data type supported by the form item in the form structure information, type conversion is performed on the first content data to obtain second content data for filling in the form item.

6. The method according to claim 1, wherein The obtaining of form information of the form to be filled in includes: Calling the interface of the form to be filled in through the AI ​​interface layer on which the AI ​​component relies to query form information of the form to be filled in; Receiving, through the AI ​​interface layer, form information of the form to be filled in returned by the interface of the form to be filled in; Filling the second content data into the form item includes: The interface of the form to be filled in is called through the AI ​​interface layer, and the second content data is filled into the form item.

7. The method according to claim 6, characterized in that The form to be filled in is inherited from a predefined form class, and the AI ​​interface layer is implemented based on the form class.

8. The method according to claim 1, characterized in that After the second content data is filled into the form item, the method further includes: In response to a second trigger operation on the form filling interface, generating a target form based on the form information and the second content data; In response to a third triggering operation on the form filling interface, the second content data filled in the form item is discarded.

9. The method according to claim 1, characterized in that The source data includes at least one type of data: voice, text, and picture.

10. The method according to claim 1, characterized in that If the form to be filled in is not a form filled in for the first time, the form information further includes historical content data of the form items.

11. A form filling method, characterized in that: include: Receive a form filling request sent by the AI ​​component, wherein the form filling request carries form information and source data of a form to be filled in, wherein the form to be filled in includes form items to be filled in, and the form information includes form structure information; Parsing the source data based on the form information using a large model to obtain first content data of the form item; The first content data is returned to the AI ​​component, where the first content data is used to fill in the form item.

12. The method according to claim 11, characterized in that The step of parsing the source data based on the form information using the large model to obtain the first content data of the form item includes: Determining, based on the type of the source data, a target big model suitable for the source data from deployed big models; The source data is parsed through the target large model to obtain the first content data of the form item.

13. The method according to claim 11, characterized in that The step of parsing the source data based on the form information using the large model to obtain the first content data of the form item includes: Obtaining reference form structure information of the form to be filled out; Performing consistency check on the reference form structure information and the form structure information in the form information; If the verification passes, the source data is parsed based on the form information through the large model to obtain the first content data of the form item.

14. A form filling device, characterized in that: include: A first display module is used to display a form filling interface, wherein the form filling interface includes a form to be filled in, and the form to be filled in includes form items to be filled in; a first acquisition module, configured to receive source data input through the form filling interface in response to an automatic filling instruction for the form to be filled in, and acquire form information of the form to be filled in, the form information including form structure information; A filling module is used to obtain second content data for filling in the form item based on the first content data recommended by the artificial intelligence AI component for the form item, and fill the second content data into the form item, wherein the first content data is obtained by the AI ​​component from the source data based on the form information.

15. A form filling device, characterized in that: include: A first receiving module is configured to receive a form filling request sent by an AI component, wherein the form filling request carries form information and source data of a form to be filled in, wherein the form to be filled in includes form items to be filled in, and the form information includes form structure information; a parsing module, configured to parse the source data based on the form information using a large model to obtain first content data of the form item; The first sending module is used to return the first content data to the AI ​​component, where the first content data is used to fill in the form item.

16. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the form filling method according to any one of claims 1 to 10, or to implement the form filling method according to any one of claims 11 to 13.

17. A computer-readable storage medium, characterized in that When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the form filling method as described in any one of claims 1 to 10, or execute the form filling method as described in any one of claims 11 to 13.

18. A computer program product, characterized in that The computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps in the form filling method described in any one of claims 1 to 10, or to execute some or all of the steps in the form filling method described in any one of claims 11 to 13.

Images