Questionnaire survey method and device based on large model, electronic equipment and storage medium

Abstract

The invention discloses a questionnaire survey method and device based on a large model, electronic equipment and a storage medium, and relates to the technical field of computers, in particular to the artificial intelligence fields of large models, natural language processing, deep learning and the like. According to the specific implementation scheme, first reply information for a current question in a target questionnaire is obtained; wherein the current question has an associated first word slot; performing information extraction on the first reply information according to the attribute information of the first word slot to obtain key information, and filling the first word slot according to the key information; responding to a second word slot which is not filled in the first word slot, generating a guiding problem according to attribute information of the second word slot, and displaying the guiding problem; acquiring second reply information of the guide question; and performing information extraction on the second reply information according to the attribute information of the second word slot so as to fill the second word slot.

Description

Technical Field

[0001] The present application relates to the field of computer technology, in particular to artificial intelligence fields such as large models, natural language processing, and deep learning, and specifically to a questionnaire survey method, device, electronic device, and storage medium based on a large model. Background Art

[0002] Questionnaire survey is a common data collection method. It obtains the results of opinion survey by collecting the answers of the respondents to the questions. Questionnaire survey has a wide range of applications in various fields, such as marketing, social science research, and medical fields. Summary of the invention

[0003] The present application provides a questionnaire survey method, device, electronic device and storage medium based on a large model.

[0004] According to one aspect of the present application, a questionnaire survey method based on a large model is provided, comprising:

[0005] Acquire first reply information for a current question in a target questionnaire; wherein the current question has an associated first word slot;

[0006] Extracting information from the first reply information according to the attribute information of the first word slot to obtain key information, and filling the first word slot according to the key information;

[0007] In response to the presence of an unfilled second word slot in the first word slot, generating a guiding question according to attribute information of the second word slot, and displaying the guiding question;

[0008] Obtaining second reply information of the guiding question;

[0009] According to the attribute information of the second word slot, information extraction is performed on the second reply information to fill the second word slot.

[0010] According to another aspect of the present application, a questionnaire survey device based on a large model is provided, comprising:

[0011] A first acquisition module is used to acquire first reply information for a current question in a target questionnaire; wherein the current question has an associated first word slot;

[0012] An information extraction and filling module, configured to extract information from the first reply information according to the attribute information of the first word slot to obtain key information; and fill the first word slot according to the key information;

[0013] a question generating module, configured to generate a guiding question according to attribute information of a second word slot in response to the presence of an unfilled second word slot in the first word slot;

[0014] A display module, used to display the guiding question;

[0015] A second acquisition module, used to acquire second reply information of the guiding question;

[0016] The information extraction and filling module is further used to extract information from the second reply information according to the attribute information of the second word slot to fill the second word slot.

[0017] According to another aspect of the present application, there is provided an electronic device, including:

[0018] at least one processor; and

[0019] a memory communicatively connected to the at least one processor; wherein,

[0020] The memory stores instructions that can be executed by the at least one processor. The instructions are executed by the at least one processor so that the at least one processor can perform the method described in the above embodiment.

[0021] According to another aspect of the present application, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the method according to the above embodiment.

[0022] According to another aspect of the present application, a computer program product is provided, including a computer program, wherein the computer program implements the steps of the method described in the above embodiment when executed by a processor.

[0023] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present application.

[0025] Figure 1 A flowchart of a questionnaire survey method based on a large model provided in one embodiment of the present application;

[0026] Figure 2 A flowchart of a questionnaire survey method based on a large model provided in another embodiment of the present application;

[0027] Figure 3A flowchart of a questionnaire survey method based on a large model provided in another embodiment of the present application;

[0028] Figure 4 A flowchart of a questionnaire survey method based on a large model provided in another embodiment of the present application;

[0029] Figure 5 A flowchart of a questionnaire survey provided in an embodiment of the present application;

[0030] Figure 6 A schematic diagram of the structure of a questionnaire survey device based on a large model provided in one embodiment of the present application;

[0031] Figure 7 It is a block diagram of an electronic device used to implement the large model-based questionnaire survey method of an embodiment of the present application. DETAILED DESCRIPTION

[0032] The following is a description of exemplary embodiments of the present application in conjunction with the accompanying drawings, including various details of the embodiments of the present application to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0033] It should be noted that the acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations and do not violate public order and good morals.

[0034] The following describes the large model-based questionnaire survey method, device, electronic device and storage medium of an embodiment of the present application with reference to the accompanying drawings.

[0035] In some embodiments, a set of fixed questions and options may be provided during a questionnaire survey, and the user may select responses from the preset options. However, this static questionnaire method has poor flexibility.

[0036] Based on this, an embodiment of the present application provides a questionnaire survey method based on a large model. Figure 1 A flowchart of a large model-based questionnaire survey method provided in one embodiment of the present application.

[0037] The large model-based questionnaire survey method of the embodiment of the present application can be executed by the large model-based questionnaire survey device of the embodiment of the present application, and the device can be configured in an electronic device.

[0038] Among them, the electronic device can be any device with computing capabilities, such as a personal computer, a mobile terminal, a server, etc. The mobile terminal can be, for example, a vehicle-mounted device, a mobile phone, a tablet computer, a personal digital assistant, a wearable device, and other hardware devices with various operating systems, touch screens and / or display screens.

[0039] like Figure 1 As shown, the questionnaire survey method based on the large model includes:

[0040] Step 101, obtaining the first response information for the current question in the target questionnaire.

[0041] In the present application, the target questionnaire may be any questionnaire, which may include at least one preset question, and a word slot may be set in advance for each preset question, that is, each preset question may be associated with one or more word slots.

[0042] In the present application, the current question in the target questionnaire can be displayed to the user, and the user can enter a reply to the current question, that is, the first reply information, thereby obtaining the first reply information of the current question. The current question has an associated word slot, that is, the first word slot. In addition, the number of the first word slots can be one or more, and there is no limitation on this.

[0043] Exemplarily, the user may input a reply to a question by voice or text. If the reply is in voice form, the reply may be voice recognized and converted into text.

[0044] Step 102: extract information from the first reply information according to the attribute information of the first word slot to obtain key information, and fill the first word slot according to the key information.

[0045] The attribute information of the word slot may include but is not limited to the name of the word slot, description information, answer type, answer value range, filling requirements, etc. The description information indicates the content that guides AI (Artificial Intelligence) to understand, and the filling requirements indicate whether the word slot must be filled.

[0046] For example, the attribute information of a word slot is as follows:

[0047] {

[0048] "slot":"Image report has been uploaded",

[0049] "description":"You need to determine what the uploaded report is. If the content of the uploaded report is "imaging report" (such as CT, B-ultrasound, etc.), please fill in the slot "Yes"; if the user clearly indicates that he cannot provide the report, fill in the slot "Unwilling to upload"; in other cases, there is no need to fill in this slot.",

[0050] "type":"enumeration type",

[0051] "answer_range":"[\"Yes\",\"Don't want to upload\"]",

[0052] "optional":true

[0053] }

[0054] For example, the attribute information of another word slot is as follows:

[0055] {

[0056] "slot":"Age",

[0057] "description":"Collect user's age information",

[0058] "type":"Open question and answer type",

[0059] "answer_range":"\"Age, such as 20, 35\"",

[0060] "optional":true

[0061] }

[0062] In the above example, "optional" indicates whether the word slot is optional. The attribute value of "optional" is "true" which means that the word slot is not required to be filled, and the attribute value of "optional" is "false" which means that the word slot must be filled. In the example, the attribute value of "optional" for the word slot named "age" is "true", which means that the user can choose not to provide age information.

[0063] As a possible implementation method, the first reply information can be segmented to obtain each segmented word contained in the first reply information, and the attribute information of the first word slot can be matched with each segmented word to determine the segmented word that matches the first word slot from the various segmented words, and based on the segmented word that matches the first word slot, key information that matches the first segmented word is obtained, and then the key information is used to fill the first word slot so that the word slot has slot information.

[0064] Step 103: In response to the existence of an unfilled second word slot in the first word slot, a guiding question is generated according to attribute information of the second word slot, and the guiding question is displayed.

[0065] In the present application, if there is an unfilled second word slot in the first word slot, in order to meet the needs of the questionnaire survey, a guiding question can be generated based on the attribute information of the second word slot and displayed to the user to conduct targeted follow-up questions to the user.

[0066] The second word slot may be part of or all of the word slots in the first word slot, and the second word slot may be one or more.

[0067] For example, the question is "How long do you use this product every day and how often do you use it?" The word slots associated with this question are "daily usage time" and "daily usage frequency." The user's response is "I use it about two hours a day." Based on the word slots associated with this question, the key information "two hours" that matches the word slot "daily usage time" is extracted from the question, so that the word slot "daily usage time" can be filled. If the word slot "daily usage frequency" is not filled, then based on this word slot, a leading question "How many times do you use this product every day" can be generated to continue asking the user.

[0068] For example, the question is "What is your name and how old are you?" The word slots associated with this question are "name" and "age." The user inputs a reply "Can you repeat it again?". No key information is extracted from this reply, and "name" and "age" are not filled. Then, based on these two word slots, the question can be re-polished to generate a leading question, such as "I want to ask what your name is and how old you are this year." In this way, new questions can be re-generated for indirect replies, effectively handling unexpected answers.

[0069] Step 104: Obtain second reply information to the guiding question.

[0070] In the present application, the reply input by the user to the guiding question may be collected, that is, the second reply information of the guiding question may be obtained.

[0071] Step 105: extract information from the second reply information according to the attribute information of the second word slot to fill the second word slot.

[0072] In the present application, the method for extracting information from the second reply information based on the attribute information of the second word slot can refer to the method for extracting information from the first reply information based on the attribute information of the first word slot, and will not be repeated here.

[0073] In the present application, the second word slot is filled with key information extracted from the second reply information. If there are still unfilled word slots in the second word slot, the leading question can be continuously generated based on the unfilled word slots in the second word slot, so as to extract key information from the reply information of the leading question and fill the unfilled word slots in the second word slot until the first word slots associated with the current question are all filled. If the second word slots are all filled, it means that the first word slots of the current question are all filled. Here, the first word slots are all filled, which means that the first word slots all have slot information.

[0074] Exemplarily, after the first word slot of the current question is filled, the next question of the current question in the target questionnaire can be displayed, and the above method can be continued to fill the word slot associated with the next question until all the word slots associated with the questions in the target questionnaire are filled, and the session ends.

[0075] In the embodiment of the present application, information extraction is performed on the reply information of the current question according to the attribute information of the word slot associated with the current question in the target questionnaire, and the word slot associated with the current question is filled with the extracted key information. If there is an unfilled word slot in the word slot associated with the current question, the unfilled word slot is used to generate a leading question, and information extraction is performed on the reply information of the leading question according to the attribute information of the unfilled word slot to fill the unfilled word slot. Thus, the natural language input by the user can be accurately parsed through the word slot associated with the question, and if there is an unfilled word slot, a leading question is generated, so as to dynamically adjust the subsequent question-and-answer process, thereby improving the flexibility of the questionnaire survey and the natural communication sense of dynamic question-and-answer, and generating leading questions based on the unfilled word slot and conducting targeted follow-up questions can improve the efficiency of the questionnaire.

[0076] Figure 2 A flowchart of a large model-based questionnaire survey method provided in another embodiment of the present application.

[0077] like Figure 2 As shown, the questionnaire survey method based on the large model includes:

[0078] Step 201, obtaining the first response information for the current question in the target questionnaire.

[0079] In the present application, step 201 can be implemented in any of the embodiments of the present application, so it will not be described in detail here.

[0080] Optionally, the attribute information of the current question may be obtained first. If the attribute information of the current question includes a skip condition and the skip condition is not met, the current question is displayed to the user to obtain the reply information entered by the user for the current question. If the skip condition is met, the current question may be skipped, and based on the attribute information of the next question of the current question, it is determined whether the next question meets the skip condition, thereby determining whether to display the next question of the current question.

[0081] For example, the skip condition for a question is as follows:

[0082]

[0083] The skip condition indicates that if the slot information of the word slot "image report sheet uploaded" is "yes", that is, the image report sheet has been uploaded, the question can be skipped.

[0084] Therefore, by setting skip conditions for questions, the number of interactions can be controlled, repeated information collection can be avoided, and the efficiency of the questionnaire survey can be improved.

[0085] Step 202: extract information from the first reply information according to the attribute information of the first word slot to obtain key information, and fill the first word slot according to the key information.

[0086] In the present application, step 202 can be implemented in any of the embodiments of the present application, so it will not be described in detail here.

[0087] Exemplarily, the information extraction task prompt template can be filled based on the attribute information of the first word slot to obtain the second prompt information, and the second prompt information can be input into the large model, so that the large model can extract information from the first reply information based on the attribute information of the first word slot to obtain key information, and then use the extracted key information to fill the first word slot.

[0088] The second prompt information can be used to prompt the large model to perform an information extraction task based on the attribute information of the first word slot.

[0089] Therefore, the large model uses the attribute information of the word slot associated with the current question to extract the reply information of the current question, which simplifies the task scope of the large model and allows the large model to focus on executing the recognition of the content of a single word slot, thereby improving the stability and response speed of the overall system.

[0090] Since the reply information of the current question may contain slot information of other word slots except the word slot associated with the current question, based on this, illustratively, the unfilled word slots in the word slot set corresponding to the target questionnaire can be determined, and the unfilled word slots in the word slot set can include the unfilled word slots in the first word slot, and according to the attribute information of the unfilled word slots in the word slot set, the first reply information is extracted to obtain key information, and the unfilled word slots in the word slot set are filled using the extracted key information.

[0091] Among them, the word slot set corresponding to the target questionnaire is the set of word slots associated with each question in the target questionnaire, that is, the word slot set is the set formed by the word slots associated with each question in the target questionnaire.

[0092] Therefore, by utilizing the currently unfilled word slots of the target questionnaire to extract information from the reply information of the current question, not only can the word slots associated with the current question be filled, but also other word slots can be filled when the reply information contains slot information of other word slots, thereby improving the efficiency of the questionnaire survey.

[0093] Step 203, in response to the existence of an unfilled second word slot in the first word slot, generating first prompt information according to attribute information of the second word slot, the current question and the first reply information.

[0094] In this application, if there is an unfilled second word slot in the first word slot, the question generation task prompt template can be filled according to the attribute information of the second word slot, the current question, the first reply information, etc., to obtain the first prompt information. The first prompt information can be used to prompt the large model to perform the question generation task.

[0095] The first prompt information may include but is not limited to the attribute information of the second word slot, the current question, the first reply information, the question generation task instruction statement, the model output requirements, etc. The question generation task instruction statement is a statement that indicates the task to be performed by the large model, such as the question generation task instruction statement is "Based on the above content, generate a question for the word slot "daily frequency of use"".

[0096] Step 204: input the first prompt information into the large model, use the large model to generate a guiding question for the second word slot, and display the guiding question.

[0097] In the present application, a large model can be used to generate a guiding question for the second word slot according to the first prompt information. It can be understood that the guiding question is associated with the second word slot.

[0098] Step 205: Obtain second reply information to the guiding question.

[0099] Step 206: extract information from the second reply information according to the attribute information of the second word slot to fill the second word slot.

[0100] In the present application, step 205-step 206 can be implemented in any of the embodiments of the present application, so they will not be described in detail here.

[0101] In the embodiment of the present application, based on the attribute information of the unfilled word slots in the currently associated word slots, the current question and the reply information of the current question, the first prompt information is constructed, the first prompt information is input into the large model, the large model is used to generate guiding questions for the unfilled word slots, and based on the reply information of the guiding questions, the unfilled word slots in the word slots associated with the current question are filled. Thus, by using the large model to generate guiding questions for the unfilled word slots, the accuracy of the guiding questions is improved.

[0102] Figure 3 A flowchart of a large model-based questionnaire survey method provided in another embodiment of the present application.

[0103] like Figure 3 As shown, the questionnaire survey method based on the large model includes:

[0104] Step 301, obtaining the first response information for the current question in the target questionnaire.

[0105] Step 302: extract information from the first reply information according to the attribute information of the first word slot to obtain key information, and fill the first word slot according to the key information.

[0106] In the present application, step 301-step 302 can be implemented in any of the embodiments of the present application, so they will not be described in detail here.

[0107] Step 303: In response to the existence of an unfilled second word slot in the first word slot and the slot information type of the first word slot being a picture type, determine a third word slot associated with the second word slot.

[0108] In the present application, the attribute information of the word slot may include the slot information type of the word slot, and the slot information type includes, for example, image type, text type, etc.

[0109] Exemplarily, if the slot information of a word slot is a picture type, the picture is identified to extract the required information from the picture, and the word slot corresponding to the information to be extracted from the picture is the word slot associated with the word slot whose slot information is a picture type.

[0110] For example, the slot information type of the word slot "laboratory report" is a picture type. By identifying the picture of "laboratory report", the indicator value of each test indicator can be obtained from the picture. Then each test indicator is a word slot associated with the word slot "laboratory report".

[0111] In the present application, if there is an unfilled second word slot in the first word slot, it can be determined whether the attribute information of the second word slot includes the slot information type of the second word slot. If the attribute information of the second word slot includes the slot information type of the second word slot, and the slot information type of the second word slot is a picture type, the third word slot associated with the second word slot can be determined.

[0112] Step 304: Generate a guiding question based on the attribute information of the third word slot, and display the guiding question.

[0113] In the present application, the question generation task prompt template can be filled in according to the attribute information of the third word slot to obtain corresponding prompt information, which is input into the large model, and the large model is used to generate guiding questions.

[0114] For example, the word slot is "laboratory report", when the user is unwilling or unable to provide a laboratory report, guiding questions can be generated based on the word slots associated with the word slot "laboratory report", that is, each laboratory indicator, so as to inquire about the indicator value of each laboratory indicator through guiding questions.

[0115] Step 305: Obtain second reply information to the guiding question.

[0116] Step 306: extract information from the second reply information according to the attribute information of the second word slot to fill the second word slot.

[0117] In the present application, steps 305 and 306 may be implemented in any of the embodiments of the present application, and therefore will not be described in detail here.

[0118] In an embodiment of the present application, the attribute information of the second word slot includes the slot information type of the second word slot. If the slot information type of the second word slot is a picture type, a guiding question can be generated based on the third word slot associated with the second word slot, thereby converting the slot information of the picture type into information indirectly obtained through the picture through direct inquiry, thereby improving the flexibility of data collection.

[0119] Figure 4 A flowchart of a large model-based questionnaire survey method provided in another embodiment of the present application.

[0120] like Figure 4 As shown, the questionnaire survey method based on the large model includes:

[0121] Step 401, obtaining the first response information for the current question in the target questionnaire.

[0122] Step 402: extract information from the first reply information according to the attribute information of the first word slot to obtain key information, and fill the first word slot according to the key information.

[0123] In the present application, step 401-step 402 can be implemented in any of the embodiments of the present application, so they will not be described in detail here.

[0124] Step 403 , in response to the existence of an unfilled second word slot in the first word slot and the filling requirement of the second word slot being mandatory, generating a guiding question according to the attribute information of the second word slot and displaying the guiding question.

[0125] For example, in the word slots associated with each question in the questionnaire, some word slots may have slot information that must be collected, while some word slots may have slot information that is optional. Based on this, if there is an unfilled second word slot in the first word slot, it can be determined whether the attribute information of the second word slot includes the filling requirement of the second word slot. If the attribute information of the second word slot includes the filling requirement of the second word slot, and the filling requirement is required, it means that the slot information of the second word slot is necessary data, and then a guiding question can be generated according to the attribute information of the second word slot.

[0126] Step 404: Obtain second reply information to the guiding question.

[0127] Step 405: extract information from the second reply information according to the attribute information of the second word slot to fill the second word slot.

[0128] In the present application, step 404-step 405 can be implemented in any of the embodiments of the present application, so they will not be described in detail here.

[0129] In the embodiment of the present application, if there is an unfilled second word slot in the first word slot, it is determined whether the attribute information of the second word slot includes a filling requirement for the second word slot, and if the filling requirement is included and the filling requirement is required, a guiding question is generated based on the attribute information of the second word slot. Thus, when the slot information of the second word slot is a required item, a guiding question is generated based on the attribute information of the second word slot, which can improve the efficiency of the questionnaire survey.

[0130] In one embodiment of the present application, the user's responses can also be extracted based on all word slots associated with each question in the questionnaire. After the word slots are filled based on the responses to each question in the questionnaire, if there are unfilled word slots, guiding questions can be generated based on the unfilled word slots to conduct targeted follow-up questions. Figure 5 To explain, Figure 5A flowchart of a questionnaire survey provided in an embodiment of the present application.

[0131] like Figure 5 As shown, the process includes:

[0132] Step 501, initializing the questionnaire.

[0133] In this embodiment, the questionnaire may be initialized and word slots may be allocated to each question in the questionnaire.

[0134] Step 502, display the question to the user.

[0135] Step 503: Receive a user response.

[0136] Step 504, parse the reply and fill the word slots.

[0137] In this embodiment, a large model can be used to parse the responses based on all word slots corresponding to the questionnaire, and fill the word slots based on the extracted key information.

[0138] Step 505: Check whether all necessary items have been collected.

[0139] The necessary items here can be understood as the word slots that are required to be filled in. In this embodiment, it can be determined whether all the necessary word slots of the questionnaire have been filled in with slot information.

[0140] Step 506: Generate a new guiding question.

[0141] If there are unfilled words, the large model is used to generate new guiding questions based on the unfilled word slots, and the questions are displayed to the user to receive user responses and continue to fill the filled word slots.

[0142] Step 507, end the session.

[0143] If all necessary word slots of the questionnaire are filled with slot information, the session ends.

[0144] The questionnaire survey method based on the large model in the embodiment of the present application can bring at least the following four beneficial effects:

[0145] Improve user experience: By allowing users to communicate in a more natural way, it reduces the barriers in the traditional questionnaire filling process;

[0146] Enhanced information collection capabilities: When faced with answers containing multiple pieces of information, they can effectively disassemble and extract them, ensuring that the required data is obtained comprehensively and accurately;

[0147] Reduce costs and maintenance difficulty: It combines the advantages of easy management of static questionnaires and flexible interaction of dynamic questions and answers, while reducing the need to design complex dialogue logic for different business scenarios;

[0148] Improved efficiency and reliability: The task scope of large language models is simplified to focus on performing simple but critical small tasks, namely identifying information fragments in a specified format, thereby improving the stability and responsiveness of the overall system.

[0149] In order to implement the above embodiment, the embodiment of the present application also proposes a questionnaire survey device based on a large model. Figure 6 A schematic diagram of the structure of a large model-based questionnaire survey device provided in one embodiment of the present application.

[0150] like Figure 6 As shown, the questionnaire survey device 600 based on the large model includes:

[0151] A first acquisition module 610 is used to acquire first reply information for a current question in a target questionnaire; wherein the current question has an associated first word slot;

[0152] The information extraction and filling module 620 is used to extract information from the first reply information according to the attribute information of the first word slot to obtain key information; and fill the first word slot according to the key information;

[0153] A question generating module 630, configured to generate a guiding question according to attribute information of a second word slot in response to the presence of an unfilled second word slot in the first word slot;

[0154] A display module 640 is used to display the guiding question;

[0155] A second acquisition module 650 is used to acquire second reply information of the guiding question;

[0156] The information extraction and filling module 620 is further used to extract information from the second reply information according to the attribute information of the second word slot to fill the second word slot.

[0157] Optionally, the question generating module 630 is used to:

[0158] Generate first prompt information according to the attribute information of the second word slot, the current question and the first reply information; wherein the first prompt information is used to prompt the large model to perform the question generation task;

[0159] The first prompt information is input into the large model, and the guiding question for the second word slot is generated using the large model.

[0160] Optionally, the attribute information of the second word slot includes a slot information type of the second word slot, and the question generating module 630 is configured to:

[0161] In response to the slot information type being a picture type, determining a third word slot associated with the second word slot;

[0162] The guiding question is generated according to the attribute information of the third word slot.

[0163] Optionally, the question generating module 630 is used to:

[0164] In response to the presence of an unfilled second word slot in the first word slot, determining whether the attribute information of the second word slot includes a filling requirement for the second word slot;

[0165] In response to the attribute information of the second word slot including a filling requirement for the second word slot and the filling requirement being mandatory, the guiding question is generated according to the attribute information of the second word slot.

[0166] Optionally, the information extraction and filling module 620 is used to:

[0167] Generate second prompt information according to the attribute information of the first word slot; wherein the second prompt information is used to prompt the large model to perform an information extraction task;

[0168] The second prompt information is input into the large model, and the large model is used to extract information from the first reply information to obtain the key information.

[0169] Optionally, the information extraction and filling module 620 is used to:

[0170] Determine unfilled word slots in the word slot set corresponding to the target questionnaire; wherein the word slot set is a set of word slots associated with each question in the target questionnaire, and the unfilled word slots in the word slot set include unfilled word slots in the first word slot;

[0171] Information extraction is performed on the first reply information according to the attribute information of the unfilled word slots in the word slot set to obtain the key information.

[0172] Optionally, the device may further include:

[0173] A third acquisition module is used to acquire attribute information of the current question;

[0174] The display module 640 is also used to display the current question in response to the attribute information of the current question including a skip condition and the skip condition is not met; in response to the skip condition being met, skip the current question, and determine whether to display the next question of the current question based on the attribute information of the next question of the current question.

[0175] It should be noted that the explanation of the aforementioned embodiment of the questionnaire survey method based on a large model is also applicable to the questionnaire survey device based on a large model of this embodiment, so it will not be repeated here.

[0176] In the embodiment of the present application, information extraction is performed on the reply information of the current question according to the attribute information of the word slot associated with the current question in the target questionnaire, and the word slot associated with the current question is filled with the extracted key information. If there is an unfilled word slot in the word slot associated with the current question, the unfilled word slot is used to generate a leading question, and information extraction is performed on the reply information of the leading question according to the attribute information of the unfilled word slot to fill the unfilled word slot. Thus, the natural language input by the user can be accurately parsed through the word slot associated with the question, and if there is an unfilled word slot, a leading question is generated, so as to dynamically adjust the subsequent question-and-answer process, thereby improving the flexibility of the questionnaire survey and the natural communication sense of dynamic question-and-answer, and generating leading questions based on the unfilled word slot and conducting targeted follow-up questions can improve the efficiency of the questionnaire.

[0177] According to an embodiment of the present application, the present application also provides an electronic device, a readable storage medium and a computer program product.

[0178] Figure 7 A schematic block diagram of an example electronic device 700 that can be used to implement an embodiment of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or required herein.

[0179] like Figure 7 As shown, the device 700 includes a computing unit 701, which can perform various appropriate actions and processes according to a computer program stored in a ROM (Read-Only Memory) 702 or a computer program loaded from a storage unit 708 to a RAM (Random Access Memory) 703. In the RAM 703, various programs and data required for the operation of the device 700 can also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An I / O (Input / Output) interface 705 is also connected to the bus 704.

[0180] A number of components in the device 700 are connected to the I / O interface 705, including: an input unit 706, such as a keyboard, a mouse, etc.; an output unit 707, such as various types of displays, speakers, etc.; a storage unit 708, such as a disk, an optical disk, etc.; and a communication unit 709, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 709 allows the device 700 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0181] The computing unit 701 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a CPU (Central Processing Unit), a GPU (Graphic Processing Units), various dedicated AI (Artificial Intelligence) computing chips, various computing units running machine learning model algorithms, a DSP (Digital Signal Processor), and any appropriate processor, controller, microcontroller, etc. The computing unit 701 performs the various methods and processes described above, such as a questionnaire survey method based on a large model. For example, in some embodiments, the questionnaire survey method based on a large model may be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 708. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 700 via the ROM 702 and / or the communication unit 709. When the computer program is loaded into the RAM 703 and executed by the computing unit 701, one or more steps of the questionnaire survey method based on the large model described above may be executed. Alternatively, in other embodiments, the computing unit 701 may be configured to execute the large model-based questionnaire survey method in any other appropriate manner (eg, by means of firmware).

[0182] Various embodiments of the systems and techniques described above herein may be implemented in digital electronic circuit systems, integrated circuit systems, FPGAs (Field Programmable Gate Arrays), ASICs (Application-Specific Integrated Circuits), ASSPs (Application Specific Standard Products), SOCs (System On Chips), CPLDs (Complex Programmable Logic Devices), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include: being implemented in one or more computer programs that may be executed and / or interpreted on a programmable system including at least one programmable processor that may be a dedicated or general-purpose programmable processor that may receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0183] The program code for implementing the method of the present application can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, implements the functions / operations specified in the flow chart and / or block diagram. The program code can be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0184] In the context of the present application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include electrical connections based on one or more lines, portable computer disks, hard disks, RAM, ROM, EPROM (Electrically Programmable Read-Only-Memory) or flash memory, optical fiber, CD-ROM (Compact Disc Read-Only Memory), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0185] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (Cathode-Ray Tube) or LCD (Liquid Crystal Display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0186] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: LAN (Local Area Network), WAN (Wide Area Network), the Internet, and blockchain networks.

[0187] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services (Virtual Private Server). The server may also be a server of a distributed system, or a server combined with a blockchain.

[0188] According to an embodiment of the present application, the present application also provides a computer program product, which, when an instruction processor in the computer program product executes, executes the large model-based questionnaire survey method proposed in the above embodiment of the present application.

[0189] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this application can be executed in parallel, sequentially or in different orders, as long as the expected results of the technical solution disclosed in this application can be achieved, and this document is not limited here.

[0190] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A questionnaire survey method based on a large model, characterized in that: include: Acquire first reply information for a current question in a target questionnaire; wherein the current question has an associated first word slot; Extracting information from the first reply information according to the attribute information of the first word slot to obtain key information, and filling the first word slot according to the key information; In response to the presence of an unfilled second word slot in the first word slot, generating a guiding question according to attribute information of the second word slot, and displaying the guiding question; Obtaining second reply information of the guiding question; According to the attribute information of the second word slot, information extraction is performed on the second reply information to fill the second word slot.

2. The method of claim 1, wherein: The step of generating a guiding question according to the attribute information of the second word slot includes: Generate first prompt information according to the attribute information of the second word slot, the current question and the first reply information; wherein the first prompt information is used to prompt the large model to perform the question generation task; The first prompt information is input into the large model, and the guiding question for the second word slot is generated using the large model.

3. The method of claim 1, wherein: The attribute information of the second word slot includes the slot information type of the second word slot, and generating a guiding question according to the attribute information of the second word slot includes: In response to the slot information type being a picture type, determining a third word slot associated with the second word slot; The guiding question is generated according to the attribute information of the third word slot.

4. The method of claim 1, wherein: In response to the presence of an unfilled second word slot in the first word slot, generating a guiding question according to attribute information of the second word slot includes: In response to the presence of an unfilled second word slot in the first word slot, determining whether the attribute information of the second word slot includes a filling requirement for the second word slot; In response to the attribute information of the second word slot including a filling requirement for the second word slot and the filling requirement being mandatory, the guiding question is generated according to the attribute information of the second word slot.

5. The method of claim 1, wherein: The extracting the first reply information according to the attribute information of the first word slot to obtain key information includes: Generate second prompt information according to the attribute information of the first word slot; wherein the second prompt information is used to prompt the large model to perform an information extraction task; The second prompt information is input into the large model, and the large model is used to extract information from the first reply information to obtain the key information.

6. The method of claim 1, wherein: The extracting the first reply information according to the attribute information of the first word slot to obtain key information includes: Determine unfilled word slots in the word slot set corresponding to the target questionnaire; wherein the word slot set is a set of word slots associated with each question in the target questionnaire, and the unfilled word slots in the word slot set include unfilled word slots in the first word slot; Information extraction is performed on the first reply information according to the attribute information of the unfilled word slots in the word slot set to obtain the key information.

7. The method according to any one of claims 1 to 6, further comprising: Obtaining attribute information of the current problem; In response to the attribute information of the current question including a skip condition and the skip condition is not satisfied, displaying the current question; In response to satisfying the skip condition, the current question is skipped, and it is determined whether to display the next question of the current question according to the attribute information of the next question of the current question.

8. A questionnaire survey device based on a large model, characterized in that: include: A first acquisition module is used to acquire first reply information for a current question in a target questionnaire; wherein the current question has an associated first word slot; An information extraction and filling module, configured to extract information from the first reply information according to the attribute information of the first word slot to obtain key information; and fill the first word slot according to the key information; a question generating module, configured to generate a guiding question according to attribute information of a second word slot in response to the presence of an unfilled second word slot in the first word slot; A display module, used to display the guiding question; A second acquisition module, used to acquire second reply information of the guiding question; The information extraction and filling module is further used to extract information from the second reply information according to the attribute information of the second word slot to fill the second word slot.

9. The device of claim 8, wherein: The question generating module is used for: Generate first prompt information according to the attribute information of the second word slot, the current question and the first reply information; wherein the first prompt information is used to prompt the large model to perform the question generation task; The first prompt information is input into the large model, and the guiding question for the second word slot is generated using the large model.

10. The device of claim 8, wherein: The attribute information of the second word slot includes the slot information type of the second word slot, and the question generating module is used to: In response to the slot information type being a picture type, determining a third word slot associated with the second word slot; The guiding question is generated according to the attribute information of the third word slot.

11. The device of claim 8, wherein: The question generating module is used for: In response to the presence of an unfilled second word slot in the first word slot, determining whether the attribute information of the second word slot includes a filling requirement for the second word slot; In response to the attribute information of the second word slot including a filling requirement for the second word slot and the filling requirement being mandatory, the guiding question is generated according to the attribute information of the second word slot.

12. The device of claim 8, wherein: The information extraction and filling module is used to: Generate second prompt information according to the attribute information of the first word slot; wherein the second prompt information is used to prompt the large model to perform an information extraction task; The second prompt information is input into the large model, and the large model is used to extract information from the first reply information to obtain the key information.

13. The device of claim 8, wherein: The information extraction and filling module is used to: Determine unfilled word slots in the word slot set corresponding to the target questionnaire; wherein the word slot set is a set of word slots associated with each question in the target questionnaire, and the unfilled word slots in the word slot set include unfilled word slots in the first word slot; Information extraction is performed on the first reply information according to the attribute information of the unfilled word slots in the word slot set to obtain the key information.

14. The apparatus according to any one of claims 8 to 13, further comprising: A third acquisition module is used to acquire attribute information of the current question; The display module is also used to display the current question in response to the attribute information of the current question including a skip condition and the skip condition is not met; in response to the skip condition being met, skip the current question, and determine whether to display the next question of the current question based on the attribute information of the next question of the current question.

15. An electronic device, comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.

16. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-7.

17. A computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 7.

Images