Methods, apparatus, computer equipment, and storage media for generating customized question banks

Abstract

This application relates to a method, apparatus, computer device, and storage medium for generating a customized question bank. The method includes: acquiring preset merchant requirements and basic user information; obtaining a user profile based on the user information, where the merchant requirements include merchant keywords manually entered by the merchant; acquiring a preset question bank and determining target question tags in the preset question bank that match the merchant keywords; generating an initial customized question bank based on the target question tags; acquiring user answer data corresponding to the initial customized question bank; inputting the user answer data into a preset collaborative recommendation model; revising the user profile using the collaborative recommendation model to obtain a first user tag; the collaborative recommendation model is trained based on the user's historical answer data; and optimizing the initial customized question bank based on the first user tag to generate a first target customized question bank. This method enables the rapid generation of customized question banks for users.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus, computer device, and storage medium for generating customized question banks. Background Technology

[0002] In order to convey brand image and improve the effectiveness of every communication with consumers, more and more quiz games have emerged, which can help merchants quickly establish a basic understanding of consumers.

[0003] In traditional technology, existing quiz games usually randomly select questions from a shared question bank configured in the backend. The number and categories of questions are very limited, which cannot match the needs of all merchants and consumers. As a result, merchants obtain very little effective consumer information through existing quiz games, and consumers have a poor sense of participation in the quiz games.

[0004] Currently, no effective solution has been proposed for the problem that questions in related technologies can only be randomly selected, resulting in merchants obtaining little effective information through quiz games and poor consumer engagement. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, apparatus, computer device, and computer-readable storage medium capable of generating customized question banks to address the aforementioned technical problems.

[0006] Firstly, this application provides a method for generating a customized question bank. The method includes:

[0007] Obtain preset merchant requirements and user basic information, and generate user profiles based on user basic information. Merchant requirements include merchant keywords, which are manually entered by the merchant.

[0008] Obtain a preset question bank and identify target question tags in the preset question bank that match the merchant's keywords. Generate an initial customized question bank based on the target question tags.

[0009] Obtain user answer data corresponding to the initial customized question bank, input the user answer data into the preset collaborative recommendation model, use the collaborative recommendation model to revise the user profile, and obtain the first user tag. The collaborative recommendation model is generated by training based on the user's historical answer data.

[0010] The initial customized question bank is optimized based on the first user tag to generate the first target customized question bank.

[0011] In one embodiment, the merchant requirements also include a preset quantity threshold, determining target question tags in the preset question bank that match the merchant's keywords, and generating an initial customized question bank based on the target question tags, including:

[0012] A basic customized question bank is obtained by performing a basic search in the preset question bank based on the merchant's keywords;

[0013] When the number of questions in the basic customized question bank is greater than or equal to the preset number threshold, the basic customized question bank will be used as the initial customized question bank.

[0014] When the number of questions in the basic customized question bank is less than the preset threshold, an extended search is performed in the preset question bank to obtain an extended customized question bank. The basic customized question bank and the extended customized question bank are used together as the initial customized question bank.

[0015] In one embodiment, when the number of questions in the basic customized question bank is less than a preset threshold, an extended search is performed in the preset question bank to obtain an extended customized question bank, including:

[0016] Obtain the merchant part-of-speech category corresponding to the merchant keywords. The merchant part-of-speech category includes fuzzy synonyms associated with the merchant keywords. Then, search the preset question bank based on the fuzzy synonyms to obtain secondary search results.

[0017] Obtain the keywords corresponding to the basic questions in the basic customized question bank, as well as the part-of-speech categories of the questions corresponding to the keywords. The part-of-speech categories include fuzzy keywords associated with the keywords. Then, search the preset question bank based on the fuzzy keywords to obtain the joint search results.

[0018] The secondary search results and combined search results are used together to form an expanded and customized question bank.

[0019] In one embodiment, user answer data is input into a preset collaborative recommendation model, and the user profile is revised using the collaborative recommendation model to obtain the first user tag, which includes:

[0020] When it is detected that a user is participating in the quiz for the first time, the user profile is revised based on the user's answer data and the target question tags to obtain the first user tag;

[0021] When it is detected that a user is not participating in the quiz for the first time, historical quiz data is obtained. The current user's quiz data, target question tags, historical quiz data, and user profile are input into the collaborative recommendation model. The collaborative recommendation model is used to match the user profile with the target question tags to obtain the first matching result. The first matching result is corrected based on the historical quiz data and the current user's quiz data to obtain the first user tag.

[0022] In one embodiment, the collaborative recommendation model includes at least one weight type parameter. After correcting the matching results based on historical answer data and current user answer data to obtain the first user tag, the method further includes:

[0023] Based on the answer data, at least one type of weight parameter in the collaborative recommendation model is modified to obtain the modified target collaborative recommendation model. The answer data includes the current user answer data and historical answer data.

[0024] The answer data and user basic information are input into the target collaborative recommendation model, and the user profile is matched with the target question tags to obtain the second matching result. The second matching result is corrected according to the user's answer results to obtain the second user tag.

[0025] The initial customized question bank is optimized based on the second user tag to generate a second target customized question bank.

[0026] In one embodiment, obtaining a user profile based on basic user information includes:

[0027] If it is detected that the user is participating in the quiz for the first time, a user profile is obtained based on the user's basic information;

[0028] If it is detected that the user is not participating in the quiz for the first time, a user profile is obtained based on the user's basic information and historical quiz data.

[0029] In one embodiment, there are multiple users. After optimizing the initial customized question bank based on the first user tag and generating the first target customized question bank, the above method further includes:

[0030] Obtain the average accuracy value based on the merchant's needs;

[0031] The questions with large deviation values ​​are calculated based on the answer data of multiple users, and the user's answer accuracy rate is obtained based on the questions with large deviation values ​​and the first user tag corresponding to the user.

[0032] When a user's answer accuracy rate is higher than or equal to the average accuracy rate, a first new question that does not correspond to the first user tag is selected from the questions with large deviation values, and the first new question is added to the first target customized question bank to generate the third target customized question bank.

[0033] When a user's correct answer rate is lower than the average correct answer rate, a second new question corresponding to the first user tag is selected from the questions with large deviation values, and the second new question is added to the first target customized question bank to generate the fourth target customized question bank.

[0034] Secondly, this application also provides an apparatus for automating the creation of a customized question bank. The apparatus includes:

[0035] The acquisition module is used to acquire preset merchant requirements, user basic information and preset question bank. It obtains user profiles based on user basic information. Merchant requirements include merchant keywords, which are manually entered by the merchant. It determines the target question tags in the preset question bank that match the merchant keywords and generates an initial customized question bank based on the target question tags.

[0036] The calculation module is used to obtain user answer data corresponding to the initial customized question bank, input the user answer data into the preset collaborative recommendation model, and use the collaborative recommendation model to revise the user profile to obtain the first user tag. The collaborative recommendation model is generated by training based on the user's historical answer data.

[0037] The generation module is used to optimize the initial customized question bank based on the first user tags and generate the first target customized question bank.

[0038] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0039] Obtain preset merchant requirements and user basic information, and generate user profiles based on user basic information. Merchant requirements include merchant keywords, which are manually entered by the merchant.

[0040] Obtain a preset question bank and identify target question tags in the preset question bank that match the merchant's keywords. Generate an initial customized question bank based on the target question tags.

[0041] Obtain user answer data corresponding to the initial customized question bank, input the user answer data into the preset collaborative recommendation model, use the collaborative recommendation model to revise the user profile, and obtain the first user tag. The collaborative recommendation model is generated by training based on the user's historical answer data.

[0042] The initial customized question bank is optimized based on the first user tag to generate the first target customized question bank.

[0043] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0044] Obtain preset merchant requirements and user basic information, and generate user profiles based on user basic information. Merchant requirements include merchant keywords, which are manually entered by the merchant.

[0045] Obtain a preset question bank and identify target question tags in the preset question bank that match the merchant's keywords. Generate an initial customized question bank based on the target question tags.

[0046] Obtain user answer data corresponding to the initial customized question bank, input the user answer data into the preset collaborative recommendation model, use the collaborative recommendation model to revise the user profile, and obtain the first user tag. The collaborative recommendation model is generated by training based on the user's historical answer data.

[0047] The initial customized question bank is optimized based on the first user tag to generate the first target customized question bank.

[0048] The aforementioned method, apparatus, computer equipment, and storage medium for generating customized question banks obtain an initial customized question bank and generate a user profile based on the user's basic information. The user profile is then revised using a collaborative recommendation model to obtain a first user tag. The initial customized question bank is then optimized based on the first user tag to generate a first target customized question bank. This solves the problem that merchants obtain little effective information through quiz games and that consumer engagement is poor, thereby improving user engagement. Attached Figure Description

[0049] Figure 1 This is a flowchart illustrating a method for generating a customized question bank in one embodiment;

[0050] Figure 2 This is a flowchart illustrating a method for generating a customized question bank in another embodiment;

[0051] Figure 3 This is a flowchart illustrating a preferred embodiment of a method for generating a customized question bank.

[0052] Figure 4 A structural block diagram of a custom question bank device is provided for one embodiment.

[0053] Figure 5 An internal structure diagram of a customized question bank computer device is generated for one embodiment. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0055] In one embodiment, such as Figure 1 As shown, a method for generating a customized question bank is provided. Figure 1This is a flowchart of a method for generating a customized question bank according to an embodiment of this application. The process includes the following steps:

[0056] Step S110: Obtain preset merchant requirements and basic user information; obtain a user profile based on the basic user information; the merchant requirements include merchant keywords, which are manually entered by the merchant.

[0057] On one hand, the user's basic information includes gender, age, etc., and the user can participate in the survey through various means such as scanning a QR code or participating in a questionnaire. On the other hand, the merchant's requirements include the merchant keywords they need to be surveyed and their requirements for the number of questions.

[0058] Step S120: Obtain a preset question bank, determine the target question tags in the preset question bank that match the merchant's keywords, and generate an initial customized question bank based on the target question tags.

[0059] The initial customized question bank is generated by searching a pre-set question bank based on the merchant keywords manually entered by the merchants. Specifically, the pre-set question bank consists of questions pre-entered by us and questions entered by the merchants. All questions have defined keywords, and each keyword is assigned a category.

[0060] Step S130: Obtain user answer data corresponding to the initial customized question bank, input the user answer data into a preset collaborative recommendation model, and use the collaborative recommendation model to revise the user profile to obtain a first user tag. The collaborative recommendation model is generated by training based on the user's historical answer data.

[0061] Specifically, when a user is participating in the quiz for the first time, the user profile is revised based on the user's current answer data to obtain the first user tag. When a user is not participating in the quiz for the first time, the collaborative recommendation model is trained based on the user's historical answer data. After training, the current answer data is input into the collaborative recommendation model, and the collaborative recommendation model is used to match the user profile with the target question tag to obtain the first matching result. The first matching result is then corrected based on the current answer data and historical answer data to obtain the first user tag.

[0062] Step S140: Optimize the initial customized question bank according to the first user tag and generate the first target customized question bank.

[0063] Specifically, the initial customized question bank is further filtered using the first user tag to obtain the first target customized question bank. This first target customized question bank, based on the user's first user tag, is distributed to each user to push content of interest to them, thus achieving accurate and targeted question delivery. Furthermore, algorithmic analysis of past answer data can be used to obtain the first user tag containing answer accuracy rates. Based on this first user tag, a first target customized question bank that guarantees a certain accuracy rate is obtained and distributed to the corresponding users. This first target customized question bank can maintain a high average accuracy rate for users. Optimizing the initial customized question bank and generating the first target customized question bank can ensure that the average accuracy rate of multiple users remains at a high level.

[0064] Through steps S110 to S140 above, firstly, basic user information is obtained, and a user profile is generated based on this information, including gender, age, etc.; secondly, preset merchant requirements are obtained, including the number of questions, merchant keywords, etc.; thirdly, a preset question bank is obtained, which includes questions pre-entered by us and questions pre-entered by merchants, and target question tags matching merchant keywords in the preset question bank are determined, and an initial customized question bank is generated based on these target question tags; finally, the initial customized question bank is distributed to each user, and the user corresponding to the initial customized question bank is obtained. The user's answer data is input into a pre-set collaborative recommendation model, which is trained based on the user's historical answer data. The model matches the user profile with the target question tag to obtain a first matching result. This first matching result is then corrected based on the answer data to obtain a first user tag. The answer data includes historical and current answer data. Finally, the initial customized question bank is optimized based on this first user tag to generate a first target customized question bank. This ensures that the average accuracy rate of multiple users remains at a high level. Therefore, unlike existing quiz activity methods, this application defines different customized question banks by accurately assigning different user tags to different users, maintaining a relatively high average accuracy rate for multiple users. This not only increases user engagement but also maximizes the user information obtained from a limited number of questions, significantly improving merchants' understanding of users and strengthening the connection between users and merchants.

[0065] In one embodiment, the method for generating a customized question bank further includes:

[0066] A basic customized question bank is obtained by performing a basic search in the preset question bank based on the merchant's keywords;

[0067] When the number of questions in the basic customized question bank is greater than or equal to the preset number threshold, the basic customized question bank will be used as the initial customized question bank.

[0068] When the number of questions in the basic customized question bank is less than the preset threshold, an extended search is performed in the preset question bank to obtain an extended customized question bank. The basic customized question bank and the extended customized question bank are used together as the initial customized question bank.

[0069] Specifically, a basic search is performed in the preset question bank using the merchant keywords initially entered by the merchant to obtain a basic customized question bank. When the number of questions in the basic customized question bank is greater than or equal to the preset threshold in the merchant's requirements after only the basic search, the search ends and the basic customized question bank is used as the initial customized question bank. When the number of questions in the basic customized question bank is less than the preset threshold after the basic search, an extended search is performed in the preset question bank to obtain an extended customized question bank. At this time, the extended customized question bank and the basic customized question bank are used together as the initial customized question bank. The preset question bank includes the question bank that we have entered in advance and the questions that the merchant has entered in advance.

[0070] In one embodiment, the method for generating a customized question bank further includes:

[0071] Obtain the merchant part-of-speech category corresponding to the merchant keywords. The merchant part-of-speech category includes fuzzy synonyms associated with the merchant keywords. Then, search the preset question bank based on the fuzzy synonyms to obtain secondary search results.

[0072] Obtain the keywords corresponding to the basic questions in the basic customized question bank, as well as the part-of-speech categories of the questions corresponding to the keywords. The part-of-speech categories include fuzzy keywords associated with the keywords. Then, search the preset question bank based on the fuzzy keywords to obtain the joint search results.

[0073] The secondary search results and combined search results are used together to form an expanded and customized question bank.

[0074] Specifically, when establishing the pre-set question bank, each question is assigned its own keywords. A question can have multiple keywords, and each keyword is assigned a different part-of-speech category. First, the part-of-speech category corresponding to the merchant's keywords is obtained. This part-of-speech category includes fuzzy synonyms associated with the merchant's keywords. There can be multiple fuzzy synonyms. The pre-set question bank is searched based on these fuzzy synonyms to obtain secondary search results. If the number of questions after the secondary search meets the requirements specified in the merchant's needs, the extended search ends, and the secondary search results are used as the extended customized question bank. If the number of questions after the secondary search still does not meet the requirements specified in the merchant's needs, a joint search is performed to obtain the question keywords corresponding to the basic questions in the basic customized question bank, as well as the question part-of-speech categories corresponding to these questions. There can be multiple question keywords and question part-of-speech categories. These question part-of-speech categories include fuzzy keywords associated with the question keywords. There can be multiple fuzzy keywords. The pre-set question bank is searched based on these fuzzy keywords to obtain joint search results. At this point, the secondary search results and the joint search results are used together as the extended customized question bank. This demonstrates that, on the one hand, this application can obtain an initial customized question bank from both basic and extended searches. The extended search is further divided into two steps: secondary search and combined search. In the secondary search, the fuzzy synonyms are still similar in meaning to the merchant's keywords, while in the combined search, the fuzzy keywords may differ significantly in meaning from the merchant's keywords. However, the combined search yields a larger number of questions. This step can meet the merchant's requirements for the number of questions from different perspectives and better adapt to different situations. On the other hand, traditional question bank entry methods suffer from high costs, long processing times, and narrow applicability. In contrast, the customized question bank in this application can be quickly generated based on user tags, resulting in higher efficiency. Furthermore, different customized question banks can be generated for different users, providing a better answering experience.

[0075] In one embodiment, the method for generating a customized question bank further includes:

[0076] When it is detected that a user is participating in the quiz for the first time, the user profile is revised based on the user's answer data and question tags to obtain the first user tag. When it is detected that a user is not participating in the quiz for the first time, historical answer data is obtained. The current user answer data, target question tags, historical answer data, and user profile are input into the collaborative recommendation model. The collaborative recommendation model is used to match the user profile with the target question tags to obtain the first matching result. The first matching result is corrected based on historical answer data and current user answer data to obtain the first user tag.

[0077] Specifically, when it is detected that a user has no historical answer data, the user profile is revised only based on the current user answer data and question tags to obtain the first user tag;

[0078] When historical answer data is detected, this data is retrieved. The current user profile, target question tags, historical answer data, and user profile are input into the collaborative recommendation model. The model matches the user profile with the target question tags to obtain a first matching result. This first matching result is then corrected based on the answer data, which includes both current and historical answer data, to obtain a first user tag. Training the collaborative recommendation model involves acquiring training data, including question tags, user answering time feedback, and user answering accuracy. This training data is divided into a training set and a validation set for training and validation of the collaborative recommendation model. Therefore, this application first obtains a user profile from basic user information, then revises the user profile to obtain a first user tag. Through these steps, relatively accurate user tags can be obtained, facilitating the delivery of customized question banks to users based on their specific circumstances. This approach aims to obtain more basic user information and enhance user engagement in the quiz activity.

[0079] In one embodiment, a method for generating a customized question bank is provided. Figure 2 The flowchart for another method of generating a customized question bank for this application is shown in the figure. The process includes the following steps:

[0080] Step S240: Based on the answer data, at least one weight type parameter in the collaborative recommendation model is modified to obtain a modified target collaborative recommendation model. The answer data includes current user answer data and historical answer data. The answer data and user basic information are input into the target collaborative recommendation model, and the user profile is matched with the question tags using the target collaborative recommendation model to obtain a second matching result. The second matching result is modified based on the user answer data to obtain a second user tag. The initial customized question bank is optimized based on the second user tag to generate a second target customized question bank.

[0081] Specifically, after the collaborative recommendation model is trained, it can be optimized. This optimization mainly involves adjusting at least one type of weight parameter in the model based on actual question-answering data. This type of weight parameter includes feedback time, etc. After the collaborative recommendation model is optimized, a target collaborative recommendation model is obtained. Question-answering data and user basic information are input into the target collaborative recommendation model. The target collaborative recommendation model matches user profiles with question tags to obtain a second matching result. The second matching result is then adjusted based on the user's question-answering data to obtain a second user tag. Finally, a second target customized question bank is generated based on the second user tag.

[0082] It can be seen that this application not only obtains more accurate user tags by correcting the user tags themselves, but also corrects the collaborative recommendation model by modifying its weight type parameters according to different actual situations, so that the collaborative recommendation model can be better applied in real-world environments and output more accurate user tags.

[0083] In one embodiment, the method for generating a customized question bank further includes:

[0084] If it is detected that the user is participating in the quiz for the first time, the user profile is obtained based on the user's basic information; if it is detected that the user is not participating in the quiz for the first time, the user profile is obtained based on the user's basic information and the historical quiz data.

[0085] In one embodiment, the method for generating a customized question bank further includes:

[0086] The system obtains the average accuracy rate based on merchant requirements; calculates questions with large deviation values ​​based on the answer data of multiple users, and obtains the user's answer accuracy rate based on the questions with large deviation values ​​and the first user tag corresponding to the user; when the user's answer accuracy rate is higher than or equal to the average accuracy rate, it filters out the questions with large deviation values ​​to obtain the first new question that does not correspond to the first user tag, and adds the first new question to the first target customized question bank to generate the third target customized question bank; when the user's answer accuracy rate is lower than the average accuracy rate, it filters out the questions with large deviation values ​​to obtain the second new question that corresponds to the first user tag, and adds the second new question to the first target customized question bank to generate the fourth target customized question bank.

[0087] Specifically, based on merchant needs, an average accuracy rate is obtained, which can be set to 75%, 70%, etc. Based on the answer data, questions with large deviation values ​​are identified. These large deviation values ​​are defined as follows: when the question tag corresponds to a first user tag, the question's accuracy rate is greater than a preset first accuracy rate value; when the question tag does not correspond to the first user tag, the question's accuracy rate is less than a preset second accuracy rate value, and the first accuracy rate value is greater than the second accuracy rate value. Multiple large deviation value questions can exist. User answer data, including accuracy rates, is obtained. When a user's accuracy rate is higher than or equal to the average accuracy rate value, a first new question that does not correspond to the first user tag is selected from the large deviation value questions and added to the first target customized question bank to generate a third target customized question bank. When a user's accuracy rate is lower than the average accuracy rate value, a second new question that corresponds to the first user tag is selected from the large deviation value questions and added to the first target customized question bank to generate a fourth target customized question bank. This shows that by adding questions with large deviation values ​​that correspond to or do not correspond to the first user's tag, the average accuracy of multiple users can be further adjusted. This can ensure that the accuracy of more users meets the needs of merchants, provide more users with a better answering experience, and help merchants obtain more user information from a limited number of questions.

[0088] This embodiment also provides a specific implementation of a method for generating a customized question bank, such as... Figure 3 As shown, Figure 3 This is a flowchart of a preferred embodiment of a method for generating a customized question bank.

[0089] Step S310: Obtain merchant requirements and basic user information. This basic user information may include gender, age, order data, etc. If the user has previously participated in the quiz, their basic information also includes historical quiz data. Merchant requirements include a preset threshold for the number of questions and merchant keywords. Obtain a preset question bank, which consists of questions pre-entered by us and questions pre-entered by merchants. Pre-entered questions can be from past activities or questions from other merchants, extracted through comparison. The pre-entered questions can have their length, keywords, number of answers, etc., set, and can be associated with relevant basic questions. Merchants may also choose not to enter questions.

[0090] Step S320: Based on the merchant's keywords, perform a basic search in the preset question bank to obtain a basic customized question bank. When the number of questions in the basic customized question bank is greater than or equal to a preset threshold, use the basic customized question bank as the initial customized question bank. When the number of questions in the basic customized question bank is less than the preset threshold, perform a further extended search in the preset question bank to obtain an extended customized question bank. Use the extended customized question bank and the basic customized question bank together as the initial customized question bank. Furthermore, the extended search includes secondary search and combined search. First, a secondary search is performed to obtain the part-of-speech category corresponding to the merchant's input keywords. This part-of-speech category includes fuzzy synonyms associated with the merchant keywords. A search is then conducted in a preset question bank based on these fuzzy synonyms to obtain secondary search results. If the number of questions after the secondary search is greater than or equal to a preset threshold, the search ends, and the secondary search results are used as the extended customized question bank. If the number of questions after the secondary search is still less than the preset threshold, a combined search is performed to obtain the question keywords corresponding to the basic questions in the basic customized question bank, as well as the part-of-speech categories corresponding to these keywords. These part-of-speech categories include fuzzy keywords associated with the question keywords. While the fuzzy synonyms in the secondary search are still semantically similar to the merchant keywords, the fuzzy keywords in the combined search may have significantly different meanings from the merchant keywords. However, the combined search yields a larger number of questions. After the combined search, both the secondary search results and the combined search results are used as the extended customized question bank.

[0091] Step S330: After obtaining the initial customized question bank, distribute the initial customized question bank to the corresponding users, obtain the user answer data corresponding to the initial customized question bank, input the user answer data into the preset collaborative recommendation model, and use the collaborative recommendation model to match the user profile with the target question tags to obtain the first matching result. The first matching result is then corrected based on historical answer data and current user answer data to obtain the first user tag. Subsequently, at least one weight type parameter in the collaborative recommendation model is corrected based on the answer data to obtain the corrected target collaborative recommendation model. This answer data includes current user answer data and historical answer data. The user's answer data and basic user information are input into the target collaborative recommendation model to obtain the second user tag. A second target customized question bank is then generated based on the second user tag.

[0092] Step S340 further refines the user-customized question bank; based on merchant needs, it obtains the average accuracy rate, questions with large deviation values, and the user's answer accuracy rate based on the questions with large deviation values ​​and the user's corresponding second user tag. When the user's answer accuracy rate is higher than or equal to the average accuracy rate, a first new question that does not correspond to the second user tag is selected from the questions with large deviation values, and this first new question is added to the second target customized question bank to generate a third target customized question bank; when the user's answer accuracy rate is lower than the average accuracy rate, a second new question that corresponds to the second user tag is selected from the questions with large deviation values, and this second new question is added to the second target customized question bank to generate a fourth target customized question bank. The method for generating customized question banks in this application, on the one hand, can efficiently generate customized question banks for different users; on the other hand, it allows merchants to obtain as much basic information and preference information as possible from users through a limited number of questions, and can greatly enhance user engagement and strengthen the connection between merchants and users.

[0093] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0094] Based on the same inventive concept, this application also provides a device for generating a customized question bank to implement the above-described method for generating a customized question bank. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the device for generating a customized question bank provided below can be found in the limitations of the method for generating a customized question bank described above, and will not be repeated here.

[0095] In one embodiment, such as Figure 4 As shown, a device for generating a customized question bank is provided, comprising: an acquisition module 41, a calculation module 42, and a generation module 43, wherein:

[0096] The acquisition module 41 is used to acquire preset merchant requirements, user basic information and preset question bank, obtain user profile based on the user basic information, the merchant requirements include merchant keywords, the merchant keywords are manually entered by the merchant, determine the target question tags in the preset question bank that match the merchant keywords, and generate an initial customized question bank based on the target question tags;

[0097] The calculation module 42 is used to obtain user answer data corresponding to the initial customized question bank, input the user answer data into a preset collaborative recommendation model, and use the collaborative recommendation model to revise the user profile to obtain a first user tag. The collaborative recommendation model is generated by training based on the user's historical answer data.

[0098] The generation module 43 is used to optimize the initial customized question bank based on the first user tag and generate a first target customized question bank.

[0099] Specifically, the acquisition module 41 is connected to the calculation module 42. The acquisition module 41 collects preset merchant requirements, user basic information, and a preset question bank. Based on the user's basic information and historical answer data, the acquisition module 41 obtains a user profile and, based on merchant keywords, obtains an initial customized question bank through basic and extended searches. This initial customized question bank is then distributed to the corresponding users, and the user's answer data is sent to the calculation module 42. The calculation module 42 acquires the user's answer data corresponding to the initial customized question bank and inputs it into a preset collaborative recommendation model. This collaborative recommendation model is trained and generated based on the user's historical answer data. The collaborative recommendation model matches the user profile with the target question tag to obtain a first matching result. Based on historical answer data and current user answer data, the first matching result is corrected to obtain a first user tag. This first user tag is input into the generation module 43. The generation module 43 optimizes the initial customized question bank based on the first user tag and generates a first target customized question bank.

[0100] The aforementioned device for generating customized question banks offers several advantages. Firstly, traditional question entry methods suffer from high costs, long processing times, and limited applicability. In contrast, the customized question bank in this application can be generated quickly based on user tags, resulting in higher efficiency and greater applicability to users. Secondly, generating customized question banks based on user tags ensures that users' answer accuracy is neither too low nor too high, maintaining a relatively high score. This significantly enhances user engagement and allows merchants to gather as much basic information and preferences as possible from a limited number of questions.

[0101] Each module in the aforementioned device for generating a customized question bank can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0102] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 5 As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements an interface information synchronization method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0103] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the segmented data statistics device to which the present application is applied. A specific segmented data statistics device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0104] Those skilled in the art should understand that the technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments have been described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0105] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for generating a customized question bank, characterized in that, The method includes: Obtain preset merchant requirements and user basic information, and obtain user profiles based on the user basic information. The merchant requirements include merchant keywords, which are manually entered by the merchant. Obtain a preset question bank, determine the target question tags in the preset question bank that match the merchant's keywords, and generate an initial customized question bank based on the target question tags, including: Based on the merchant keywords, a basic search is performed in the preset question bank to obtain a basic customized question bank; When the number of questions in the basic customized question bank is greater than or equal to the preset number threshold, the basic customized question bank is used as the initial customized question bank. When the number of questions in the basic customized question bank is less than the preset threshold, an extended search is performed in the preset question bank to obtain an extended customized question bank, including: Obtain the merchant part-of-speech category corresponding to the merchant keyword. The merchant part-of-speech category includes fuzzy synonyms associated with the merchant keyword. Then, search the preset question bank based on the fuzzy synonyms to obtain secondary search results. Obtain the question keywords corresponding to the basic questions in the basic customized question bank, and the question part-of-speech categories corresponding to the question keywords. The question part-of-speech categories include fuzzy keywords associated with the question keywords. Then, search the preset question bank according to the fuzzy keywords to obtain joint search results. The secondary search results and the combined search results are used together as the extended customized question bank; The basic customized question bank and the extended customized question bank are used together as the initial customized question bank; Obtain user answer data corresponding to the initial customized question bank, input the user answer data into a preset collaborative recommendation model, use the collaborative recommendation model to revise the user profile, and obtain the first user tag. The collaborative recommendation model is generated by training based on the user's historical answer data. The initial customized question bank is optimized based on the first user tag to generate a first target customized question bank.

2. The method according to claim 1, characterized in that, The step of inputting the user's answer data into a preset collaborative recommendation model, and using the collaborative recommendation model to revise the user profile to obtain the first user tag includes: When it is detected that the user is participating in the quiz for the first time, the user profile is revised based on the user's quiz data and the target question tag to obtain the first user tag; When it is detected that the user is not participating in the quiz for the first time, the historical quiz data is obtained, and the current user quiz data, the target question tag, the historical quiz data, and the user profile are input into the collaborative recommendation model. The collaborative recommendation model is used to match the user profile with the target question tag to obtain a first matching result. The first matching result is corrected according to the historical quiz data and the current user quiz data to obtain the first user tag.

3. The method according to claim 2, characterized in that, The collaborative recommendation model includes at least one weight type parameter. After correcting the first matching result based on the historical answer data and the current user answer data to obtain the first user tag, the method further includes: Based on the answer data, at least one of the weight type parameters in the collaborative recommendation model is modified to obtain a modified and complete target collaborative recommendation model. The answer data includes the current user answer data and the historical answer data. The answer data and the user's basic information are input into the target collaborative recommendation model, and the user profile is matched with the target question tag using the target collaborative recommendation model to obtain a second matching result. The second matching result is then corrected based on the answer data to obtain a second user tag. The initial customized question bank is optimized based on the second user tag to generate a second target customized question bank.

4. The method according to claim 1, characterized in that, The process of obtaining a user profile based on the user's basic information includes: If it is detected that the user is participating in the quiz for the first time, the user profile is obtained based on the user's basic information; If it is detected that the user is not participating in the quiz for the first time, the user profile is obtained based on the user's basic information and the historical quiz data.

5. The method according to any one of claims 1 to 4, characterized in that, There are multiple users. After optimizing the initial customized question bank based on the first user tag and generating the first target customized question bank, the method further includes: Obtain the average accuracy value based on the merchant's requirements; The question with a large deviation value is calculated based on the answer data of multiple users, and the user's answer accuracy rate is obtained based on the question with a large deviation value and the first user tag corresponding to the user. When the user's answer accuracy rate is higher than or equal to the average accuracy rate value, a first new question that does not correspond to the first user tag is selected from the questions with large deviation values, and the first new question is added to the first target customized question bank to generate a third target customized question bank. When the user's answer accuracy rate is lower than the average accuracy rate, a second new question corresponding to the first user tag is selected from the questions with large deviation values, and the second new question is added to the first target customized question bank to generate a fourth target customized question bank.

6. An apparatus for generating a customized question bank, characterized in that, The device includes: The acquisition module is used to acquire preset merchant requirements, user basic information and preset question bank, obtain user profile based on the user basic information, the merchant requirements include merchant keywords, the merchant keywords are manually entered by the merchant, determine the target question tags in the preset question bank that match the merchant keywords, and generate an initial customized question bank based on the target question tags; The acquisition module is further configured to perform a basic search in the preset question bank based on the merchant keywords to obtain a basic customized question bank; when the number of questions in the basic customized question bank is greater than or equal to the preset number threshold, the basic customized question bank is used as the initial customized question bank; when the number of questions in the basic customized question bank is less than the preset number threshold, an extended search is performed in the preset question bank to obtain an extended customized question bank, and the basic customized question bank and the extended customized question bank are used together as the initial customized question bank. The acquisition module is further configured to acquire the merchant part-of-speech category corresponding to the merchant keyword, wherein the merchant part-of-speech category includes fuzzy synonyms associated with the merchant keyword, and to search the preset question bank based on the fuzzy synonyms to obtain secondary search results; acquire the question keywords corresponding to the basic questions in the basic customized question bank, and the question part-of-speech category corresponding to the question keywords, wherein the question part-of-speech category includes fuzzy keywords associated with the question keywords, and to search the preset question bank based on the fuzzy keywords to obtain joint search results; and use the secondary search results and the joint search results together as the extended customized question bank; The calculation module is used to obtain user answer data corresponding to the initial customized question bank, input the user answer data into the preset collaborative recommendation model, and use the collaborative recommendation model to revise the user profile to obtain the first user tag. The collaborative recommendation model is generated by training based on the user's historical answer data. The generation module is used to optimize the initial customized question bank based on the first user tag and generate a first target customized question bank.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images