Processing method and device of questionnaire, electronic equipment and storage medium

Abstract

The application provides a processing method and device of a questionnaire, electronic equipment and a storage medium, and relates to the technical field of computers. The method obtains the reply results of the research users for each question in the questionnaire, and respectively inputs the game feature data of the research users corresponding to each question in the questionnaire into a first model generated in advance to obtain the prediction results of each question in the questionnaire. Then, according to the prediction results of each question in the questionnaire, it is determined whether the reply results of the research users for each question are valid, so as to retain the valid reply results in the questionnaire or eliminate the invalid reply results in the questionnaire. The technical solution saves the tedious sorting and identification process of the answered questionnaire by manual work, and improves the recycling efficiency of the questionnaire.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, electronic device, and storage medium for processing questionnaires. Background Technology

[0002] With the continuous development of Internet technology, more and more users are making higher demands on the services provided by suppliers when using the Internet. In order to meet user experience and improve service design, suppliers establish contact with users to solicit their opinions, thereby continuously improving their services.

[0003] In the existing technology, after the supplier prepares the questionnaire, it sends the questionnaire to the user terminal to allow the user to respond. After the response is completed, the supplier obtains the response results, and relevant personnel screen the response results to obtain usable questionnaire response results.

[0004] However, due to the large number of questions in the questionnaire and / or the large number of responses resulting from the large number of users participating in the questionnaire, manual screening is inefficient. Summary of the Invention

[0005] In view of this, this application provides a method, apparatus, electronic device and storage medium for processing questionnaires, in order to overcome the problem that the large number of questions in the questionnaire, or / and the large number of users participating in the questionnaire result in a large number of responses, and that the method of manually screening valid responses is inefficient.

[0006] The first aspect of this application provides a method for processing a questionnaire, the method comprising:

[0007] Obtain the responses of survey users to each question in the questionnaire, which includes at least one question;

[0008] The game feature data of the survey users corresponding to each question of the questionnaire are respectively input into the pre-generated first model to obtain the prediction results of each question in the questionnaire. The first model is trained based on the game feature data of at least one user corresponding to each question.

[0009] Based on the predicted results of each question in the questionnaire, determine whether the responses of the survey users to each question are valid, so as to retain the valid responses in the questionnaire or remove the invalid responses in the questionnaire.

[0010] A second aspect of this application provides a questionnaire processing apparatus, the apparatus comprising:

[0011] The acquisition module is used to acquire the responses of survey users to each question in the questionnaire, which includes at least one question.

[0012] The determination module is used to input the game feature data of the survey users corresponding to each question of the questionnaire into a pre-generated first model to obtain the prediction results of each question in the questionnaire. The first model is trained based on the game feature data of at least one user corresponding to each question.

[0013] The processing module is used to determine whether the survey user's response to each question is valid based on the predicted results of each question in the questionnaire, so as to retain the valid response results in the questionnaire or remove the invalid response results in the questionnaire.

[0014] A third aspect of the embodiments of this application also provides an electronic device, including: a processor and a memory;

[0015] The memory stores computer-executed instructions;

[0016] The processor executes the computer execution instructions, causing the electronic device to perform the questionnaire processing method as described in the first aspect above.

[0017] A fourth aspect of this application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the questionnaire processing method described in the first aspect above.

[0018] The technical solution provided in this application obtains the responses of survey users to each question in a questionnaire, which includes at least one question. The game feature data of the survey users corresponding to each question is input into a pre-generated first model to obtain prediction results for each question. The first model is trained based on the game feature data of at least one user corresponding to each question. Then, based on the prediction results for each question, the validity of the survey user's responses is determined, retaining valid responses or removing invalid responses. This technical solution uses the game feature data of the user corresponding to the question to train a model predicting possible questionnaire results, thereby judging whether the responses meet expectations. This avoids situations where users maliciously or randomly fill out the questionnaire, saving the tedious manual process of sorting and identifying completed questionnaires and improving questionnaire return efficiency. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A schematic diagram illustrating an application scenario for processing questionnaires provided in this application embodiment;

[0021] Figure 2 A flowchart illustrating an embodiment of the questionnaire processing method provided in this application;

[0022] Figure 3 A flowchart illustrating Embodiment 2 of the questionnaire processing method provided in this application;

[0023] Figure 4 A flowchart illustrating Embodiment 3 of the questionnaire processing method provided in this application;

[0024] Figure 5 A flowchart illustrating Embodiment 4 of the questionnaire processing method provided in this application;

[0025] Figure 6 A flowchart illustrating Embodiment 5 of the questionnaire processing method provided in this application;

[0026] Figure 7 A flowchart illustrating Embodiment Six of the questionnaire processing method provided in this application;

[0027] Figure 8 A flowchart illustrating Embodiment Seven of the questionnaire processing method provided in this application;

[0028] Figure 9 A schematic diagram of a questionnaire processing device provided in an embodiment of this application;

[0029] Figure 10 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0030] To enable those skilled in the art to better understand the technical solutions of this application, the application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. However, this application can be implemented in many other ways different from those described above. Therefore, based on the embodiments provided in this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0031] It should be noted that the terms "first," "second," "third," etc., in the claims, specification, and drawings of this application are used to distinguish similar objects and are not used to describe a specific order or sequence. Such data are interchangeable where appropriate so that the embodiments of this application described herein can be implemented in a sequence other than that shown or described herein. Furthermore, the terms "comprising," "having," and their variations are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products, or apparatuses.

[0032] Before introducing the embodiments of this application, the technical terms and background technology involved in this application will be explained first:

[0033] User personas: an effective tool for identifying target users and aligning user needs with design direction. User personas abstract specific information about a user into tags, using these tags to concretize the user's image, thereby providing targeted services.

[0034] Common user profiling methods include: quantitative group classification, cluster analysis, and qualitative interviews.

[0035] With the continuous development of Internet technology, more and more users are making increasing demands on the services provided by suppliers when using the Internet. In order to improve user experience and improve service design, suppliers establish contact with users to solicit user opinions and continuously improve their services.

[0036] Taking game services as an example, in order to improve game design and enhance the player experience, it is crucial to establish direct or indirect communication between game developers and players to obtain player feedback. Currently, game developers typically choose to obtain relevant player information by distributing questionnaires designed by user researchers within the game system.

[0037] The existing technical solutions have the following main shortcomings:

[0038] 1) Although there are triggering conditions in the questionnaire, some questions need to be triggered by specific options in the preceding questions (for example, if a preceding question in the questionnaire is set to "Do you often buy skins?", when the user selects "yes", it will trigger "Are the skins you often buy legendary skins?").

[0039] This triggering method can generate questionnaire diversity, but it is limited by the number of combinations of triggering conditions. It lacks flexibility and diversity and loses the relevance of the questionnaire.

[0040] 2) After the system collects the questionnaires, technical staff need to spend extra time manually removing invalid questions or questionnaires that affect the survey results, which is time-consuming and labor-intensive.

[0041] 3) When there are too many invalid questionnaires, it is necessary to conduct the questionnaire survey again (at least once) to increase the cost of obtaining valid information.

[0042] This application addresses the aforementioned technical problems. The inventors have discovered that in existing technologies, questionnaire distribution lacks specificity, thus reducing questionnaire effectiveness. If questionnaires are tagged during distribution and sent to users corresponding to those tags, effectiveness can be improved. Furthermore, to prevent users from simply completing questionnaires to obtain rewards, big data analysis can be used to obtain feature data from one or more users corresponding to a given tag. This data can then generate a model capable of judging the effectiveness of a user's questionnaire responses, allowing for the retention of only valid questionnaires and avoiding the heavy workload of manually selecting invalid ones.

[0043] Based on the problems existing in the above-mentioned prior art Figure 1 This diagram illustrates an application scenario for processing questionnaires provided in an embodiment of this application, used to solve the aforementioned technical problems. For example... Figure 1 As shown in the diagram, the application scenario includes: terminal device 11 and electronic device 12.

[0044] The number of terminal devices 11 can be one or more, that is, the number of user accounts can be one or more; the terminal devices 11 can be devices with display functions such as mobile phones, computers, tablets, and smartwatches.

[0045] Taking game services as an example, electronic device 12 can be a network terminal device that provides services for a certain game application installed on terminal device 11.

[0046] In one possible implementation, to improve the user experience of the game application, such as the shopping recommendation experience, technicians might use high shopping frequency as a tag. Electronic device 12 can then read player tags from a user profile database and target players with high shopping frequency as recipients of the corresponding shopping recommendation questionnaires, i.e., survey users.

[0047] Furthermore, the terminal corresponding to the survey user can be terminal device 11. Electronic device 12 sends the questionnaire to terminal device 11. After the survey user responds, terminal device 11 returns the response result to electronic device 12.

[0048] The electronic device 12 contains a pre-recorded predicted result of the questionnaire. The predicted result is compared with the response result to determine the validity of the questionnaire.

[0049] It should be understood that the above application scenarios are merely examples and are not intended to limit this application. Furthermore, for any content not disclosed in these application scenarios, please refer to the following embodiments.

[0050] The technical solution of this application will be described in detail below through specific embodiments. It should be noted that the following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0051] Figure 2 This is a flowchart illustrating an embodiment of the questionnaire processing method provided in this application. Figure 2 As shown, the processing method for this questionnaire may include the following steps:

[0052] Step 21: Obtain the responses from survey users to each question in the questionnaire.

[0053] In this step, it is necessary to determine the validity of the responses to the questionnaires filled out by the survey users. First, it is necessary to obtain the responses to each question in the questionnaire.

[0054] Each questionnaire includes at least one question, and the responses to each question are the answers obtained by the survey users when they fill out the questionnaire. These responses can be the options that the survey users choose when answering the questions, and can be single-choice, multiple-choice, or open-ended questions.

[0055] Step 22: Input the characteristic data of the survey users corresponding to each question of the questionnaire into the pre-generated first model to obtain the prediction results of each question in the questionnaire.

[0056] The first model was trained based on the game feature data of at least one user corresponding to each question.

[0057] In this step, in order to compare the prediction results with the response results, the feature data of the survey users corresponding to the questions in the response results need to be input into the pre-generated first model to obtain the prediction results for the questions.

[0058] Optionally, in implementation, for the same question, the required model input is the game feature data of at least one user corresponding to the question, and the output is the options of the question, so as to construct the first model. When in use, the game feature data of the surveyed user is obtained, and the game data features of the surveyed user are input into the first model to obtain the possible options of the question, i.e., the prediction result.

[0059] In one possible implementation, taking the question of whether a product is satisfactory as an example, the user's game characteristic data could include whether they participated / purchased, the interval between their first participation and the initial launch, the frequency of use / participation, whether they recently participated / used the product, and the time since their last use / participation.

[0060] Optionally, a machine learning algorithm, such as a classification algorithm, can be used when training the first model.

[0061] In one possible implementation, consider the prediction of the answer to the question of whether someone is satisfied with a new activity or product:

[0062] The predicted results for this question can be ranked and represented as: very satisfied (60%), moderately satisfied (20%), average (10%), and dissatisfied (10%).

[0063] In another possible implementation, consider an open-ended suggestion question (e.g., "Please suggest something for this activity"):

[0064] The prediction results for this question can be sorted and represented as either positive / negative or positive / negative / neutral.

[0065] Example 1: Positive can mean: to have a suggestion; neutral can mean: to have no suggestion; negative can mean: to give a random answer.

[0066] Example 2: Positive can mean positive advice; neutral can mean fair advice; negative can mean negative advice.

[0067] Step 23: Based on the predicted results of each question in the questionnaire, determine whether the survey users' responses to each question are valid, so as to retain the valid responses in the questionnaire or remove the invalid responses.

[0068] In this step, after the electronic device receives the response, it needs to determine whether the response is valid, that is, whether it is the result of the survey user's serious response, and then either retain or remove the response.

[0069] Optionally, the survey user's responses to each question can be compared with the predicted results of each question in the corresponding questionnaire for that survey user to determine whether the responses to each question are valid.

[0070] In one possible implementation, if the response matches the preset expectation of the predicted result, the response is determined to be valid and the valid response is retained.

[0071] In another possible implementation, if the response does not meet the preset expectation of the prediction result, a further judgment is required to determine whether to retain or reject the response result. This process is given in the following embodiments and will not be repeated here.

[0072] Furthermore, retain questionnaires that do not contain invalid responses, or remove questionnaires that contain invalid responses.

[0073] The technical solution provided in this application involves obtaining the responses of survey users to each question in a questionnaire. The questionnaire includes at least one question. The game feature data of the survey users corresponding to each question are input into a pre-generated first model to obtain prediction results for each question. The first model is trained based on the game feature data of at least one user corresponding to each question. Then, based on the prediction results for each question, the validity of the survey user's responses is determined, retaining valid responses or removing invalid responses. This technical solution uses the game feature data of the user corresponding to the question to train a model predicting possible questionnaire results, thus enabling the judgment of expected responses. This avoids situations where users maliciously or randomly fill out the questionnaire, saving the tedious manual process of sorting and identifying completed questionnaires and improving questionnaire return efficiency.

[0074] Based on the above embodiments, Figure 3 This is a flowchart illustrating Embodiment Two of the questionnaire processing method provided in this application. Figure 3 As shown, prior to step 21, the questionnaire processing method may further include the following steps:

[0075] Step 31: Based on user profiles, determine the user tags of the surveyed users.

[0076] In this step, to increase the effectiveness and relevance of the questionnaire, it is first necessary to determine the user tags of the survey users, that is, which type of users the questionnaire should be sent to in order to obtain feedback.

[0077] Optionally, user profiles in the user profile database can be drawn using techniques such as cluster analysis combined with qualitative interviews and quantitative data analysis, which identify the user's corresponding user tags.

[0078] For example, Table 1 is a user tag illustration table of survey users provided in the embodiments of this application, as shown in Table 1:

[0079] Table 1:

[0080]

[0081]

[0082] Step 32: Extract questions from the question bank that correspond to the questionnaire tags that match the user tags, and generate a questionnaire.

[0083] In this step, the question bank contains multiple questions, each with a corresponding questionnaire tag. Then, questions corresponding to the questionnaire tags that match the user tags are extracted from the questionnaire question bank to generate a survey questionnaire.

[0084] Optionally, as an example, Table 2 lists the questions corresponding to the questionnaire tags provided in the embodiments of this application, as shown in Table 2:

[0085] Table 2:

[0086] Questionnaire Tags Question Number Mom 1、3、5、8、9、11、…… EasyBuy Men's Clothing 2、3、6、9、16、…… E-commerce Appliances 1、6、9、8、7、2、3 Easybuy Wine 3、5、6、12、25、45 Active players 48、69、58、44、96 New registered users 5、9、11、16、17、……

[0087] Furthermore, questions corresponding to the questionnaire tags that match the user tags are extracted. Taking User 1 (the survey user) in Table 1 as an example, the extracted questionnaire tags are: stay-at-home mom, Yigou men's clothing and Yigou appliances, and the corresponding question numbers are: 1, 3, 5, 8, 9, 11, 2, 6, 16, 7, ...

[0088] Optionally, the selection rules can be random selection, selection based on odd-numbered numbers, or selection of infrequently used questions, and the number of questions selected can also be set arbitrarily.

[0089] Specifically, the questions could be the first 9, 11, or 20 questions, or any 9, 12, or 20 questions with odd numbers, and then a questionnaire would be generated from these selected questions.

[0090] Each tag can correspond to multiple questions, and a question can also correspond to multiple tags; there are no restrictions on the correspondence.

[0091] Step 33: Send the survey questionnaire to the survey users.

[0092] In this step, after the questionnaire is generated in the above steps, the questionnaire is sent to the survey user.

[0093] Optionally, during the implementation of the entire plan, the number of survey users can be one or more (i.e., the questionnaire is sent to multiple users). Here, we use three survey users as an example to illustrate the process. Table 3 shows an example of questionnaire delivery.

[0094] Table 3:

[0095]

[0096] Furthermore, after generating the questionnaire, the electronic device pushes the questionnaire to the survey user. After the survey user selects to fill in the questionnaire, the questionnaire is arranged in a certain rule on the graphical user interface of the survey user's terminal device, and then the survey user answers it.

[0097] In one possible implementation, each page of the graphical user interface displays 10 questions. If there are more than 10 questions in the questionnaire, multiple pages are displayed; if there are no more than 10 questions, they are displayed on one page.

[0098] The technical solution provided in this application, based on user profiles, determines the user tags of survey users, extracts questions corresponding to questionnaire tags that match the user tags from the question bank, generates a questionnaire, and pushes the questionnaire to the survey users. In this technical solution, by pushing questions corresponding to questionnaire tags that match the survey users' user tags to the survey users, the targeting of the questionnaire delivery is improved, and the reliability of the questionnaire response results is enhanced.

[0099] The following is a brief introduction to the construction and application of the first model:

[0100] Figure 4 This is a flowchart illustrating Embodiment 3 of the questionnaire processing method provided in this application. Figure 4 As shown, the processing of this questionnaire also includes the following steps:

[0101] Step 41: Obtain game feature data for at least one user corresponding to the question.

[0102] In this step, game feature data of at least one user that matches the questionnaire tags are obtained from game data, game logs and other information, so as to provide training data for the subsequent construction of the first model.

[0103] Optionally, as an example, Table 4 shows examples of game feature data that need to be obtained for different questions, as shown in Table 4:

[0104] Table 4:

[0105] topic At least one user's game characteristic data Are you satisfied with the product? Whether to purchase, number of purchases per quarter... Are you satisfied with the new activities? The length of the interval between first participation in the launch event... Do you like buying electronic products? Number of purchases per quarter, whether purchases were made in the last 30 days... Are active players? Login count and data usage in the last 30 days...

[0106] In one possible implementation, consider the prediction of the answer to the question of whether someone is satisfied with a new activity or product:

[0107] Extract data such as whether or not you participated or purchased, the interval between your first participation and the launch, the frequency of use or participation, whether or not you participated or used recently, and the time since your last use or participation as game feature data.

[0108] In another possible implementation, consider an open-ended suggestion question (e.g., "Please suggest something for this activity"):

[0109] Game data such as player participation frequency, player winning percentage, recent participation status, churn time, and return status are extracted as game feature data.

[0110] Step 42: Train a machine learning algorithm model based on the game feature data of at least one user to obtain the first model.

[0111] In this step, the first model is a designated model for a certain question. Its function is to classify the output results of the question using game feature data from multiple users. When used later, by inputting the game feature data of a certain user, the specified prediction result can be obtained.

[0112] Optionally, as an example, Table 5 shows an example of the output results of the game feature data after training, as shown in Table 5 (the topic is the frequency of purchasing bottled water):

[0113] Table 5:

[0114]

[0115]

[0116] Among them, the trained features can be used as the criteria for judging the classification results.

[0117] Optionally, during the generation of the first model, the game feature data of at least one user can be used to classify the game feature data using a classification algorithm. Each classification interval corresponds to a different classification result. After inputting the game feature data of a certain user, the classification result, i.e. the prediction result, can be obtained.

[0118] In one possible implementation, game logs can provide a wealth of player behavior data. Big data engineers can select different data and algorithms to try, train different models, and then evaluate the models to obtain the one with the most accurate prediction, which is then used as the first model.

[0119] Specifically, behavioral data is used as game feature data. The process involves building training and testing sets, training models using different algorithms, testing on the testing set to see the results, adjusting features, obtaining the model, testing again, and repeating this process until a relatively good model is obtained.

[0120] Step 43: Input the game feature data of the survey users corresponding to the question into the first model to obtain the prediction result of the question.

[0121] In this step, in order to compare the prediction results with the response results, the game feature data of the survey users corresponding to the question needs to be input into the first model. The trained features can be compared with the game feature data to obtain the prediction result of the question.

[0122] Optionally, as an example, Table 6 shows a sample of user game characteristic data and corresponding prediction results, as shown in Table 6 (it should be understood that the users in Table 6 are users who completed the questionnaire, and some prediction results for different questions):

[0123] Table 6:

[0124]

[0125] It should be understood that, for a given question, the first model is trained using game feature data from at least one user corresponding to that question. This model is able to determine what data features lead to what possible options.

[0126] For example, the question "What is your favorite outfit?" actually requires the first model to infer the current player's favorite outfit using the player's current behavior data.

[0127] Therefore, a first model is trained using data from at least one user. This first model will produce a specified result when given specified features. The process of predicting the result involves finding the solution process corresponding to the question, extracting data from the survey users who completed the questionnaire, determining the feature data needed for the model, and inputting the survey users' feature data into the model to obtain the possible choices of the survey participants for this question (i.e., the predicted result).

[0128] In one possible implementation, let's take the prediction of the answer to the question of whether someone is satisfied with a new activity or product as an example:

[0129] The predicted results for this question can be ranked and represented as: very satisfied (60%), moderately satisfied (20%), average (10%), and dissatisfied (10%).

[0130] In another possible implementation, consider the above suggestion-type open-ended question (e.g., please suggest something for this activity):

[0131] The prediction results for this question can be sorted and represented as either positive / negative or positive / negative / neutral.

[0132] Example 1: Positive can mean: to have a suggestion; neutral can mean: to have no suggestion; negative can mean: to give a random answer.

[0133] Example 2: Positive can mean positive advice; neutral can mean fair advice; negative can mean negative advice.

[0134] The technical solution provided in this application involves acquiring game feature data of at least one user corresponding to a question, training a machine learning algorithm model based on the game feature data of at least one user to obtain a first model, and then inputting the game feature data of the surveyed user corresponding to the question into the first model to obtain the prediction result of the question. In this technical solution, by using the game feature data of multiple users to train the model, the accuracy of the model when making predictions using the game feature data of the surveyed user is increased, and a basis is also provided for judging the subsequent valid response results.

[0135] Based on the above embodiments, the questionnaire may include one or more of the following types of questions:

[0136] 1) Single choice;

[0137] Specifically, a question can have only one answer.

[0138] 2) Multiple choice;

[0139] Specifically, a question can have at least two options.

[0140] 3) Open-ended.

[0141] Specifically, the answer to a question is to be completed by the user; there are no options.

[0142] Furthermore, Figure 5 This is a flowchart illustrating Embodiment 4 of the questionnaire processing method provided in this application. Figure 5 As shown, if the response result in step 23 above meets the preset expectation of the predicted result, the response result is determined to be valid, and the valid response result is retained.

[0143] In this step, since the prediction results are obtained by using a model trained with the game feature data of users with the user tag, it can reflect the behavior of users corresponding to this type of user tag to a greater extent, which can improve the validity of the response results. At this time, the response results are compared with the prediction results to determine whether the response results fall within the preset expectation of the prediction results, thus confirming the validity of the response results of this questionnaire. Then, the corresponding service of the questionnaire can be improved and perfected based on the response results.

[0144] Optionally, the preset expectation can be the option with the highest probability in the prediction results.

[0145] In one possible implementation, the predicted results could be 70% (low button sensitivity setting), 20% (moderate button sensitivity setting), 5% (high button sensitivity setting), and 5% (too high button sensitivity setting). The preset expectation for the predicted results is the first two items, that is, if the response result is low button sensitivity setting or moderate button sensitivity setting, then the response result of the questionnaire is considered valid. If the response result is one of the latter items, it is necessary to further determine whether the response result is usable, which will be given in subsequent embodiments.

[0146] The following possible implementations are not in any particular order, and the triggering may differ depending on the type of question:

[0147] The answer type for question 1 is single choice:

[0148] For example, compare whether the response results match the set of highest probabilities in the predicted results (that is, the expected number of the top X options with the highest probabilities, where X is a positive integer greater than or equal to 1).

[0149] Optionally, as an example, Table 7 shows the ranking of prediction results obtained after different questions are input into the first model and the corresponding preset expectations, as shown in Table 7 (taking X as the first two options as an example):

[0150] Table 7:

[0151]

[0152] That is, if the predicted results are 70% (very satisfied), 20% (satisfied), 5% (neutral), and 5% (dissatisfied), then if either of the first two predicted results matches the actual response, the response to the questionnaire is considered valid, and the valid response is retained.

[0153] The second question's answer type is multiple choice:

[0154] For example, compare whether the response results do not match the set of probabilities that are lower in the predicted results (that is, the preset expectation does not include the Y options with lower probabilities, where Y is a positive integer greater than or equal to 1).

[0155] Optionally, as an example, Table 8 shows a second example of the ranking of prediction results obtained after different questions are input into the first model and the corresponding preset expectations, as shown in Table 8 (taking Y as the last option as an example):

[0156] Table 8:

[0157]

[0158] That is, the predicted results are 70% (fun), 15% (satisfied), 10% (would recommend to others), and 5% (not fun). If the last Y (e.g., 1) options of the predicted results do not overlap with any of the options in the response results, it means that the response results of the questionnaire are valid, and the valid response results are retained.

[0159] Third, the type of answer to the question is open-ended:

[0160] For example, the response result is first segmented using a sentence segmentation algorithm (such as jieba segmentation) to obtain the segmentation result. Then, a pre-defined custom stop word list is used to remove stop words from the segmentation result to obtain the target result. The target result is then input into a model (i.e., the second model) built by a machine learning algorithm (such as an open-source version control system (subversion, SVN)) to obtain the user's answer sentiment.

[0161] Optionally, the target result can be input into a second model to obtain the sentiment of the answer. This second model is used to determine the sentiment expressed by the survey users in the open-ended questions. If the sentiment of the answer is consistent with the sentiment expressed in the predicted result, the response result is determined to be valid.

[0162] Furthermore, we compare whether the sentiment of the answers aligns with the pre-set expectations of the predicted results.

[0163] In one possible implementation, the predicted outcome can be positive sentiment. If the processed responses also show positive sentiment, then the questionnaire responses are considered valid.

[0164] The technical solution provided in this application, by classifying the question types in the questionnaire into single-choice, multiple-choice, and / or open-ended questions, sets different judgment methods for different question types, thereby more accurately determining whether the survey user's response is a valid response.

[0165] Based on the above embodiments, Figure 6 This is a flowchart illustrating Embodiment 5 of the questionnaire processing method provided in this application. Figure 6 As shown, this paper explains how to determine the validity of the questionnaire response results if the response results do not meet the preset expectations of the predicted results in step 23 above.

[0166] Optionally, if the response does not meet the preset expectations of the predicted result, the following steps can be performed:

[0167] Step 61: Obtain the number of times the question corresponding to the response result has been answered in the questionnaire.

[0168] In this step, to prevent survey users from answering questions randomly (e.g., filling in answers randomly in order to receive rewards from the system), when it is determined that the response results do not meet the preset expectations of the predicted results, the question can be added back to the questionnaire so that survey users can respond to the question again, in order to consider whether the survey users answered questions randomly.

[0169] At this point, the number of times the question corresponding to the expected response that does not conform to the prediction result has been answered is obtained.

[0170] In one possible implementation, the number of times the question corresponding to the obtained response has been answered in the questionnaire is 1.

[0171] Step 62: If the number of times is less than or equal to the number of times threshold, the order of the options in the question corresponding to the response result that does not meet the preset expectation of the prediction result is readjusted.

[0172] In this step, to avoid the same question appearing too many times in the questionnaire and causing emotional fluctuations among users, a frequency threshold is set to limit the number of times the same question appears in the questionnaire.

[0173] Furthermore, when the number of responses is less than or equal to the threshold, the order of the options in the questions will be readjusted to prevent survey users from filling in answers randomly and to improve the survey user experience.

[0174] Optionally, as an example, Table 9 illustrates the number of times a question has appeared and the threshold for that number of appearances, as shown in Table 9:

[0175] Table 9:

[0176] Question Number Number of times answered Number of times threshold Do you want to adjust the order of the options? Z 1 3 yes Q 4 3 no

[0177] That is, the number of times threshold can be 3. Taking the above steps as an example, where 1 is less than 3, the order of the options in the question corresponding to the answer result is adjusted.

[0178] For example, if the options for a question are "Satisfied, Neutral, Unsatisfied", then it should be changed to "Unsatisfied, Satisfied, Neutral".

[0179] Step 63: Add the questions with the adjusted option order to the list of questions to be sent in the survey questionnaire so that the survey users can respond again.

[0180] In this step, the questions with the adjusted selection order are added to the list of questions to be sent out in the questionnaire, so that they can be sent to the survey users again and they can answer the questions again, so as to make effective judgment on the response results of the questions in the future.

[0181] The technical solution provided in this application involves obtaining the number of times a question corresponding to a response result has been answered in a questionnaire. If the number of answers is less than or equal to a threshold, the order of the options in the questions corresponding to responses that do not meet the preset expectations is readjusted. The questions with the readjusted options are then added to the questionnaire's delivery list so that survey users can respond again. In this technical solution, adjusting the order of options for questions corresponding to responses whose validity is uncertain and then redeploying them to users provides a reference for more accurately determining the validity of the responses.

[0182] Based on the above embodiments, Figure 7 This is a flowchart illustrating Embodiment Six of the questionnaire processing method provided in this application. Figure 7 As shown, after step 61, the method for processing the questionnaire may further include the following steps:

[0183] In this embodiment, when the number of responses exceeds a threshold, the validity of the responses is determined based on all responses to the questions that do not meet the preset expected response results. In this way, valid responses are retained in the questionnaire or invalid responses are removed.

[0184] Step 71: When the number of times exceeds the threshold, convert the options of the questions corresponding to the answers that do not meet the preset expectation of the prediction results into equally spaced sequence numbers to obtain the numbers corresponding to different options.

[0185] In this step, to ensure the reliability of the answers to the same question, the options corresponding to the question are converted into equally spaced sequence numbers to obtain the numbers corresponding to different options.

[0186] In one possible implementation, the options for question Z could be U, V, or W.

[0187] In one possible implementation, the options for question Q could be U, V, W, or E.

[0188] Optionally, as an example, Table 10 shows examples of all the response results for different questions, as shown in Table 10:

[0189] Table 10:

[0190] Question Number The options for this question Converted numbers Z U, V, W 1、2、3 Q U, V, W, E 1、2、3、4

[0191] That is, for question Z, option U corresponds to 1, option V corresponds to 2, and option W corresponds to 3; for question Q, option U corresponds to 1, option V corresponds to 2, option W corresponds to 3, and option E corresponds to 4.

[0192] Step 72: Determine the variance of the numbers corresponding to the options indicated by all the response results.

[0193] In this step, for a given question, if the number of times the answer has been given exceeds a threshold, the numbers corresponding to the options indicated by the answers to the given question are counted, and the variance of these numbers is calculated.

[0194] Variance is used to measure the degree of deviation between each number and the mean of all numbers, that is, the degree of variation of multiple answers to the same question. Based on the above multiple numbers, the variance of multiple numbers is obtained.

[0195] In one possible implementation, the numbers corresponding to the three answered options for question Z are 1, 2, and 3, respectively, with a variance of approximately 0.67.

[0196] In one possible implementation, the numbers corresponding to the three options that have been answered in question Q are either 4, 4, 4 or 1, 1, 1, with a variance of 0.

[0197] Step 73: Based on the variance and tolerance threshold, determine whether the response results are valid, so as to retain the valid response results in the questionnaire or remove the invalid response results in the questionnaire.

[0198] In this step, if the deviation is greater than the tolerance threshold, all responses that do not meet the preset expectation of the prediction result are invalid; if the deviation is not greater than the tolerance threshold, all responses that do not meet the preset expectation of the prediction result are valid.

[0199] In one possible implementation, the tolerance threshold is set to 0.1. The variance of the numbers corresponding to the options indicated by all responses to question Q is 0. Since 0 is less than 0.1, the surveyed user has expressed their true thoughts and is marked as a valid response, so that the valid response is retained.

[0200] In one possible implementation, the tolerance threshold is set to 0.1. The variance of all responses to question Z is 0.67. Since 0.67 is greater than 0.1, the surveyed user did not express their true thoughts and is marked as an invalid response, which is then removed.

[0201] The technical solution provided in this application converts the options of questions corresponding to responses that do not meet the preset expected results into equally spaced numerical sequences when the number of responses exceeds a threshold. This yields numbers corresponding to different options. Based on the numbers corresponding to the options indicated by all responses, the variance of these numbers is determined. Using the variance and a tolerance threshold, the validity of the responses is determined, thus retaining valid responses or removing invalid ones. This technical solution utilizes the degree of variation among all responses to the same question to determine the user's true thoughts on the same question, more accurately determining the validity of the responses. It automatically removes dirty data that affects the survey results, effectively improving the validity of the questionnaire and significantly increasing the efficiency of questionnaire collection.

[0202] Based on the above embodiments, Figure 8 This is a flowchart illustrating Embodiment Seven of the questionnaire processing method provided in this application. An example of the overall process is given below. Figure 8 As shown, the processing method for this questionnaire includes the following steps:

[0203] Step 1, Begin;

[0204] Step 2: Extract user tags from the user profile database;

[0205] Step 3: Determine if the user tag matches the questionnaire tag. If yes, proceed to step 4; otherwise, proceed to step 17.

[0206] Step 4: Obtain the model corresponding to the label;

[0207] Step 5: Extract game feature data from the database within the time interval during which the surveyed users most recently satisfied their picks;

[0208] Step 6: Input the game feature data into the model to obtain the prediction results;

[0209] The first model is trained based on the game feature data of at least one player.

[0210] Step 7: Distribute the questions to the survey users using certain rules;

[0211] Step 8: Wait for the questionnaire to be completed (check if you are done and click to turn the page);

[0212] Step 9: Compare the prediction results with the response results to see if there is any discrepancy. If yes, proceed to step 10; otherwise, proceed to step 12.

[0213] Step 10: Determine if the number of times the question appears has not exceeded the threshold. If yes, proceed to step 11; otherwise, proceed to step 12.

[0214] Step 11: Change the number of options for the question and release the question again;

[0215] Step 12: Wait for the above process to finish, then collect all questions and their corresponding answers;

[0216] Step 13: Compare the biased responses and determine the variance of the numbers corresponding to the biased responses;

[0217] Step 14: Determine if the variance is less than the tolerance threshold. If yes, proceed to step 15; otherwise, proceed to step 16.

[0218] Step 15: Mark as a valid response and proceed to Step 17;

[0219] Step 16: Mark the response as invalid;

[0220] Step 17, End.

[0221] The technical solutions provided in this application have the same beneficial effects as those described in the above embodiments.

[0222] Based on the above method embodiments, Figure 9 A schematic diagram of a questionnaire processing device provided in an embodiment of this application is shown below. Figure 9 As shown, the questionnaire processing device includes: an acquisition module 91, a determination module 92, and a processing module 93.

[0223] The acquisition module 91 is used to acquire the responses of survey users to each question in the questionnaire, which includes at least one question.

[0224] The determination module 92 is used to input the game feature data of the survey users corresponding to each question of the questionnaire into the pre-generated first model to obtain the prediction results of each question in the questionnaire. The first model is trained based on the game feature data of at least one user corresponding to each question.

[0225] Processing module 93 is used to determine whether the survey users' responses to each question are valid based on the predicted results of each question in the questionnaire, so as to retain the valid responses in the questionnaire or remove the invalid responses in the questionnaire.

[0226] In one possible design of this application embodiment, the processing module 93 is specifically used for:

[0227] If the response meets the preset expectation of the predicted result, the response is determined to be valid and the valid response is retained;

[0228] or,

[0229] If the response does not meet the expected prediction, obtain the number of times the corresponding question has been answered in the questionnaire.

[0230] If the number of responses exceeds the threshold, the validity of the responses is determined based on all responses to the questions that do not meet the predicted results. This allows for the retention of valid responses or the removal of invalid responses from the questionnaire.

[0231] In this possible design, processing module 93 determines the validity of all responses to questions corresponding to responses that do not meet the predicted results, in order to retain valid responses or remove invalid responses from the questionnaire. Specifically, it is used for:

[0232] The options for questions that do not meet the predicted results are converted into equally spaced sequence numbers to obtain the numbers corresponding to different options;

[0233] Based on the numbers corresponding to the options indicated by all the responses, determine the variance of the numbers corresponding to the options indicated by all the responses.

[0234] Based on variance and tolerance threshold, determine whether the response results are valid, in order to retain valid responses in the questionnaire or eliminate invalid responses.

[0235] Optionally, processing module 93 determines the validity of the response results based on variance and tolerance threshold, to retain valid responses in the questionnaire or remove invalid responses, specifically for:

[0236] If the variance is less than the tolerance threshold, all responses that do not meet the preset expectation of the prediction results are deemed valid, and valid responses are retained.

[0237] If the variance is greater than or equal to the tolerance threshold, all responses that do not meet the preset expectations of the prediction results are deemed invalid and are removed.

[0238] Optionally, processing module 93 is also used for:

[0239] If the number of times is less than or equal to the number of times threshold, the order of the options in the question corresponding to the response result that does not meet the preset expectation of the prediction result will be readjusted;

[0240] The questions with the adjusted option order will be added to the list of questions to be sent out in the survey questionnaire so that survey users can respond again.

[0241] In another possible design of this application embodiment, the response results of the questions in the questionnaire are of the single-choice type, and the preset expectation is: to hit the top X sets of the prediction results in terms of probability, where X is an integer greater than 0;

[0242] Processing module 93 determines the validity of the questionnaire responses if the responses match the predicted results, and retains the valid responses. Specifically, it is used for:

[0243] If the response matches one of the top X sets in the probability ranking of the prediction results, the response is considered valid and is retained.

[0244] In another possible design of this application embodiment, the response results of the questions in the questionnaire are of the multiple-choice type, and the preset expectation is: not to hit the last Y sets in the probability ranking of the prediction results, where Y is an integer greater than 0;

[0245] Processing module 93 determines the validity of the questionnaire responses if the responses match the predicted results, and retains the valid responses. Specifically, it is used for:

[0246] If the response does not match the last Y sets of the predicted results in terms of probability, the response is considered valid and is retained.

[0247] In another possible design of this application embodiment, the responses to the questions in the questionnaire are open-ended, with the preset expectation that they are consistent with the sentiment expressed in the predicted results.

[0248] Processing module 93 determines the validity of the questionnaire responses if the responses match the predicted results, and retains the valid responses. Specifically, it is used for:

[0249] The answer to the question is segmented using sentence segmentation to obtain the segmentation result;

[0250] The target result is obtained by removing stop words from the word segmentation results using a pre-defined custom stop word list;

[0251] The target results are input into the second model to obtain the sentiment of the answers. The second model is used to determine the sentiment expressed by the survey users in open-ended questions.

[0252] If the sentiment of the answer is consistent with the sentiment expressed in the prediction result, the response is deemed valid and the valid response is retained.

[0253] In another possible design of this application embodiment, the processing module 93 is further configured to:

[0254] Based on user profiles, determine user tags for the surveyed users;

[0255] Extract questions from the question bank that correspond to the same user tags, and generate a survey questionnaire.

[0256] The push module is used to push survey questionnaires to survey users.

[0257] In another possible design of this application embodiment, the acquisition module 91 is further used to acquire game feature data of at least one user corresponding to the question;

[0258] The processing module 93 is also used to train a machine learning algorithm model based on the game feature data of at least one user to obtain a first model.

[0259] The questionnaire processing device provided in this application embodiment can be used to execute the technical solution corresponding to the questionnaire processing method in the above embodiment. Its implementation principle and technical effect are similar, and will not be described again here.

[0260] It should be noted that the division of the various modules in the above device is merely a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, these modules can be implemented entirely in software through processing element calls; they can be fully implemented in hardware; or some modules can be implemented through processing element calls in software, while others are implemented in hardware. Moreover, these modules can be fully or partially integrated together, or implemented independently. The processing element mentioned here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above modules can be completed through the integrated logic circuits in the hardware of the processor element or through software instructions.

[0261] Figure 10 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 10 As shown, the electronic device may include: a processor 101, a memory 102, and computer program instructions stored in the memory 102 and executable on the processor 101.

[0262] The processor 101 executes computer execution instructions stored in the memory 102, causing the processor 101 to perform the scheme in the above embodiments. The processor 101 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0263] The memory 102 is connected to the processor 101 via the system bus and completes communication between them. The memory 102 is used to store computer program instructions.

[0264] Optionally, the structure of the electronic device also includes a transceiver 103, which is connected to the processor 101 via a system bus and performs communication with it.

[0265] In implementation, the transceiver 103 can correspond to Figure 9 The embodiment shown includes the acquisition module 91 and the push module.

[0266] The system bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The system bus can be divided into address bus, data bus, control bus, etc. For ease of representation, only one thick line is used in the diagram, but this does not indicate that there is only one bus or one type of bus.

[0267] The electronic device provided in this application embodiment can be used to execute the technical solution corresponding to the questionnaire processing method in the above embodiment. Its implementation principle and technical effect are similar, and will not be described again here.

[0268] This application also provides a chip for executing instructions, which is used to execute the technical solution of the questionnaire processing method in the above embodiments.

[0269] This application also provides a computer-readable storage medium storing computer instructions. When the computer instructions are executed on an electronic device, the electronic device performs the technical solution of the questionnaire processing method described in the above embodiments.

[0270] The aforementioned computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to general-purpose or special-purpose electronic devices.

[0271] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

[0272] Although this application discloses preferred embodiments as described above, it is not intended to limit this application. Any person skilled in the art can make possible changes and modifications without departing from the spirit and scope of this application. Therefore, the scope of protection of this application should be determined by the scope defined in the claims of this application.

Claims

1. A method of processing a questionnaire, characterized by, The method comprises: obtaining a reply result of each question in a questionnaire of a survey user for a game application, the questionnaire comprising at least one question; inputting game feature data of the survey user corresponding to each question of the questionnaire into a first model generated in advance respectively to obtain a prediction result of each question in the questionnaire, the first model being trained according to game feature data of at least one user corresponding to each question respectively; the game feature data is used to reflect a behavior of the survey user in the game application; determining whether the reply result of each question in the questionnaire is valid according to the prediction result of each question in the questionnaire, to retain a valid reply result in the questionnaire or eliminate an invalid reply result in the questionnaire.

2. The method of claim 1, wherein, The determining whether the reply result of each question in the questionnaire is valid according to the prediction result of each question in the questionnaire, to retain a valid reply result in the questionnaire or eliminate an invalid reply result in the questionnaire, comprises: if the reply result meets a preset expectation of the prediction result, determining that the reply result is valid, and retaining the valid reply result; or, if the reply result does not meet the preset expectation of the prediction result, obtaining a number of times that the question corresponding to the reply result has been answered in the questionnaire; if the number of times is greater than a number threshold, determining whether the reply result is valid according to all reply results of the question corresponding to the reply result that does not meet the preset expectation of the prediction result, to retain a valid reply result in the questionnaire or eliminate an invalid reply result in the questionnaire.

3. The method of claim 2, wherein, The determining whether the reply result is valid according to all reply results of the question corresponding to the reply result that does not meet the preset expectation of the prediction result, to retain a valid reply result in the questionnaire or eliminate an invalid reply result in the questionnaire, comprises: converting options of the question corresponding to the reply result that does not meet the preset expectation of the prediction result into equal-interval sequence digital representations respectively to obtain numbers corresponding to different options; determining a variance of the numbers corresponding to the options pointed by all the reply results according to the numbers corresponding to the options pointed by all the reply results; determining whether the reply result is valid according to the variance and a tolerance threshold, to retain a valid reply result in the questionnaire or eliminate an invalid reply result in the questionnaire.

4. The method of claim 3, wherein, The determining whether the reply result is valid according to the variance and the tolerance threshold, to retain a valid reply result in the questionnaire or eliminate an invalid reply result in the questionnaire, comprises: if the variance is less than the tolerance threshold, determining that all the reply results that do not meet the preset expectation of the prediction result are valid, and retaining the valid reply results; if the variance is greater than or equal to the tolerance threshold, determining that all the reply results that do not meet the preset expectation of the prediction result are invalid, and eliminating the invalid reply results.

5. The method of claim 2, wherein, The method further comprises: if the number of times is less than or equal to the number threshold, re-adjusting an order of the options in the question corresponding to the reply result that does not meet the preset expectation of the prediction result. The question after the adjustment of the order of the adjustment options is supplemented to a list of to-be-delivered questions in the survey questionnaire, so that the survey user answers again.

6. The method according to any one of claims 2-5, characterized in that, The answer result of the question in the survey questionnaire is of a single selection type, and the preset expectation is that the answer result hits the top X sets in the probability size order of the prediction result, where X is an integer greater than 0. If the answer result meets the preset expectation of the prediction result, it is determined that the answer result of the survey questionnaire is valid, and the valid answer result is retained, including: If the answer result hits the top X sets in the probability size order of the prediction result, it is determined that the answer result is valid, and the valid answer result is retained.

7. The method according to any one of claims 2-5, characterized in that, The answer result of the question in the survey questionnaire is of a multiple selection type, and the preset expectation is that the answer result does not hit the last Y sets in the probability size order of the prediction result, where Y is an integer greater than 0. If the answer result meets the preset expectation of the prediction result, it is determined that the answer result of the survey questionnaire is valid, and the valid answer result is retained, including: If the answer result does not hit the last Y sets in the probability size order of the prediction result, it is determined that the answer result is valid, and the valid answer result is retained.

8. The method according to any one of claims 2-5, characterized in that, The answer result of the question in the survey questionnaire is of an open type, and the preset expectation is that the answer result is consistent with an expression emotion in the prediction result. If the answer result meets the preset expectation of the prediction result, it is determined that the answer result of the survey questionnaire is valid, and the valid answer result is retained, including: The answer result of the question is segmented by using sentence segmentation, to obtain a segmentation result. The stop words in the segmentation result are removed by using a preset self-defined stop word table, to obtain a target result. The target result is input into a second model to obtain an answer emotion, and the second model is used to determine the emotion expressed by the survey user in the open question. If the answer emotion is consistent with the expression emotion in the prediction result, it is determined that the answer result is valid, and the valid answer result is retained.

9. The method according to any one of claims 1 to 5, characterized in that, Before the answer result of each question in the survey questionnaire by the survey user is obtained, the method further includes: Based on the user portrait, a user tag of the survey user is determined. In the question bank, a question corresponding to a questionnaire tag consistent with the user tag is extracted, to generate the survey questionnaire. The survey questionnaire is pushed to the survey user.

10. The method according to any one of claims 1 to 5, characterized in that, Before it is determined whether the answer result of the user is valid according to the prediction result, the method further includes: Game feature data of at least one user corresponding to the question is obtained. A machine learning algorithm model is trained according to the game feature data of the at least one user, to obtain the first model.

11. A processing device for a questionnaire, characterized by The method includes: An obtaining module is configured to obtain an answer result of each question in a survey questionnaire of a survey user for a game application, the survey questionnaire including at least one question. The determining module is configured to input game feature data of the survey user corresponding to each question of the questionnaire into a first model generated in advance respectively to obtain a prediction result of each question in the questionnaire, and the first model is trained according to game feature data of at least one user corresponding to each question respectively; and the game feature data is used to reflect behavior of the survey user in the game application. The processing module is configured to determine whether a reply result of each question of the survey user is valid according to the prediction result of each question in the questionnaire, to retain a valid reply result in the questionnaire or to eliminate an invalid reply result in the questionnaire.

12. An electronic device, comprising: Comprise: A processor, a memory, and computer program instructions stored on the memory and executable on the processor; The processor executes the computer program instructions to implement the processing method of the questionnaire according to any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer execution instructions, and the computer execution instructions are executed by the processor to implement the processing method of the questionnaire according to any one of claims 1 to 10.

Images