Intelligent control method and device for questionnaire layout for survey services
The intelligent control method and device for questionnaire layout enhance user participation and data quality by personalizing survey questionnaires through customer clustering, element configuration, and interest-based question ranking, addressing the limitations of traditional unified layouts.
Patent Information
- Application Number
- CN202411311272.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-20
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-09-20
AI Technical Summary
Traditional user survey questionnaires often adopt a unified layout, leading to low user participation and failure to collect important information, especially in complex questionnaires with large numbers of questions, due to lack of personalization and relevance to users.
An intelligent control method and device for questionnaire layout that analyzes user interests by clustering customers based on historical data, configuring relevant questionnaire elements, matching questions, and ranking them according to interest levels to optimize the questionnaire layout.
Improves user enthusiasm for filling out questionnaires, increasing the probability of collecting important information by personalizing the questionnaire content and structure based on user interests.
Smart Images

Figure CN118822618B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of intelligent data processing technology, and in particular to an intelligent control method and device for questionnaire layout for survey services. Background Art
[0002] User survey questionnaires are key tools for market research and user feedback in product development and improvement. Traditional user survey questionnaires usually adopt a unified layout, that is, the same questionnaire is distributed to all participants. It works well when dealing with simple or small-scale surveys, but it has significant limitations when facing complex questionnaires with a large number of questions and options. The questionnaire design methods in the prior art often ignore the personalized needs and interests of users. When users are faced with a lengthy questionnaire that is irrelevant to their personal interests, they usually feel bored or overwhelmed, which increases the probability of giving up filling it out, which means that important information related to the user cannot be collected, which directly affects the integrity and accuracy of the survey data. In addition, the questionnaire design in the prior art lacks an effective incentive mechanism for user participation. Due to the lack of pertinence in the layout and content of the questionnaire, users are not very motivated to fill it out, which not only affects the quality of the data, but also leads to deviations in the survey results.
[0003] In summary, there are technical problems in the prior art that traditional user survey questionnaires usually adopt a unified layout, resulting in low user participation when facing complex questionnaires and failure to collect important information related to the users. Summary of the invention
[0004] The purpose of this application is to provide an intelligent control method and device for questionnaire layout for survey services, in order to solve the technical problems in the prior art that traditional user survey questionnaires usually adopt a unified layout, resulting in low user participation when faced with complex questionnaires and inability to collect important information related to the user.
[0005] In view of the above problems, the present application provides an intelligent control method and device for questionnaire layout for survey services.
[0006] In a first aspect, the present application provides an intelligent control method for questionnaire layout for survey services, which is implemented by an intelligent control device for questionnaire layout for survey services, wherein the intelligent control method for questionnaire layout for survey services includes: receiving a questionnaire generation request, wherein the questionnaire generation request includes a product subject to the questionnaire; according to the product subject to the questionnaire, retrieving product historical dumping data, wherein the product historical dumping data has customer portrait label record information; performing customer clustering according to the customer portrait label record information to obtain multiple customer groups; through an element matching channel, performing element configuration on the product subject to the questionnaire to obtain a calibrated questionnaire element set; through a questionnaire question type library, performing question matching on the calibrated questionnaire element set to generate a questionnaire question set to be laid out, wherein the questionnaire question set to be laid out corresponds one-to-one to the calibrated questionnaire element set; traversing the multiple customer groups, performing interest analysis on the calibrated questionnaire element set, and obtaining multiple groups of interest analysis results; according to the multiple groups of interest analysis results, performing layout analysis on the questionnaire question set to be laid out respectively to obtain multiple question ranking results; and performing questionnaire layout control according to the multiple question ranking results.
[0007] In a second aspect, the present application further provides an intelligent control device for questionnaire layout for survey services, which is used to execute the intelligent control method for questionnaire layout for survey services as described in the first aspect, wherein the intelligent control device for questionnaire layout for survey services includes: a request receiving module, the request receiving module is used to receive a questionnaire generation request, wherein the questionnaire generation request includes a product subject to the questionnaire; a data acquisition module, the data acquisition module is used to retrieve product historical dumping data based on the product subject to the questionnaire, wherein the product historical dumping data has customer portrait label record information; a clustering analysis module, the clustering analysis module is used to cluster customers based on the customer portrait label record information to obtain multiple customer clusters; an element configuration module, the element configuration module is used to, through an element matching channel, The questionnaire target product is configured with elements to obtain a calibrated questionnaire element set; a question matching module is used to match the calibrated questionnaire element set with questions through a questionnaire question type library to generate a questionnaire question set to be laid out, wherein the questionnaire question set to be laid out corresponds one-to-one to the calibrated questionnaire element set; an interest analysis module is used to traverse the multiple customer groups, perform interest analysis on the calibrated questionnaire element set, and obtain multiple groups of interest analysis results; a layout analysis module is used to perform layout analysis on the questionnaire question set to be laid out according to the multiple groups of interest analysis results, and obtain multiple question ranking results; a layout control module is used to control the questionnaire layout according to the multiple question ranking results.
[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0009] By receiving a questionnaire generation request, wherein the questionnaire generation request includes a product of the questionnaire target; according to the product of the questionnaire target, retrieving product historical dumping data, wherein the product historical dumping data has customer portrait tag record information; performing customer clustering according to the customer portrait tag record information to obtain multiple customer groups; through the element matching channel, performing element configuration on the product of the questionnaire target to obtain a calibrated questionnaire element set; through the questionnaire question type library, performing question matching on the calibrated questionnaire element set to generate a questionnaire question set to be laid out, wherein the questionnaire question set to be laid out corresponds to the calibrated questionnaire element set one by one; traversing the multiple customer groups, performing interest analysis on the calibrated questionnaire element set to obtain multiple groups of interest degree analysis results; according to the multiple groups of interest degree analysis results, performing layout analysis on the questionnaire question set to be laid out respectively to obtain multiple question ranking results; and performing questionnaire layout control according to the multiple question ranking results. That is, by analyzing the user's interest and ranking the elements with higher interest in front, the user's enthusiasm for filling out is improved, thereby increasing the probability of collecting important information related to the user.
[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.
[0012] Figure 1 A flowchart of an intelligent control method for questionnaire layout for survey services of this application;
[0013] Figure 2 A schematic diagram of the structure of an intelligent control device for the layout of questionnaires used in the survey service of this application.
[0014] Explanation of the accompanying drawings: request receiving module 11, data acquisition module 12, clustering analysis module 13, element configuration module 14, topic matching module 15, interest analysis module 16, layout analysis module 17, layout control module 18. DETAILED DESCRIPTION
[0015] This application solves the technical problem in the prior art that traditional user survey questionnaires usually adopt a unified layout, resulting in low user participation in complex questionnaires and failure to collect important information related to users by providing an intelligent control method and device for questionnaire layout for survey services. By analyzing user interest and ranking elements with higher interest first, the user's enthusiasm for filling out the questionnaire is increased, thereby increasing the probability of collecting important information related to users.
[0016] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.
[0017] For example, please refer to the attached Figure 1 The present application provides an intelligent control method for questionnaire layout for survey services, wherein the intelligent control method for questionnaire layout for survey services is applied to an intelligent control device for questionnaire layout for survey services, and the intelligent control method for questionnaire layout for survey services specifically includes the following steps:
[0018] Step 1: Receive a questionnaire generation request, wherein the questionnaire generation request includes a product subject to the questionnaire.
[0019] Specifically, a questionnaire generation request is initiated by an individual or organization that needs to conduct market research or data collection, indicating that a new questionnaire needs to be created. The questionnaire generation request contains the product of the questionnaire, that is, the main object of this survey or research, and collects feedback or data on the specific product targeted by the questionnaire. By receiving a questionnaire generation request and specifying the product targeted by the questionnaire, it is ensured that the questionnaire content is related to a specific product, thereby improving the pertinence and relevance of the questionnaire.
[0020] Step 2: According to the product subject to the questionnaire, retrieve the product historical dumping data, wherein the product historical dumping data has customer portrait tag record information.
[0021] Specifically, historical sales data related to the product in the questionnaire is retrieved and obtained, including the product's past sales records, market performance, customer feedback and other information. These data contain label information of customer portraits, that is, information that describes the characteristics of the customer group that purchases the product, including age, gender, purchasing habits, preferences, etc., which is used to construct customer characteristics. By analyzing the product's historical sales data and understanding the product's market performance and customer feedback, it is helpful to identify the characteristics and needs of the target customer group and provide market insights for questionnaire design.
[0022] Step three: Cluster customers according to the customer portrait label record information to obtain multiple customer groups.
[0023] Specifically, the customer portrait tag record information contains multiple different customer image tags, each tag represents a characteristic or behavior pattern of a customer group. Compare every two customer portrait tag record information to obtain the customer portrait deviation value. Use clustering algorithms to divide customer groups, and group customers with similar characteristics or behaviors. Through clustering analysis, divide customer groups into multiple clusters. Customers within each cluster are similar to some extent, while customers between different clusters are quite different. Through customer clustering, the customer market can be segmented more accurately, and customer groups with similar needs and behaviors can be identified.
[0024] Step 4: Through the element matching channel, the elements of the questionnaire target product are configured to obtain a calibration questionnaire element set.
[0025] Specifically, the element matching channel refers to a set of rules or algorithms for matching product features with questionnaire elements, including the mapping relationship between product attributes and questionnaire questions, and how to select appropriate questionnaire elements based on product categories. Obtain the product category and retrieve the historical questionnaire set, that is, the questionnaire set designed and used in the past for the same or similar product categories. Perform statistical analysis on the elements in the historical questionnaire data and calculate the frequency with which each element is triggered (i.e., used) in the historical questionnaires. Delete those questionnaire element attributes with too low a triggering frequency and retain the elements with a higher triggering frequency to ensure that the new questionnaire contains the elements that users are most concerned about and improve the relevance of the questionnaire content to the product category. Group the screened questionnaire elements into a set to determine the final set of calibrated questionnaire elements. Through the element matching channel, it can be ensured that the questionnaire elements are highly relevant to the target product of the questionnaire, thereby improving the relevance and effectiveness of the questionnaire.
[0026] Step 5: Match the marked questionnaire element set with questions through the questionnaire question type library to generate a set of questionnaire questions to be arranged, wherein the set of questionnaire questions to be arranged corresponds one-to-one to the marked questionnaire element set.
[0027] Specifically, the questionnaire question type library is a database containing various question types and questions, such as single-choice questions, multiple-choice questions, scale questions, open-ended questions, etc. Each question type has a series of possible questions and related attributes, such as difficulty, relevance, applicable scenarios, etc. Match the questions that best suit each set of calibrated questionnaire elements to ensure that the selected questions can effectively collect target information. Combine the matched questions into a set, that is, the set of questions to be laid out for the questionnaire, to ensure that each question corresponds one-to-one to the corresponding element in the set of calibrated questionnaire elements. Through question matching, the questionnaire questions are highly correlated with the questionnaire elements, thereby improving the relevance and effectiveness of the questionnaire.
[0028] Step 6: traverse the multiple customer groups, perform interest analysis on the calibration questionnaire element set, and obtain multiple groups of interest analysis results.
[0029] Specifically, the number of historical questionnaire questions with a completion rate lower than 100% in the historical questionnaire is collected, and the central trend analysis is performed on them. The minimum value in the concentrated number of questions is used as the threshold of the number of questions. It is determined whether the number of questions in the questionnaire to be laid out is greater than or equal to the threshold of the number of questions. If it is less than, it means that the number of questions in the questionnaire is small, and the conventional layout can be directly performed. If it is greater than or equal to, it means that the number of questions in the questionnaire is large, and it is necessary to traverse multiple customer groups and extract the first cluster of customer portrait labels. The first cluster of customer portrait labels and the set of calibrated questionnaire elements are sent to several influence scoring nodes, and the influence of the questionnaire elements on the target customer group is scored to obtain several groups of influence scoring results. According to the order of the set of calibrated questionnaire elements, the scores of each attribute in each group of influence scoring results are summed to obtain the first set of influence scoring summed results. The influence scoring results of all attributes are summed to obtain the overall sum of influence scores. The ratio of the influence score of each element to the overall sum of influence scores is calculated to obtain the first group of interest analysis results. Follow this step in sequence to obtain the interest analysis results of other groups. Through interest analysis, a set of interest analysis results is generated for each customer group, which together constitute multiple sets of interest analysis results, reflecting the group's interest in the questionnaire elements. By conducting interest analysis on different customer groups, questionnaires that better meet the characteristics and needs of each group can be designed to improve the relevance and attractiveness of the questionnaire.
[0030] Step 7: Based on the multiple groups of interest analysis results, respectively perform layout analysis on the set of questions to be laid out in the questionnaire to obtain multiple question ranking results.
[0031] Specifically, based on multiple groups of interest analysis results, the first group of interest analysis results is extracted. Based on the first group of interest analysis results, the set of questions to be laid out in the questionnaire is sorted according to the user's interest level, and the questions with high interest are placed at the front. Multiple question sorting results include question sorting for different customer groups. By extracting the first group of interest analysis results, the interests and concerns of the first cluster of customer groups are clarified. Layout analysis is performed in sequence to obtain a result set after the interest levels of multiple different customer groups are sorted for questionnaire elements. By sorting the interest analysis results from large to small, the questions that users are most interested in can be placed at the front of the questionnaire, thereby increasing the user's enthusiasm for filling out the questionnaire and the completion rate of the questionnaire.
[0032] Step 8: Control the layout of the questionnaire according to the sorting results of the multiple questions.
[0033] Specifically, extract the respective question ranking results from the interest analysis results of multiple customer groups, which reflects the interest of different customer groups in each question in the questionnaire. Analyze the question ranking results of all customer groups, find out the questions with high general interest, determine the global ranking strategy, and give priority to the questions of general interest. Create the layout of the questionnaire, including the order of questions, the presentation of questions, page layout, etc., based on the global ranking strategy and other considerations, such as user experience. Publish the questionnaire layout to respondents to ensure that the questionnaire can be accessed by different customer groups and optimized according to customer interest. Sort the questions according to the interest of different groups, optimize the structure of the questionnaire, and ensure that the layout of the questionnaire is both logical and attractive, while taking into account the user experience, so as to improve the completion rate and data quality of the questionnaire.
[0034] Further, step three of this application includes:
[0035] The customer portrait label record information includes N customer portrait label record information; performing pairwise deviation analysis on the N customer portrait label record information to obtain a number of customer portrait deviation values; performing cluster analysis based on the several customer portrait deviation values according to a portrait deviation value threshold to obtain the multiple cluster customer groups.
[0036] Specifically, the customer portrait label record information includes N customer portrait label record information, which records multiple different customer image labels, each label represents a characteristic or behavior pattern of a customer group. The N customer portrait labels are compared in pairs to find the differences or deviations between different labels. A suitable statistical method or distance metric is selected to calculate the deviation value between two customer portraits, such as Euclidean distance, Manhattan distance, cosine similarity, etc. The customer portrait deviation value represents the degree of difference between the two customer portrait labels, which can be quantitative or qualitative. The customer portrait labels are described and classified in detail to obtain a customer portrait label attribute set, including a customer portrait label type attribute set and a customer portrait label numerical attribute set. The first and second customer portrait label record information are extracted from the N customer portrait label record information. It is determined whether the type attributes of the two customer image label records are the same. If one is different, the customer portrait deviation value is configured to be greater than the portrait deviation value threshold; if both are the same, the statistical deviation set does not meet the ratio of the numerical attribute set, and this ratio is set as the customer portrait deviation value.
[0037] A portrait deviation value threshold is set in advance to determine whether the differences between customer characteristics are significant enough. When the customer portrait deviation value exceeds this threshold, it is considered to be significantly different, which affects the clustering results. Based on several customer portrait deviation values that exceed the threshold, a clustering algorithm is used to divide the customer group, and customers with similar characteristics or behaviors are grouped together. Cluster analysis is a data mining technology used to divide objects in a data set into multiple clusters so that objects in the same cluster are as similar as possible, while objects in different clusters are as different as possible. Through cluster analysis, the customer group is divided into multiple clusters. Customers in each cluster are similar to some extent, while customers in different clusters are quite different. Through pairwise deviation analysis, we can deeply understand the characteristics and needs of customers, accurately divide customer groups, and thus improve the relevance and effectiveness of the questionnaire.
[0038] Furthermore, the present application also includes the following steps:
[0039] Obtain a customer portrait label attribute set, wherein the customer portrait label attribute set includes a customer portrait label type attribute set and a customer portrait label numerical attribute set; configure a fault-tolerant deviation interval set for the customer portrait label numerical attribute set; extract the first customer portrait label record information and the second customer portrait label record information according to the N customer portrait label record information; when any type attribute of the customer portrait label type attribute set of the first customer portrait label record information and the second customer portrait label record information is different, configure the customer portrait deviation value to be greater than the portrait deviation value threshold; when the customer portrait label type attribute set of the first customer portrait label record information and the second customer portrait label record information are the same, count the ratio of the customer portrait label numerical attribute deviation set of the first customer portrait label record information and the second customer portrait label record information that does not satisfy the customer portrait label numerical attribute set, and set it as the customer portrait deviation value.
[0040] Specifically, the customer portrait label is described and classified in detail to obtain a customer portrait label attribute set, including various characteristics and attributes of the customer. The customer portrait label type attribute set refers to the type attribute that describes the customer characteristics, including gender, occupation, marital status, etc., which is usually discrete. The customer portrait label numerical attribute set refers to the attributes with numerical characteristics, including age, income, purchase frequency, etc. A set of intervals configured for each numerical attribute is used to define the acceptable error range of the attribute. For example, for an age attribute, the tolerance deviation interval may be ±5 years old. According to specific requirements or standards, the first and second customer portrait label record information are extracted from N customer portrait label record information and compared. When any type attribute of the first customer portrait label record information and the second customer portrait label record information is different, it means that the two customer groups have significant differences in this specific attribute, and a numerical value is obtained, namely the customer portrait deviation value. The customer portrait deviation value is configured to be greater than the portrait deviation value threshold, indicating that the two customer portraits belong to different clusters.
[0041] When the type attribute sets of the two customer portrait tag record information are the same, calculate the set of numerical attribute deviations of the first and second customer portrait tag record information. For each numerical attribute, calculate the deviation between the first and second customer portrait tag record information, and compare it with the fault tolerance deviation interval. Count the ratio of the numerical attribute deviation sets that do not meet the fault tolerance deviation interval, that is, the ratio of the number of attributes that do not meet the conditions to the total number of attributes, set the ratio as the customer portrait deviation value, quantify the difference in numerical attributes between the two customer groups, and reflect the difference between different customer characteristics. By configuring the fault tolerance deviation interval, the inaccuracy of the data can be tolerated to a certain extent, thereby improving the accuracy of data analysis. Quantifying the difference in numerical attributes between the two customer groups by counting the ratio helps to analyze customer characteristics more accurately.
[0042] Further, step 4 of this application includes:
[0043] Obtain the product category to which the product subject of the questionnaire belongs; according to the product category, retrieve a historical questionnaire set, wherein the historical questionnaire set includes a plurality of sets of recorded questionnaire elements; perform element trigger frequency statistics based on the plurality of sets of recorded questionnaire elements to obtain a questionnaire element attribute list and a questionnaire element trigger frequency list; based on the questionnaire element trigger frequency list, delete questionnaire element attributes whose questionnaire element trigger frequency is less than or equal to a trigger frequency threshold from the questionnaire element attribute list to obtain the calibrated questionnaire element set.
[0044] Specifically, the category of the product to which the questionnaire target belongs is determined by the product identification or name. According to the product category, the questionnaire data related to the same or similar products in the past are retrieved and called from the product or market environment. The historical questionnaire set includes multiple sets of records, each of which contains a set of questionnaire elements, including questions, options, layout and other information. Statistical analysis is performed on multiple sets of recorded questionnaire element sets to calculate the frequency of each element being triggered (i.e., used) in the historical questionnaire. The questionnaire element attribute list and the questionnaire element trigger frequency list are obtained. The questionnaire element attribute list refers to a detailed description of the questionnaire element, including the type of question, options, content, etc.; the questionnaire element trigger frequency list records the frequency of each questionnaire element appearing in the historical questionnaire, reflecting the importance or relevance of the element. The trigger frequency threshold is set in advance to determine whether the questionnaire element is frequently used. If the questionnaire element trigger frequency is less than or equal to the trigger frequency threshold, the questionnaire element attribute is deleted and the element with a higher trigger frequency is retained. After screening, the remaining questionnaire element attributes constitute the calibration questionnaire element set, which has a higher trigger frequency in the historical questionnaire and is more valuable for the current questionnaire design.
[0045] For example, assume that the product category is smartphones. The historical questionnaire set includes multiple sets of recorded questionnaire element sets: Questionnaire 1: appearance design, performance, price, operating system, camera function, battery life; Questionnaire 2: appearance design, performance, price, operating system, memory size, battery life; Questionnaire 3: appearance design, performance, price, operating system, camera function, screen size; Questionnaire 4: appearance design, performance, price, operating system, memory size, screen size; Questionnaire 5: appearance design, performance, price, operating system, camera function, battery life. The element triggering frequency statistics show: appearance design: 4 times; performance: 4 times; price: 4 times; operating system: 4 times; camera function: 3 times; memory size: 2 times; battery life: 3 times; screen size: 2 times. Assume that the threshold is 2, that is, when the triggering frequency of a questionnaire element is less than or equal to 2, it will be deleted from the questionnaire element attribute list. Therefore, the obtained calibration questionnaire element set includes appearance design, performance, price, operating system, camera function, and battery life. By using historical questionnaire data to optimize the questionnaire design, ensuring that the elements included in the questionnaire are relevant to the product category and have high importance, and retaining the elements with higher triggering frequency, it can be ensured that the new questionnaire contains the elements that users are most concerned about, thereby improving the relevance of the questionnaire content to the product category.
[0046] Furthermore, step six of this application includes:
[0047] Collect a set of historical questionnaire question numbers with a completion rate less than 1; perform a central trend analysis on the set of historical questionnaire question numbers to obtain a concentrated question number set; extract the minimum value of the concentrated question number set and set it as a question number threshold; determine whether the number of questionnaire question sets to be arranged is greater than or equal to the question number threshold; if greater than or equal to, traverse the multiple customer clusters, perform interest analysis with the calibrated questionnaire element set, and obtain multiple groups of interest analysis results; if less than, directly perform conventional layout according to the questionnaire question set to be arranged.
[0048] Specifically, questionnaires with a completion degree less than 1 (i.e., not fully filled) are screened out from the historical questionnaire database, and the number of questions in the historical questionnaires with a low completion degree is recorded to form a set of historical questionnaire question numbers. The set of historical questionnaire question numbers is analyzed using statistical methods to determine the central tendency of the questions in the historical questionnaires. Central tendency analysis usually includes calculating statistics such as the mean and median. Through central tendency analysis, a set reflecting the central tendency of historical questionnaire questions is generated, including statistical results such as the average number of questions and the median number of questions. The minimum value is extracted from the set of concentrated question numbers as the lower limit of the number of questions in the historical questionnaire, i.e., the threshold of the number of questions. The number of the set of questions to be laid out in the questionnaire is compared with the threshold of the number of questions. If the number of the set of questions to be laid out in the questionnaire is greater than or equal to the threshold, the multi-cluster customer groups obtained by cluster analysis are traversed, and interest analysis is performed with the set of calibrated questionnaire elements to determine the interest of the group in the questionnaire elements. A set of interest analysis results is generated for each customer group through interest analysis, reflecting the interest of the group in the questionnaire elements. If it is less than the threshold, conventional layout is performed directly according to the set of questions to be laid out in the questionnaire, which means that the questionnaire length is short and complex interest analysis is not required, thus avoiding unnecessary complex analysis.
[0049] In a specific example, by analyzing historical questionnaires, historical questionnaires with a completion rate less than 1 are collected to obtain: Questionnaire 1: the total number of questions is 20, the number of questions completed by user 1 is 15, the number of questions completed by user 2 is 15, and the number of questions completed by user 3 is 18; Questionnaire 2: the total number of questions is 15, the number of questions completed by user 4 is 10, the number of questions completed by user 5 is 14, and the number of questions completed by user 6 is 12; Questionnaire 3: the total number of questions is 30, the number of questions completed by user 7 is 25, the number of questions completed by user 8 is 26, and the number of questions completed by user 9 is 24. The central tendency analysis of the number of questions in the historical questionnaires is performed: Questionnaire 1: the average number of questions completed is 16, and the median number of questions completed is 15; Questionnaire 2: the average number of questions completed is 12, and the median number of questions completed is 12; Questionnaire 3: the average number of questions completed is 25, and the median number of questions completed is 25. It can be seen that the minimum value of the concentrated number of questions is 12, so 12 is used as the question number threshold. Assume that the number of questions to be laid out in the questionnaire is 30, which is greater than 12. Conventional layout cannot be performed. It is necessary to traverse multiple customer groups and perform interest analysis with the calibration questionnaire element set to obtain multiple sets of interest analysis results. By analyzing historical data and customer interests, the questionnaire design can be optimized to improve the quality and effectiveness of the questionnaire. When the number of questionnaire questions is large, interest analysis can help increase the attractiveness of the questionnaire, thereby increasing the user's enthusiasm for filling out the questionnaire and the completion rate of the questionnaire.
[0050] Furthermore, the present application also includes the following steps:
[0051] According to the multiple clusters of customer groups, extract a first cluster of customer portrait labels; send the first cluster of customer portrait labels and the calibration questionnaire element set to several influence scoring nodes to obtain several groups of influence scoring results, and the information of any two of the several influence scoring nodes is not interoperable; according to the calibration questionnaire element set, add the influence scoring results of each attribute of the several groups of influence scoring results to obtain a first influence scoring sum result set; add the several groups of influence scoring results to obtain an overall influence scoring sum result; traverse the first influence scoring sum result set, calculate the ratio with the overall influence scoring sum result, obtain a first group of interest analysis results, and add them into the multiple groups of interest analysis results.
[0052] Specifically, the first cluster of customers is selected from multiple customer groups as the analysis object, and the portrait labels of the first cluster of customers are extracted, including age, gender, purchasing habits, etc., which reflect the basic characteristics and preferences of customers. The first cluster of customer portrait labels and the calibration questionnaire element set are sent to several influence scoring nodes, which are responsible for scoring the influence of questionnaire elements on the target customer group. Each influence scoring node calculates the influence score of each element on the first cluster of customers according to its internal algorithm, and obtains several groups of influence scoring results, which reflect the influence of the questionnaire elements on the target customer group. The information between these influence scoring nodes is not interoperable, which ensures the independence and objectivity of the scoring results. According to the order of the calibration questionnaire element set, the score of each attribute in each group of influence scoring results is added up to obtain the first influence score sum result set, which reflects the overall influence of the questionnaire elements on the target customer group.
[0053] Sum up all attribute scores in the first influence score sum result set to obtain a comprehensive influence score sum result. Traverse the first influence score sum result set and calculate the ratio of the influence score of each element to the overall influence score sum result. By calculating the ratio, the first set of interest analysis results is obtained, which reflects the interest level of the first cluster of customer groups in each questionnaire element. The first set of interest analysis results is added to multiple sets of interest analysis results, each of which contains an assessment of the interest of different customer groups in the questionnaire elements. By extracting the first cluster of customer portrait labels and performing influence scoring, a questionnaire that better meets the characteristics and needs of the target customer group is designed to improve the relevance and attractiveness of the questionnaire.
[0054] Further, step seven of this application includes:
[0055] According to the multiple groups of interest analysis results, a first group of interest analysis results is extracted; according to the first group of interest analysis results, the set of questions to be arranged in the questionnaire is sorted from large to small to obtain a first question sorting result, which is added to the multiple question sorting results.
[0056] Specifically, select the first group from multiple groups of interest analysis results as the basis for sorting, and sort the set of questions to be laid out in the questionnaire. Select a sorting algorithm, such as bubble sort, quick sort, merge sort, etc., to ensure that the sorting algorithm sorts the questions from large to small according to the first group of interest analysis results. According to the customer's interest in the questions, put the questions with high interest in front. Through sorting, get the sorting result of the first question, add the sorting result of the first question to the multiple question sorting results, and the multiple question sorting results include the sorting of questions for different customer groups. Sort from large to small according to the interest analysis results, and put the questions that users are most interested in at the front of the questionnaire to improve the user's enthusiasm for filling out the questionnaire and the completion rate of the questionnaire.
[0057] In summary, the intelligent control method for questionnaire layout for survey services provided in this application has the following technical effects:
[0058] By receiving a questionnaire generation request, wherein the questionnaire generation request includes a product of the questionnaire target; according to the product of the questionnaire target, retrieving product historical dumping data, wherein the product historical dumping data has customer portrait tag record information; performing customer clustering according to the customer portrait tag record information to obtain multiple customer groups; through the element matching channel, performing element configuration on the product of the questionnaire target to obtain a calibrated questionnaire element set; through the questionnaire question type library, performing question matching on the calibrated questionnaire element set to generate a questionnaire question set to be laid out, wherein the questionnaire question set to be laid out corresponds to the calibrated questionnaire element set one by one; traversing the multiple customer groups, performing interest analysis on the calibrated questionnaire element set to obtain multiple groups of interest degree analysis results; according to the multiple groups of interest degree analysis results, performing layout analysis on the questionnaire question set to be laid out respectively to obtain multiple question ranking results; and performing questionnaire layout control according to the multiple question ranking results. That is, by analyzing the user's interest and ranking the elements with higher interest in front, the user's enthusiasm for filling out is improved, thereby increasing the probability of collecting important information related to the user.
[0059] Embodiment 2: Based on the same inventive concept as the intelligent control method for questionnaire layout for survey service in the above embodiment, the present application also provides an intelligent control device for questionnaire layout for survey service, please refer to the attached Figure 2 , the intelligent control device for questionnaire layout for survey service includes:
[0060] The request receiving module 11 is used to receive a questionnaire generation request, wherein the questionnaire generation request includes a product subject of the questionnaire.
[0061] The data acquisition module 12 is used to retrieve the product historical sales data according to the product of the questionnaire target, wherein the product historical sales data has customer portrait tag record information.
[0062] The cluster analysis module 13 is used to cluster customers according to the customer portrait tag record information to obtain multiple customer groups.
[0063] The element configuration module 14 is used to configure the elements of the questionnaire target product through an element matching channel to obtain a calibration questionnaire element set.
[0064] The topic matching module 15 is used to match the topic of the calibrated questionnaire element set through the questionnaire question type library to generate a set of questionnaire topics to be arranged, wherein the set of questionnaire topics to be arranged corresponds one-to-one to the set of calibrated questionnaire elements.
[0065] The interest analysis module 16 is used to traverse the multiple customer groups, perform interest analysis on the calibration questionnaire element set, and obtain multiple groups of interest analysis results.
[0066] The layout analysis module 17 is used to perform layout analysis on the questionnaire question sets to be laid out according to the multiple groups of interest analysis results, and obtain multiple question ranking results.
[0067] The layout control module 18 is used to control the layout of the questionnaire according to the sorting results of the multiple questions.
[0068] Furthermore, the cluster analysis module 13 in the intelligent control device for questionnaire layout for survey service is also used for:
[0069] The customer portrait label record information includes N customer portrait label record information; performing pairwise deviation analysis on the N customer portrait label record information to obtain a number of customer portrait deviation values; performing cluster analysis based on the several customer portrait deviation values according to a portrait deviation value threshold to obtain the multiple cluster customer groups.
[0070] Furthermore, the cluster analysis module 13 in the intelligent control device for questionnaire layout for survey service is also used for:
[0071] Obtain a customer portrait label attribute set, wherein the customer portrait label attribute set includes a customer portrait label type attribute set and a customer portrait label numerical attribute set; configure a fault-tolerant deviation interval set for the customer portrait label numerical attribute set; extract the first customer portrait label record information and the second customer portrait label record information according to the N customer portrait label record information; when any type attribute of the customer portrait label type attribute set of the first customer portrait label record information and the second customer portrait label record information is different, configure the customer portrait deviation value to be greater than the portrait deviation value threshold; when the customer portrait label type attribute set of the first customer portrait label record information and the second customer portrait label record information are the same, count the ratio of the customer portrait label numerical attribute deviation set of the first customer portrait label record information and the second customer portrait label record information that does not satisfy the customer portrait label numerical attribute set, and set it as the customer portrait deviation value.
[0072] Furthermore, the element configuration module 14 in the intelligent control device for questionnaire layout for survey service is also used for:
[0073] Obtain the product category to which the product subject of the questionnaire belongs; according to the product category, retrieve a historical questionnaire set, wherein the historical questionnaire set includes a plurality of sets of recorded questionnaire elements; perform element trigger frequency statistics based on the plurality of sets of recorded questionnaire elements to obtain a questionnaire element attribute list and a questionnaire element trigger frequency list; based on the questionnaire element trigger frequency list, delete questionnaire element attributes whose questionnaire element trigger frequency is less than or equal to a trigger frequency threshold from the questionnaire element attribute list to obtain the calibrated questionnaire element set.
[0074] Furthermore, the interest analysis module 16 in the intelligent control device for questionnaire layout for survey service is also used for:
[0075] Collect a set of historical questionnaire question numbers with a completion rate less than 1; perform a central trend analysis on the set of historical questionnaire question numbers to obtain a concentrated question number set; extract the minimum value of the concentrated question number set and set it as a question number threshold; determine whether the number of questionnaire question sets to be arranged is greater than or equal to the question number threshold; if greater than or equal to, traverse the multiple customer clusters, perform interest analysis with the calibrated questionnaire element set, and obtain multiple groups of interest analysis results; if less than, directly perform conventional layout according to the questionnaire question set to be arranged.
[0076] Furthermore, the interest analysis module 16 in the intelligent control device for questionnaire layout for survey service is also used for:
[0077] According to the multiple clusters of customer groups, extract a first cluster of customer portrait labels; send the first cluster of customer portrait labels and the calibration questionnaire element set to several influence scoring nodes to obtain several groups of influence scoring results, and the information of any two of the several influence scoring nodes is not interoperable; according to the calibration questionnaire element set, add the influence scoring results of each attribute of the several groups of influence scoring results to obtain a first influence scoring sum result set; add the several groups of influence scoring results to obtain an overall influence scoring sum result; traverse the first influence scoring sum result set, calculate the ratio with the overall influence scoring sum result, obtain a first group of interest analysis results, and add them into the multiple groups of interest analysis results.
[0078] Furthermore, the layout analysis module 17 in the intelligent control device for questionnaire layout for survey service is also used for:
[0079] According to the multiple groups of interest analysis results, a first group of interest analysis results is extracted; according to the first group of interest analysis results, the set of questions to be arranged in the questionnaire is sorted from large to small to obtain a first question sorting result, which is added to the multiple question sorting results.
[0080] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The intelligent control method and specific examples of the questionnaire layout for survey services in the first embodiment are also applicable to the intelligent control device for the questionnaire layout for survey services in this embodiment. Through the detailed description of the intelligent control method for the questionnaire layout for survey services, those skilled in the art can clearly understand the intelligent control device for the questionnaire layout for survey services in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.
[0081] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
[0082] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.
Claims
1. An intelligent control method for questionnaire layout for survey services, characterized in that: include: Receiving a questionnaire generation request, wherein the questionnaire generation request includes a product subject to the questionnaire; According to the product subject to the questionnaire, retrieve the product historical dumping data, wherein the product historical dumping data has customer portrait tag record information; Clustering customers according to the customer portrait tag record information to obtain multiple customer groups; Through the element matching channel, the element configuration is performed on the target product of the questionnaire to obtain a calibration questionnaire element set; Through the questionnaire question type library, the calibrated questionnaire element set is matched with the questions to generate a set of questions to be arranged for the questionnaire, wherein the set of questions to be arranged for the questionnaire corresponds one to one with the calibrated questionnaire element set; Traversing the plurality of customer groups, performing interest analysis on the calibration questionnaire element set, and obtaining a plurality of interest degree analysis results; According to the multiple groups of interest analysis results, respectively perform layout analysis on the set of questions to be laid out in the questionnaire to obtain multiple question ranking results; Controlling the layout of the questionnaire according to the sorting results of the multiple questions; Among them, through the element matching channel, the element configuration of the questionnaire target product is performed to obtain a calibration questionnaire element set, including: Obtain the product category to which the product subject to the questionnaire belongs; Retrieving a historical questionnaire set according to the product category, wherein the historical questionnaire set includes a plurality of sets of recorded questionnaire element sets; Perform element trigger frequency statistics according to the plurality of sets of recorded questionnaire element sets to obtain a questionnaire element attribute list and a questionnaire element trigger frequency list; Based on the questionnaire element trigger frequency list, the questionnaire element attributes whose questionnaire element trigger frequency is less than or equal to the trigger frequency threshold are deleted from the questionnaire element attribute list to obtain the calibrated questionnaire element set; Traversing the multiple customer groups, performing interest analysis with the calibration questionnaire element set, and obtaining multiple groups of interest analysis results, the previous steps include: Collect the number of historical questionnaire questions with a completion rate less than 1; Performing a central tendency analysis on the set of historical questionnaire questions to obtain a set of concentrated question numbers; Extracting the minimum value of the set of questions in the set and setting it as the question quantity threshold; Determine whether the number of the set of questions to be arranged in the questionnaire is greater than or equal to the threshold number of questions; When it is greater than or equal to, traverse the multiple customer groups, perform interest analysis with the calibration questionnaire element set, and obtain multiple groups of interest analysis results; When it is less than, conventional layout is directly performed according to the set of questions to be laid out in the questionnaire.
2. The intelligent control method for questionnaire layout for survey service according to claim 1, characterized in that: Customers are clustered according to the customer portrait tag record information to obtain multiple customer groups, including: The customer portrait label record information includes N customer portrait label record information; Perform pairwise deviation analysis on the N customer portrait tag record information to obtain a number of customer portrait deviation values; According to the portrait deviation value threshold, cluster analysis is performed based on the several customer portrait deviation values to obtain the multiple clusters of customer groups.
3. The intelligent control method for questionnaire layout for survey service as claimed in claim 2, characterized in that: Perform pairwise deviation analysis on the N customer portrait tag record information to obtain several customer portrait deviation values, including: Obtaining a customer portrait label attribute set, wherein the customer portrait label attribute set includes a customer portrait label type attribute set and a customer portrait label value attribute set; Configuring a set of fault tolerance deviation intervals for the customer portrait tag numerical attribute set; Extracting first customer portrait label record information and second customer portrait label record information according to the N customer portrait label record information; When any type attribute of the customer portrait tag type attribute set of the first customer portrait tag record information and the second customer portrait tag record information is different, configuring the customer portrait deviation value to be greater than the portrait deviation value threshold; When the customer portrait label type attribute sets of the first customer portrait label record information and the second customer portrait label record information are the same, the ratio of the customer portrait label numerical attribute deviation sets of the first customer portrait label record information and the second customer portrait label record information that do not satisfy the customer portrait label numerical attribute set is counted and set as the customer portrait deviation value.
4. The intelligent control method for questionnaire layout for survey service as claimed in claim 1, characterized in that: Traversing the multiple customer groups, performing interest analysis with the calibration questionnaire element set, and obtaining multiple groups of interest analysis results, including: Extracting a first cluster of customer portrait labels according to the multiple clusters of customer groups; Sending the first cluster of customer portrait labels and the calibration questionnaire element set to a plurality of influence scoring nodes to obtain a plurality of groups of influence scoring results, wherein information of any two of the plurality of influence scoring nodes is not interoperable; According to the calibration questionnaire element set, the impact score results of each attribute of the plurality of groups of impact score results are added to obtain a first impact score sum result set; Adding the plurality of groups of impact score results to obtain an overall impact score sum result; The first set of influence score summation results is traversed, and a ratio to the overall summation result of the influence scores is calculated to obtain a first set of interest analysis results, which is added to the multiple sets of interest analysis results.
5. The intelligent control method for questionnaire layout for survey service as claimed in claim 1, characterized in that: According to the multiple groups of interest analysis results, layout analysis is performed on the questionnaire question sets to be laid out respectively to obtain multiple question ranking results, including: Extracting a first group of interest analysis results according to the multiple groups of interest analysis results; According to the first group of interest analysis results, the set of questions to be arranged in the questionnaire is sorted from large to small to obtain a first question sorting result, which is added to the multiple question sorting results.
6. An intelligent control device for questionnaire layout for survey services, characterized in that: The steps for implementing the intelligent control method for questionnaire layout for survey service as claimed in any one of claims 1 to 5, wherein the intelligent control device for questionnaire layout for survey service comprises: A request receiving module, the request receiving module is used to receive a questionnaire generation request, wherein the questionnaire generation request includes a product subject to the questionnaire; A data acquisition module, the data acquisition module is used to retrieve product historical dumping data according to the product subject of the questionnaire, wherein the product historical dumping data has customer portrait tag record information; A cluster analysis module, the cluster analysis module is used to cluster customers according to the customer portrait tag record information to obtain multiple customer groups; An element configuration module, the element configuration module is used to configure elements of the questionnaire target product through an element matching channel to obtain a calibration questionnaire element set; A topic matching module, the topic matching module is used to match the topic of the calibrated questionnaire element set through the questionnaire question type library to generate a set of questions to be arranged for the questionnaire, wherein the set of questions to be arranged for the questionnaire corresponds one-to-one to the set of calibrated questionnaire elements; An interest analysis module, the interest analysis module is used to traverse the multiple customer groups, perform interest analysis on the calibration questionnaire element set, and obtain multiple groups of interest analysis results; A layout analysis module, the layout analysis module is used to perform layout analysis on the questionnaire question sets to be laid out according to the multiple groups of interest analysis results, and obtain multiple question ranking results; A layout control module is used to control the layout of the questionnaire according to the sorting results of the multiple questions.
