A consumer satisfaction survey optimization method and system based on big data

Abstract

The present invention discloses a consumer satisfaction survey optimization method and system based on big data, which belongs to the field of electronic digital data processing technology, and includes the following steps: counting the current online platforms where the product is sold, recorded as each selling platform, obtaining the basic data of each selling platform for preprocessing, and obtaining the satisfaction survey data extraction ratio; extracting data according to the satisfaction survey data extraction ratio of each selling platform, and marking it as a first valid data set; counting the source objects corresponding to the first valid data set, and determining the data available evaluation value of each source object; obtaining a second valid data set by screening based on the data available evaluation value of each source object and the first valid data set; performing comprehensive processing based on the second valid data set to obtain a satisfaction survey evaluation label for the product, and performing survey tracing optimization prompts based on the satisfaction survey evaluation label for the product, thereby solving the problem of resource waste caused by excessive data on each platform in the prior art.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic digital data processing, and in particular to a method and system for optimizing consumer satisfaction surveys based on big data. Background Art

[0002] In a consumer-driven market, customer satisfaction has become a key indicator of business success. Therefore, it's crucial to deeply mine customer data, accurately analyze customer needs and pain points, and improve customer satisfaction.

[0003] Existing satisfaction survey methods primarily rely on mobile phone users' direct product evaluations and their emotional and sentimental information. These sentimental data are quantified to assess the specific levels of product attributes, while emotional data is used to assess consumers' subjective feelings. Based on this quantified emotional and sentimental data, the system ultimately calculates product satisfaction data, which is then used to evaluate consumer satisfaction.

[0004] For example, the user satisfaction evaluation method, device, system and data display platform announced in the invention patent with announcement number: CN111523914B include: obtaining at least one comment data for a product during a product transaction; obtaining sentiment data and emotional data for the product from at least one comment data, wherein the sentiment data is used to characterize the evaluation level of the attribute information of the product, and the emotional data is used to characterize the emotional level generated when evaluating the product; determining the product satisfaction data based on the sentiment data and emotional data, wherein the satisfaction data is used to characterize the changing trend of satisfaction during the transaction process of the product.

[0005] For example, the invention patent publication number CN110914815B discloses optimizing user satisfaction when training a cognitive hierarchical storage management system, including the following: The cognitive hierarchical storage management system receives feedback describing user satisfaction with how previous data access requests were served. The system uses this feedback to associate the metadata and storage tier of each previously requested data element with the user's satisfaction level. As feedback continues to be received, the system uses machine learning methods to determine the degree to which specific metadata patterns are associated with specific user satisfaction levels and specific storage tiers. The system then uses these associations when determining whether data associated with a specific metadata pattern should be migrated to another tier.

[0006] However, in the process of implementing the technical solutions of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0007] In the existing technology, the evaluation of user satisfaction mainly relies on the review data during the product transaction process. However, there is a lot of review data on various sales platforms, and some data is not valuable for evaluation. If all of them are evaluated, it will take up a lot of resources and waste a lot of time. Therefore, there is a problem of resource waste caused by too much data on each platform. Summary of the Invention

[0008] The embodiments of the present application provide a method and system for optimizing consumer satisfaction surveys based on big data, thereby solving the problem of resource waste caused by excessive data on various platforms in the prior art and enhancing the rationality and representativeness of survey data.

[0009] An embodiment of the present application provides a consumer satisfaction survey optimization method based on big data, comprising the following steps: counting the current online platforms where the product is sold, recorded as each selling platform, and obtaining the basic data of each selling platform for preprocessing to obtain the satisfaction survey data extraction ratio of each selling platform; counting the satisfaction questionnaire data set of each selling platform, and extracting data according to the satisfaction survey data extraction ratio of each selling platform, and marking the extracted satisfaction questionnaire data set as a first valid data set; counting the source objects corresponding to the first valid data set, and counting the consumption activity data of each source object for processing to determine the data available evaluation value of each source object; based on the data available evaluation value of each source object and the first valid data set, obtaining a second valid data set through screening; based on the second valid data set, and counting the historical sales performance data of the product for comprehensive processing, obtaining a satisfaction survey evaluation label for the product, and performing survey tracing optimization prompts based on the satisfaction survey evaluation label of the product.

[0010] Furthermore, the basic data of each sales platform specifically includes: the number of active users, commodity transaction volume, transaction volume growth rate, commodity return rate, commodity evaluation volume and user complaint rate of each sales platform within the first preset time period.

[0011] Furthermore, the said obtaining of the satisfaction survey data extraction ratio of each on-sale platform includes the following specific steps: obtaining a preset user activity benchmark value, commodity transaction volume benchmark value, commodity return rate benchmark value, commodity evaluation volume benchmark value and user complaint rate benchmark value of each on-sale platform in a first preset time period, and recording them as preset basic data; comparing the preset basic data with the basic data of each on-sale platform to obtain the basic characteristic value of each on-sale platform, the said basic characteristic value is used for quantitative analysis of the basic data of each on-sale platform, and is used as an analysis basis for the satisfaction survey data extraction ratio; based on the verification basic characteristic value preset in the database, and compared with the basic characteristic value of each on-sale platform, the data is extracted from the database synchronously. The original default survey data extraction ratios of each platform on sale are compared. If the basic characteristic value of a certain platform on sale is above the preset verification basic characteristic value, the difference between the basic characteristic value of the platform on sale and the preset verification basic characteristic value is extracted, and the difference is matched with the preset difference in the database to obtain the adjustment ratio corresponding to the preset difference, and the adjustment ratio of the platform on sale is added to the original default survey data extraction ratio of the platform on sale to obtain the satisfaction survey data extraction ratio of the platform on sale. If the basic characteristic value of a certain platform on sale is less than the preset verification basic characteristic value, the original default survey data extraction ratio of the platform on sale is used as the satisfaction survey data extraction ratio of the platform on sale.

[0012] Furthermore, the data availability evaluation value of each source object is determined, and the specific steps include: obtaining the consumption activity data of each source object, the consumption activity data including: return rate, number of repurchases, product purchase conversion rate, negative review rate and purchase frequency within a first preset time period; obtaining a preset consumer activity data benchmark set, and comparing it with the consumption activity data of each source object to obtain the data availability evaluation value of each source object; the preset consumer activity data benchmark set includes: return rate verification value, repurchase number verification value, product purchase conversion rate verification value, negative review rate verification value and purchase frequency verification value; the data availability evaluation value of each source object is used to quantify the consumption activity data of each source object, and is used as a basis for analyzing consumer satisfaction.

[0013] Furthermore, the method for determining the available evaluation value of the data of each source object is:

[0014]

[0015] Where, E i represents the data availability evaluation value of the i-th source object, i represents the number of the source object, i=1, 2, ..., n, n represents the number of source objects, r i represents the return rate of the i-th source object in the first preset time period, r1 represents the return rate verification value, e represents the base of the natural logarithm, f irepresents the number of repurchases of the i-th source object within the first preset time period, f1 represents the repurchase verification value, c i represents the product purchase conversion rate of the i-th source object in the first preset time period, c1 represents the product purchase conversion rate verification value, q i represents the negative review rate of the i-th source object in the first preset time period, q1 represents the negative review rate verification value, and p i represents the purchase frequency of the i-th source object in the first preset time period, p1 represents the purchase frequency verification value, α1 represents the return rate influence weight of the source object in the first preset time period, α2 represents the repurchase number influence weight of the source object in the first preset time period, α3 represents the commodity purchase conversion rate influence weight of the source object in the first preset time period, α4 represents the negative review rate influence weight of the source object in the first preset time period, and α5 represents the purchase frequency influence weight of the source object in the first preset time period.

[0016] Furthermore, the second valid data set is obtained by screening, and the specific steps include: obtaining a data availability evaluation threshold in the database, and comparing it with the data availability evaluation value of each source object; if the data availability evaluation value of a certain source object is above the data availability evaluation threshold, then the satisfaction questionnaire data of the source object is determined to be second valid data; if the data availability evaluation value of the source object is less than the data availability evaluation threshold, then the satisfaction questionnaire data of the source object is determined to be non-valid data and deleted; all the second valid data are aggregated to obtain a second valid data set; the second valid data set includes each target source object and the satisfaction questionnaire data of each target source object.

[0017] Furthermore, the specific steps of obtaining the satisfaction survey evaluation label of the product include: obtaining the satisfaction survey evaluation value of the product based on the second valid data set and processing according to the historical sales performance data of the product; obtaining the satisfaction survey evaluation threshold of the product in the database and comparing it with the satisfaction survey evaluation value of the product; if the satisfaction survey evaluation value of the product is above the satisfaction survey evaluation threshold of the product, then defining the satisfaction survey evaluation label of the product as a high-satisfaction product; if the satisfaction survey evaluation value of the product is less than the satisfaction survey evaluation threshold of the product, then defining the satisfaction survey evaluation label of the product as a low-satisfaction product.

[0018] Furthermore, the survey and traceability optimization prompts are performed based on the satisfaction survey evaluation label of the product. The specific method is: if the satisfaction survey evaluation label of the product is defined as a low-satisfaction product, then based on the second valid data set obtained by screening, the satisfaction questionnaire data of each target source object is counted, and the satisfaction questionnaire data includes the evaluation text of each target source object on the product in each dimension; based on the evaluation text of each target source object on the product in each dimension, a number of evaluation keywords of each target source object on the product in each dimension are counted, and according to the positive keyword collection and negative keyword collection preset in the database, the total number of positive evaluation keywords and the total number of negative evaluation keywords of each target source object on the product in each dimension are screened through mapping and matching; according to the total number of positive evaluation keywords and the total number of negative evaluation keywords of each target source object on the product in each dimension, the text evaluation value of the product in each dimension is processed, and the text evaluation values are arranged in ascending order to obtain the survey and traceability arrangement order of the product, and the survey and traceability arrangement order of the product is transmitted to the central controller.

[0019] Positive keyword collections include: good, great, practical, beautiful, etc., and negative keyword collections include: bad, poor, ugly, etc. The central controller refers to an electronic device used for data processing, data analysis, and system control. The central controller can optimize the processing of products based on the order of product investigation and traceability.

[0020] Furthermore, the satisfaction survey evaluation value of the product is obtained based on the second valid data set and processed according to the historical sales performance data of the product. The specific steps are: based on the satisfaction questionnaire data of each target source object in the second valid data set, the satisfaction questionnaire data of all target source objects are summarized to obtain the satisfaction questionnaire text of the product, and a number of keywords corresponding to the product are counted therefrom, thereby matching the positive keyword collection and negative keyword collection preset in the database, and then obtaining the total number of positive evaluation keywords and the total number of negative evaluation keywords corresponding to the satisfaction questionnaire text of the product through mapping and matching; based on the total number of positive evaluation keywords and the total number of negative evaluation keywords corresponding to the satisfaction questionnaire text of the product, and processing according to the historical sales performance data of the product to obtain the satisfaction survey evaluation value of the product; the historical sales performance data of the product include the historical per capita positive evaluation keywords total number and the historical per capita negative evaluation keywords total number; the satisfaction survey evaluation value of the product is used to quantitatively analyze the second valid data set, and used as a basis for analyzing product satisfaction.

[0021] A consumer satisfaction survey optimization system based on big data, characterized by comprising: a data acquisition module, a data extraction module, a data evaluation module, a data screening module, and a label processing module; wherein the data acquisition module is used to count the current online platforms where the product is sold, denoted as each selling platform, and obtain basic data of each selling platform for preprocessing to obtain the satisfaction survey data extraction ratio of each selling platform; the data extraction module is used to count the satisfaction questionnaire data set of each selling platform, extract data according to the satisfaction survey data extraction ratio of each selling platform, and mark the extracted satisfaction questionnaire data set as a first valid data set; the data evaluation module is used to count the source objects corresponding to the first valid data set, count the consumption activity data of each source object for processing, and determine the data available evaluation value of each source object; the data screening module is used to obtain a second valid data set by screening based on the data available evaluation value of each source object and the first valid data set; the label processing module is used to perform comprehensive processing based on the second valid data set and the historical sales performance data of the product to obtain a satisfaction survey evaluation label for the product, and provide survey retrospective optimization prompts based on the satisfaction survey evaluation label of the product.

[0022] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0023] 1. The present invention provides a consumer satisfaction survey optimization method based on big data. Through multi-platform data fusion and preprocessing, it accurately calculates the satisfaction survey data extraction ratio, and extracts survey data from each platform according to the corresponding satisfaction survey data extraction ratio, thereby ensuring the rationality of the quantity and representativeness of the quality of the survey data, and effectively solving the problem of resource waste caused by excessive data on each platform in the existing technology.

[0024] 2. The present invention comprehensively evaluates consumer activity data to determine the available evaluation value of the data, thereby achieving the screening of high-quality data and avoiding the influence of uneven data quality on the accuracy of the analysis results.

[0025] 3. By obtaining the satisfaction questionnaire data of each target source object and summarizing the satisfaction questionnaire data of all target source objects, the satisfaction questionnaire survey text of the product is obtained, and then the total number of positive evaluation keywords and the total number of negative evaluation keywords corresponding to the satisfaction questionnaire survey text of the product are obtained through mapping and matching. Then, the text evaluation values are arranged in order from small to large to obtain the survey and traceability arrangement order of the product, which is beneficial for enterprises to optimize services according to the survey and traceability arrangement order. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1A flowchart of a method for optimizing a consumer satisfaction survey based on big data provided in an embodiment of the present application;

[0027] Figure 2 A module diagram of a consumer satisfaction survey optimization system based on big data provided in an embodiment of the present application. DETAILED DESCRIPTION

[0028] The embodiment of the present application solves the problem of resource waste caused by excessive data on each platform in the prior art by providing a consumer satisfaction survey optimization method and system based on big data. The method comprises: counting the current online platforms where the product is sold, recording them as each selling platform, obtaining the basic data of each selling platform for preprocessing, and obtaining the satisfaction survey data extraction ratio of each selling platform; counting the satisfaction questionnaire data set of each selling platform, and extracting data according to the satisfaction survey data extraction ratio of each selling platform, and marking the extracted satisfaction questionnaire data set as a first valid data set; counting the source objects corresponding to the first valid data set, and counting the consumption activity data of each source object for processing, and determining the data available evaluation value of each source object; based on the data available evaluation value of each source object and the first valid data set, obtaining a second valid data set through screening; based on the second valid data set, and counting the historical sales performance data of the product for comprehensive processing, obtaining the satisfaction survey evaluation label of the product, and performing survey tracing optimization prompts based on the satisfaction survey evaluation label of the product, thereby ensuring the quantity rationality and quality representativeness of the survey data.

[0029] The technical solution in the embodiment of the present application is to solve the above-mentioned problem of resource waste caused by excessive data on each platform. The overall idea is as follows: by counting the current online platforms where the product is sold, recorded as each selling platform, and obtaining the basic data of each selling platform for preprocessing, the satisfaction survey data extraction ratio of each selling platform is obtained; the satisfaction questionnaire data set of each selling platform is counted, and data is extracted according to the satisfaction survey data extraction ratio of each selling platform, and the extracted satisfaction questionnaire data set is marked as a first valid data set; the source objects corresponding to the first valid data set are counted, and the consumption activity data of each source object are counted for processing to determine the data available evaluation value of each source object; based on the data available evaluation value of each source object and the first valid data set, a second valid data set is obtained by screening; based on the second valid data set, the historical sales performance data of the product is counted for comprehensive processing to obtain the satisfaction survey evaluation label of the product, and survey tracing optimization prompts are performed based on the satisfaction survey evaluation label of the product, thereby enhancing the rationality and representativeness of the survey data.

[0030] In order to better understand the above technical solution, the above technical solution will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0031] like Figure 1 As shown, it is a flow chart of a consumer satisfaction survey optimization method based on big data provided by an embodiment of the present application, the method comprising the following steps: counting the current online platforms where the product is sold, recorded as each selling platform, and obtaining the basic data of each selling platform for preprocessing to obtain the satisfaction survey data extraction ratio of each selling platform; counting the satisfaction questionnaire data set of each selling platform, and extracting data according to the satisfaction survey data extraction ratio of each selling platform, marking the extracted satisfaction questionnaire data set as the first valid data set; counting the source objects corresponding to the first valid data set, and counting the consumption activity data of each source object for processing to determine the data available evaluation value of each source object; based on the data available evaluation value of each source object and the first valid data set, obtaining the second valid data set by screening; according to the second valid data set, and counting the historical sales performance data of the product for comprehensive processing, obtaining the satisfaction survey evaluation label of the product, and performing survey tracing optimization prompts based on the satisfaction survey evaluation label of the product.

[0032] In this embodiment, basic data from each platform is obtained and preprocessed to determine the percentage of satisfaction survey data extracted from each platform. Data extraction is performed based on the percentage of satisfaction survey data extracted from each platform, using a random sampling method. Specifically, a number of satisfaction survey data points are randomly sampled from each platform's satisfaction survey data using a corresponding percentage. These satisfaction survey data points form a satisfaction questionnaire dataset. Random sampling ensures that the sample data is representative of the overall population, making the analysis results more objective and accurate, and more realistically reflecting the overall situation. Obtaining basic data from each platform and preprocessing it to determine the percentage of satisfaction survey data extracted from each platform helps save costs. In a big data environment, processing all the data is often time-consuming and labor-intensive. By determining a reasonable percentage of satisfaction survey data, the time and resource costs of data processing can be significantly reduced while ensuring the reliability of the analysis results. By statistically analyzing the consumer activity data of each source object and determining the usable evaluation value of each source object's data, erroneous, incomplete, or abnormal data can be identified and eliminated, ensuring the accuracy and reliability of the data. Furthermore, malicious reviews or abnormal data can be identified and eliminated, ensuring the fairness and objectivity of the evaluation results. Determining the available evaluation value of the data of each source object is to ensure that this data can be effectively utilized in the subsequent analysis, decision-making and business optimization process. The purpose of providing survey and traceability optimization prompts based on the product satisfaction survey evaluation label is to enable companies to gain an in-depth understanding of customers' real feedback on the product through the satisfaction survey evaluation label, thereby identifying the shortcomings of the product and taking targeted measures to improve it to enhance customer satisfaction. At the same time, survey and traceability can help companies discover defects or unreasonable aspects in product design, provide data support for product iteration and optimization, and make products more in line with market demand and consumer preferences. By tracing the supply chain information behind the evaluation label, companies can monitor the performance of suppliers and ensure the quality of raw materials and parts, thereby optimizing supply chain management and reducing risks caused by supply chain problems.

[0033] Furthermore, the basic data of each sales platform specifically includes: the number of active users, commodity transaction volume, transaction volume growth rate, commodity return rate, commodity evaluation volume and user complaint rate of each sales platform within the first preset time period.

[0034] In this embodiment, the first preset time period refers to a relatively long period of time, for example, one month.

[0035] It should be noted that the method for obtaining the basic characteristic values of each selling platform is:

[0036]

[0037] Where Z irepresents the basic characteristic value of the jth platform on sale, j represents the number of the jth platform on sale, j = 1, 2, ..., J, J represents the number of each platform on sale, h j represents the number of active users on the jth selling platform in the first preset time period, △h j M represents the preset user activity benchmark value of the jth selling platform, j represents the commodity transaction volume of the jth selling platform in the first preset time period, △M i represents the preset commodity transaction volume benchmark value of the jth selling platform, b aj represents the transaction volume growth rate of the jth selling platform in the first preset time period, △b aj represents the preset transaction volume growth rate benchmark value of the jth selling platform in the first preset time period, g j represents the product return rate of the jth selling platform in the first preset time period, △g j represents the preset benchmark value of the product return rate on the jth selling platform, y j represents the number of product reviews on the jth selling platform in the first preset time period, △y j A represents the preset benchmark value of product evaluation on the jth selling platform, j represents the user complaint rate of the jth selling platform in the first preset time period, △A j It represents the preset user complaint rate benchmark value of the jth selling platform, β1 represents the influence weight of the number of active users, β2 represents the influence weight of the commodity transaction volume, β3 represents the influence weight of the commodity return rate, β4 represents the influence weight of the commodity evaluation volume, β5 represents the influence weight of the user complaint rate, and β6 represents the influence weight of the commodity transaction volume.

[0038] It should be noted that the basic data of each platform on sale can be obtained by querying the background data of each platform on sale, the preset basic data can be obtained by querying the database, and the influence weight can be obtained through the database, for example: the influence weight of the number of active users, which represents the numerical value of the degree of influence of the number of active users on the basic characteristic values of each platform on sale. When used, the influence weight corresponding to the number of active users of each platform on sale in the first preset time period can be directly obtained from the database. The corresponding relationship can be a pre-set mapping relationship. For example, the mapping relationship forms a mapping set with the number of active users of each platform on sale in the first preset time period and the influence weight of the number of active users. The real-time active user number is input into the mapping set to obtain the influence weight corresponding to the number of active users. The mapping relationship can be a one-to-one correspondence or a many-to-one relationship. The acquisition method of other influence weights such as the commodity transaction volume influence weight, the commodity return rate influence weight, the commodity evaluation volume influence weight, the user complaint rate influence weight and the commodity transaction volume influence weight is consistent with the acquisition method of the user activity number influence weight, and will not be repeated here.

[0039] It should also be noted that in this embodiment, the basic characteristic values of each sales platform are obtained by processing the number of active users, commodity transaction volume, transaction volume growth rate, commodity return rate, commodity evaluation volume and user complaint rate of each sales platform within the first preset time period. This is based on the mutual influence between these parameters. For example, the number of active users is the basis of commodity transaction volume. The more active users there are, the more potential buyers there are, which may drive the growth of commodity transaction volume. Conversely, the growth of commodity transaction volume may also attract more users to participate and increase user activity. An increase in commodity transaction volume usually brings more user reviews. After purchasing and experiencing the product, users tend to post reviews on the platform. Positive reviews can attract more potential users and promote the growth of commodity transaction volume; while negative reviews may reduce users' willingness to buy and affect transaction volume.

[0040] It should be understood that the preset user activity benchmark values, commodity transaction volume benchmark values, and commodity evaluation benchmark values for each selling platform represent the data minimum reference standards under ideal circumstances, namely, the ideal minimum reference standards. The user activity count, commodity transaction volume, transaction volume growth rate, and commodity evaluation count of each selling platform in the basic data of each selling platform during the first preset time period are compared with the ideal minimum reference standards. The purpose is to consider that if the user activity count, commodity transaction volume, transaction volume growth rate, and commodity evaluation count of each selling platform during the first preset time period are at a higher numerical level than the ideal minimum reference standards, then it indicates that the basic characteristic values of each selling platform are better. The preset return rate benchmark values and user complaint rate benchmark values for each selling platform represent the data reference standards under ideal circumstances, namely, the ideal reference standards. The product return rate and user complaint rate of each selling platform during the first preset time period are compared with the ideal reference standards. The purpose is to consider that if the product return rate and user complaint rate of each selling platform during the first preset time period are at a lower numerical level than the ideal reference standards, then it indicates that the basic characteristic values of each selling platform are better.

[0041] Furthermore, the satisfaction survey data extraction ratio of each sales platform is obtained, and the specific steps include: obtaining the preset user activity number benchmark value, commodity transaction volume benchmark value, commodity return rate benchmark value, commodity evaluation volume benchmark value and user complaint rate benchmark value of each sales platform in the first preset time period, and recording them as preset basic data; comparing the preset basic data with the basic data of each sales platform to obtain the basic characteristic value of each sales platform, and the basic characteristic value is used to quantitatively analyze the basic data of each sales platform, and used as the analysis basis for the satisfaction survey data extraction ratio; based on the verification basic characteristic value preset in the database, and compared with the basic characteristic value of each sales platform, the satisfaction survey data extraction ratio of each sales platform is extracted from the database synchronously. The original default survey data extraction ratio of the platforms on sale is compared. If the basic characteristic value of a certain platform on sale is above the preset verification basic characteristic value, the difference between the basic characteristic value of the platform on sale and the preset verification basic characteristic value is extracted, and the difference is matched with the preset difference in the database to obtain the adjustment ratio corresponding to the preset difference, and the adjustment ratio of the platform on sale is added to the original default survey data extraction ratio of the platform on sale to obtain the satisfaction survey data extraction ratio of the platform on sale. If the basic characteristic value of a certain platform on sale is less than the preset verification basic characteristic value, the original default survey data extraction ratio of the platform on sale is used as the satisfaction survey data extraction ratio of the platform on sale.

[0042] In this embodiment, each selling platform refers to an online platform that is authorized by each online commodity management manufacturer to sell commodities.

[0043] It should be noted that the preset basic verification characteristic value can be obtained from the database. The preset basic verification characteristic value is a judgment standard set for the basic characteristic value. When the basic characteristic value of the platform on sale is higher than the preset basic verification characteristic value, it means that the platform on sale is highly representative.

[0044] Furthermore, the data availability evaluation value of each source object is determined, and the specific steps include: obtaining the consumption activity data of each source object, the consumption activity data including: return rate, number of repurchases, product purchase conversion rate, negative review rate and purchase frequency within a first preset time period; obtaining a preset consumer activity data benchmark set, and comparing it with the consumption activity data of each source object to obtain the data availability evaluation value of each source object; the preset consumer activity data benchmark set includes: return rate verification value, repurchase number verification value, product purchase conversion rate verification value, negative review rate verification value and purchase frequency verification value; the data availability evaluation value of each source object is used to quantify the consumption activity data of each source object, and is used as a basis for analyzing consumer satisfaction.

[0045] In this embodiment, it should be noted that the consumption activity data of each source object can be obtained by big data statistics of the historical consumption of each source object. The greater the data usable evaluation value of each source object, the more valuable the data of the source object is for analysis and the more representative it is of the consumption behavior of the general public.

[0046] Furthermore, the method for determining the available evaluation value of the data of each source object is:

[0047]

[0048] Where, E i represents the data availability evaluation value of the i-th source object, i represents the number of the source object, i=1, 2, ..., n, n represents the number of source objects, r i represents the return rate of the i-th source object in the first preset time period, r1 represents the return rate verification value, e represents the base of the natural logarithm, f i represents the number of repurchases of the i-th source object within the first preset time period, f1 represents the repurchase verification value, c i represents the product purchase conversion rate of the i-th source object in the first preset time period, c1 represents the product purchase conversion rate verification value, q i represents the negative review rate of the i-th source object in the first preset time period, q1 represents the negative review rate verification value, and p i represents the purchase frequency of the i-th source object in the first preset time period, p1 represents the purchase frequency verification value, α1 represents the return rate influence weight of the source object in the first preset time period, α2 represents the repurchase number influence weight of the source object in the first preset time period, α3 represents the commodity purchase conversion rate influence weight of the source object in the first preset time period, α4 represents the negative review rate influence weight of the source object in the first preset time period, and α5 represents the purchase frequency influence weight of the source object in the first preset time period.

[0049] In this embodiment, it should be noted that the consumption activity data of each source object can be statistically obtained through the detailed data analysis function provided by the backend system of each sales platform, and the preset consumption activity data benchmark set can be obtained through the database.

[0050] It should also be noted that the return rate influence weight of the source object within the first preset time period represents the numerical value of the influence of the return rate of the source object within the first preset time period on the available evaluation value of the data of each source object. When used, the influence weight corresponding to the return rate within the first preset time period can be directly obtained from the database. The corresponding relationship can be a pre-set mapping relationship. For example, the mapping relationship forms a mapping set of the return rate within the first preset time period and the return rate influence weight of the source object within the first preset time period. The real-time return rate is input into the mapping set to obtain the influence weight corresponding to the return rate of the source object within the first preset time period. The mapping relationship can be a one-to-one correspondence or a many-to-one relationship. The acquisition method of other influence weights such as the influence weight of the number of repurchases, the influence weight of the product purchase conversion rate, the influence weight of the negative review rate, and the influence weight of the purchase frequency is consistent with the acquisition method of the return rate influence weight of the source object within the first preset time period, and will not be repeated here.

[0051] It should be understood that in this embodiment, the available evaluation value of the data of each source object obtained by processing the consumption activity data of each source object takes into account the mutual influence between these parameters. For example, the return rate, number of repurchases, product purchase conversion rate, negative review rate and purchase frequency are interrelated indicators that jointly affect the overall performance of consumption activities. A high return rate may reduce the product purchase conversion rate and the number of repurchases, while increasing the negative review rate and reducing the purchase frequency. On the contrary, a high number of repurchases and purchase frequency usually means that customers have high brand loyalty and satisfaction, which helps to improve the product purchase conversion rate and reduce the return rate and negative review rate. The level of product purchase conversion rate is directly related to the brand's ability to attract new customers and retain old customers, further affecting the return rate, number of repurchases and negative review rate. The negative review rate, as an important indicator of customer feedback, has a negative impact on the return rate, number of repurchases, purchase frequency and product purchase conversion rate.

[0052] Furthermore, a second valid data set is obtained through screening, and the specific steps include: obtaining a data availability evaluation threshold in the database, and comparing it with the data availability evaluation value of each source object; if the data availability evaluation value of a certain source object is above the data availability evaluation threshold, the satisfaction questionnaire data of the source object is determined to be the second valid data; if the data availability evaluation value of the source object is less than the data availability evaluation threshold, the satisfaction questionnaire data of the source object is determined to be non-valid data and deleted; all the second valid data are aggregated to obtain a second valid data set; the second valid data set includes each target source object and the satisfaction questionnaire data of each target source object.

[0053] In this embodiment, it should be noted that the satisfaction questionnaire data is an important basis for evaluating the service quality and customer satisfaction of the source object, and the available evaluation value of the data is of great value for the analysis of satisfaction, so it is necessary to analyze the satisfaction questionnaire data to avoid waste of resources due to excessive platform data.

[0054] Furthermore, a satisfaction survey evaluation label of the product is obtained. The specific steps include: based on the second valid data set, and processed according to the historical sales performance data of the product, to obtain the satisfaction survey evaluation value of the product; obtaining the satisfaction survey evaluation threshold of the product in the database, and comparing it with the satisfaction survey evaluation value of the product; if the satisfaction survey evaluation value of the product is above the satisfaction survey evaluation threshold of the product, then the satisfaction survey evaluation label of the product is defined as a high-satisfaction product; if the satisfaction survey evaluation value of the product is less than the satisfaction survey evaluation threshold of the product, then the satisfaction survey evaluation label of the product is defined as a low-satisfaction product.

[0055] In this embodiment, it should be noted that the method for obtaining the satisfaction survey evaluation value of the product is:

[0056]

[0057] In the formula, G represents the satisfaction survey evaluation value of the product, W P W represents the total number of historical average positive evaluation keywords for a product. N represents the total number of historical average negative evaluation keywords for a product, H represents the number of target source objects, and T P The total number of positive evaluation keywords corresponding to the product satisfaction questionnaire text, T N Indicates the total number of negative evaluation keywords corresponding to the product satisfaction questionnaire text.

[0058] It should be noted that the satisfaction survey evaluation value of a product is used to represent the consumer's satisfaction with the product. When a consumer is surveyed and evaluated, there must be a certain number of historical average positive evaluation keywords and historical average negative evaluation keywords for the product. Therefore, there is no situation where both are zero at the same time. Therefore, the historical average positive evaluation keywords and historical average negative evaluation keywords for the product are not equal to zero.

[0059] It should also be noted that the historical sales performance data of the product can be obtained through the backend management of each sales platform, and the number of target source objects can be obtained through statistical software.

[0060] Furthermore, survey and traceability optimization prompts are performed based on the satisfaction survey and evaluation labels of the products. The specific method is: if the satisfaction survey and evaluation label of the product is defined as a low-satisfaction product, then based on the second valid data set obtained through screening, the satisfaction questionnaire data of each target source object is counted, and the satisfaction questionnaire data includes the evaluation text of each target source object on the product in each dimension; based on the evaluation text of each target source object on the product in each dimension, a number of evaluation keywords of each target source object on the product in each dimension are counted, and according to the positive keyword collection and negative keyword collection preset in the database, the total number of positive evaluation keywords and the total number of negative evaluation keywords of each target source object on the product in each dimension are screened through mapping and matching; according to the total number of positive evaluation keywords and the total number of negative evaluation keywords of each target source object on the product in each dimension, the text evaluation value of the product in each dimension is processed, and the text evaluation values are arranged in ascending order to obtain the survey and traceability arrangement order of the product, and the survey and traceability arrangement order of the product is transmitted to the central controller.

[0061] In this embodiment, statistics of several evaluation keywords of each target source object on the product in each dimension can be obtained through e-commerce data analysis software, such as SellerSprite, Feiyu Data Advisor, and Oulu.

[0062] Obtaining satisfaction questionnaire data from each target source audience helps capture genuine customer feedback and perceptions of products or services, providing a deeper understanding of their expectations, needs, and dissatisfactions. Each target source audience includes all consumers on each sales platform. Examples of these data include evaluation text on service attitude, logistics speed, product value for money, and product quality. Positive keywords include "good," "great," "practical," and "beautiful," while negative keywords include "bad," "poor," and "ugly."

[0063] It should be noted that the satisfaction questionnaire data includes the evaluation texts of each target source object on the product in various dimensions, which can be obtained through the backend system statistics of each sales platform. Usually, the satisfaction questionnaire is distributed by the enterprise, and after collection, it is analyzed using data analysis software to obtain the satisfaction questionnaire data.

[0064] The method to obtain the text evaluation value of the product in each dimension is:

[0065]

[0066] Where V k Indicates the text evaluation value of the product in each dimension, k represents the dimension number, k = 1, 2, 3, ..., k max , k max Indicates the number of dimensions, P kIndicates the total number of positive evaluation keywords for the product in each dimension, P k0 Indicates the total number of positive evaluation keywords of the preset product in each dimension, N k Indicates the total number of negative evaluation keywords for the product in each dimension, N k0 It represents the verification value of the total number of negative evaluation keywords of the preset product in each dimension, γ1 represents the influence weight of the positive evaluation keywords of the product in each dimension, and γ2 represents the influence weight of the negative evaluation keywords of the product in each dimension.

[0067] It should be noted that the textual evaluation values for each dimension of a product are used to represent the satisfaction level of the product in each dimension. The total number of positive evaluation keywords for a product in the kth dimension and the total number of negative evaluation keywords for a product in the kth dimension must not be zero; if they are zero, the research is not meaningful. The preset verification value for the total number of positive evaluation keywords for a product in the kth dimension can be obtained from the database, and the preset verification value for the total number of negative evaluation keywords for a product in the kth dimension can be obtained from the database.

[0068] In this embodiment, the dimensions include but are not limited to: service attitude dimension, quality dimension, price dimension, logistics dimension, and cost-effectiveness dimension.

[0069] It should also be noted that the influence weight of the positive evaluation keywords of the product in each dimension represents the numerical value of the degree of influence of the positive evaluation keywords of the product in each dimension on the text evaluation value of the product in each dimension. When used, the influence weight corresponding to the positive evaluation keywords of the product in each dimension can be directly obtained from the database. The corresponding relationship can be a pre-set mapping relationship. For example, the positive evaluation keywords of the product in each dimension and the positive evaluation keyword influence weight of the product in each dimension form a mapping set. The real-time positive evaluation keywords of the product in each dimension are input into the mapping set to obtain the positive evaluation influence weight of the product in each dimension. The mapping relationship can be a one-to-one correspondence or a many-to-one relationship. The method of obtaining the influence weight of the negative evaluation keywords of the product in each dimension is consistent with the method of obtaining the return rate influence weight of the source object in the first preset time period, which will not be repeated here.

[0070] It should also be noted that in this embodiment, the text evaluation value of the product in each dimension is obtained by processing the total number of positive evaluation keywords and the total number of negative evaluation keywords of the product in each dimension. This takes into account the mutual influence between these parameters. For example: if the proportion of the total number of positive evaluation keywords to the total number of keywords is higher, the proportion of the total number of negative evaluation keywords to the total number of keywords will be lower. At this time, the overall evaluation of consumers is more satisfied, and vice versa.

[0071] It should be understood that the preset verification value of the total number of positive evaluation keywords for the product in each dimension represents the minimum reference standard for positive evaluation keywords under ideal circumstances, that is, the ideal minimum reference standard for positive evaluation keywords. By comparing the preset verification value of the total number of positive evaluation keywords for the product in each dimension with the ideal minimum reference standard for positive evaluation keywords, the purpose is to take into account that if the preset verification value of the total number of positive evaluation keywords for the product in each dimension is at a higher numerical level than the ideal minimum reference standard for positive evaluation keywords, it indicates that the verification value data of the total number of positive evaluation keywords for the product in each dimension is better, and the text evaluation value of the product in each dimension is higher. The preset verification value of the total number of negative evaluation keywords for the product in each dimension represents the reference standard for negative evaluation keywords under ideal circumstances, that is, the ideal negative evaluation keyword reference standard. By comparing the preset verification value of the total number of negative evaluation keywords for the product in each dimension with the ideal negative evaluation keyword reference standard, the purpose is to take into account that if the preset verification value of the total number of negative evaluation keywords for the product in each dimension is at a lower numerical level than the ideal negative evaluation keyword reference standard, it indicates that the verification value data of the total number of negative evaluation keywords for the product in each dimension is better, and the text evaluation value of the product in each dimension is higher.

[0072] Furthermore, based on the second valid data set and processed according to the historical sales performance data of the product, a satisfaction survey evaluation value of the product is obtained. The specific steps are: based on the satisfaction questionnaire data of each target source object in the second valid data set, the satisfaction questionnaire data of all target source objects are summarized to obtain the satisfaction questionnaire text of the product, and a number of keywords corresponding to the product are counted therefrom, thereby matching the positive keyword collection and negative keyword collection preset in the database, and then obtaining the total number of positive evaluation keywords and the total number of negative evaluation keywords corresponding to the satisfaction questionnaire text of the product through mapping and matching; based on the total number of positive evaluation keywords and the total number of negative evaluation keywords corresponding to the satisfaction questionnaire text of the product, and processed according to the historical sales performance data of the product to obtain the satisfaction survey evaluation value of the product; the historical sales performance data of the product include the historical per capita positive evaluation keywords total number and the historical per capita negative evaluation keywords total number; the satisfaction survey evaluation value of the product is used to quantitatively analyze the second valid data set, and used as a basis for analyzing product satisfaction.

[0073] In this embodiment, the above-described processing flow integrates the satisfaction questionnaire data from each target source object in the second valid dataset, not only obtaining rich consumer feedback, but also quantifying the positive and negative image of the product in consumers' minds through keyword statistics and mapping matching. This process not only improves the efficiency and accuracy of data analysis but also lays a solid foundation for the calculation of subsequent product satisfaction survey evaluation values. By counting the total number of positive and negative evaluation keywords, a comprehensive and detailed understanding of the product's performance in different dimensions can be achieved, including but not limited to service attitude, quality, price, logistics, and cost-effectiveness. This multi-dimensional assessment helps companies more accurately identify the strengths and weaknesses of their products, providing strong support for subsequent product improvements.

[0074] A consumer satisfaction survey optimization system based on big data is characterized by comprising: a data acquisition module, a data extraction module, a data evaluation module, a data screening module and a label processing module, wherein the data acquisition module is used to count the current online platforms where the product is sold, recorded as each selling platform, and obtain basic data of each selling platform for preprocessing to obtain the satisfaction survey data extraction ratio of each selling platform; the data extraction module is used to count the satisfaction questionnaire data set of each selling platform, and extract data according to the satisfaction survey data extraction ratio of each selling platform, and mark the extracted satisfaction questionnaire data set as a first valid data set; the data evaluation module is used to count the source objects corresponding to the first valid data set, and count the consumption activity data of each source object for processing to determine the data available evaluation value of each source object; the data screening module is used to obtain a second valid data set by screening based on the data available evaluation value of each source object and the first valid data set; the label processing module is used to perform comprehensive processing based on the second valid data set and the historical sales performance data of the product to obtain the satisfaction survey evaluation label of the product, and to provide survey retrospective optimization prompts based on the satisfaction survey evaluation label of the product.

[0075] In this embodiment, multiple modules work together to first collect and pre-process basic data from each sales platform to determine the extraction ratio of satisfaction survey data. Subsequently, a satisfaction questionnaire data set is extracted from each platform based on these ratios to form a first valid data set. Next, the evaluation module processes the consumer activity data in the first valid data set, determines the data availability, and generates an evaluation value. The data screening module uses these evaluation values to eliminate low-quality data and obtain a second valid data set. Finally, the label processing module combines the second valid data set with historical sales data to generate a satisfaction survey evaluation label for the product, while also providing survey traceability and optimization tips to support the company's accurate decision-making and continuous improvement.

[0076] To sum up, this embodiment accurately calculates the satisfaction survey data extraction ratio through multi-platform data fusion and preprocessing, and extracts survey data from each platform according to the corresponding satisfaction survey data extraction ratio, thereby ensuring the rationality of the quantity and representativeness of the quality of the survey data, and effectively solving the problem of resource waste caused by excessive data on each platform in the existing technology.

[0077] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0078] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0079] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0080] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0081] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0082] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A method for optimizing consumer satisfaction survey based on big data, characterized in that: The following steps are involved: Count the online platforms where the product is currently sold, denoted as each selling platform, and obtain basic data of each selling platform for preprocessing. Compare the preset basic data with the basic data of each selling platform to obtain basic characteristic values of each selling platform. The basic characteristic values are used to quantitatively analyze the basic data of each selling platform and serve as an analysis basis for the proportion of satisfaction survey data extracted, thereby obtaining the proportion of satisfaction survey data extracted for each selling platform; Collecting satisfaction questionnaire data sets from various platforms currently on sale, extracting data based on the satisfaction survey data extraction ratio of each platform currently on sale, and marking the extracted satisfaction questionnaire data sets as the first valid data sets; Counting each source object corresponding to the first valid data set, and processing the consumption activity data of each source object, and determining the available evaluation value of the data of each source object; Based on the data availability evaluation value of each source object and the first valid data set, a second valid data set is obtained by screening; According to the second valid data set, the historical sales performance data of the product is statistically processed to obtain the satisfaction survey evaluation value of the product, and further obtain the satisfaction survey evaluation label of the product, and perform survey and tracing optimization prompts based on the satisfaction survey evaluation label of the product. The survey and tracing optimization prompts based on the satisfaction survey evaluation label of the product are statistically analyzed according to the satisfaction survey evaluation label of the product, and the total number of positive evaluation keywords and the total number of negative evaluation keywords of each target source object for the product in each dimension are screened and processed to obtain the text evaluation value of the product in each dimension, and survey and tracing optimization prompts are performed in sequence according to the text evaluation value.

2. The method for optimizing a consumer satisfaction survey based on big data according to claim 1, characterized in that: The basic data of each sales platform specifically includes: The number of active users, product transaction volume, transaction volume growth rate, product return rate, product evaluation volume and user complaint rate of each sales platform during the first preset time period.

3. The method for optimizing a consumer satisfaction survey based on big data according to claim 1, characterized in that: The specific steps of obtaining the satisfaction survey data extraction ratio of each sales platform include: Obtain the preset user activity benchmark value, product transaction volume benchmark value, product return rate benchmark value, product review volume benchmark value, and user complaint rate benchmark value for each sales platform within the first preset time period, and record them as the preset basic data; Based on the verification basic characteristic value preset in the database and compared with the basic characteristic value of each platform on sale, the original default survey data extraction ratio of each platform on sale is simultaneously extracted from the database. After comparison, if the basic characteristic value of a certain platform on sale is above the preset verification basic characteristic value, the difference between the basic characteristic value of the platform on sale and the preset verification basic characteristic value is extracted, and the difference is matched with the difference preset in the database to obtain the adjustment ratio corresponding to the preset difference, and the adjustment ratio of the platform on sale is added to the original default survey data extraction ratio of the platform on sale to obtain the satisfaction survey data extraction ratio of the platform on sale. If the basic characteristic value of a certain platform on sale is less than the preset verification basic characteristic value, the original default survey data extraction ratio of the platform on sale is used as the satisfaction survey data extraction ratio of the platform on sale.

4. The method for optimizing a consumer satisfaction survey based on big data according to claim 1, wherein: The specific steps of determining the available evaluation value of the data of each source object include: Obtaining consumption activity data for each source object, the consumption activity data including: return rate, number of repeat purchases, product purchase conversion rate, negative review rate, and purchase frequency within a first preset time period; Obtain a preset consumer activity data benchmark set and compare it with the consumer activity data of each source object to obtain the data availability assessment value of each source object; The preset consumer activity data benchmark set includes: return rate verification value, repurchase number verification value, product purchase conversion rate verification value, negative review rate verification value and purchase frequency verification value; The data of each source object may be evaluated to quantify the consumption activity data of each source object and serve as a basis for analyzing consumer satisfaction.

5. The method for optimizing a consumer satisfaction survey based on big data according to claim 4, characterized in that: The method for determining the available evaluation value of the data of each source object is: Where, E i Indicates the data availability evaluation value of the i-th source object, i represents the number of the source object, i=1,2,...,n, n represents the number of source objects, r i represents the return rate of the i-th source object in the first preset time period, r1 represents the return rate verification value, e represents the base of the natural logarithm, f i represents the number of repurchases of the i-th source object within the first preset time period, f1 represents the repurchase verification value, c i represents the product purchase conversion rate of the i-th source object in the first preset time period, c1 represents the product purchase conversion rate verification value, q i represents the negative review rate of the i-th source object in the first preset time period, q1 represents the negative review rate verification value, and p i represents the purchase frequency of the i-th source object in the first preset time period, p1 represents the purchase frequency verification value, α1 represents the return rate influence weight of the source object in the first preset time period, α2 represents the repurchase number influence weight of the source object in the first preset time period, α3 represents the commodity purchase conversion rate influence weight of the source object in the first preset time period, α4 represents the negative review rate influence weight of the source object in the first preset time period, and α5 represents the purchase frequency influence weight of the source object in the first preset time period.

6. The method for optimizing a consumer satisfaction survey based on big data according to claim 1, characterized in that: The second valid data set is obtained by screening, and the specific steps include: Obtain a data availability assessment threshold in the database and compare it with the data availability assessment value of each source object. If the data availability assessment value of a source object is above the data availability assessment threshold, the satisfaction questionnaire data of the source object is determined to be second valid data. If the data availability assessment value of the source object is less than the data availability assessment threshold, the satisfaction questionnaire data of the source object is determined to be invalid data and deleted. Summarizing all the second valid data to obtain a second valid data set; The second valid data set includes each target source object and satisfaction questionnaire data of each target source object.

7. The method for optimizing a consumer satisfaction survey based on big data according to claim 6, characterized in that: The specific steps of obtaining the satisfaction survey evaluation label of the product include: Based on the second valid data set and processed according to the historical sales performance data of the product, a satisfaction survey evaluation value of the product is obtained; Obtain the satisfaction survey evaluation threshold of the product in the database and compare it with the satisfaction survey evaluation value of the product. If the satisfaction survey evaluation value of the product is above the satisfaction survey evaluation threshold, the satisfaction survey evaluation label of the product is defined as a high-satisfaction product. If the satisfaction survey evaluation value of the product is less than the satisfaction survey evaluation threshold, the satisfaction survey evaluation label of the product is defined as a low-satisfaction product.

8. The method for optimizing a consumer satisfaction survey based on big data according to claim 7, characterized in that: The specific method for conducting survey retrospective optimization prompts based on the product satisfaction survey evaluation label is as follows: If the satisfaction survey evaluation label of the product is defined as a low-satisfaction product, then based on the second valid data set obtained by screening, the satisfaction questionnaire data of each target source object is collected, and the satisfaction questionnaire data includes the evaluation text of each target source object on the product in each dimension; Based on the evaluation texts of each target source object on the product in each dimension, a number of evaluation keywords of each target source object on the product in each dimension are counted, and according to the positive keyword collection and negative keyword collection preset in the database, the total number of positive evaluation keywords and the total number of negative evaluation keywords of each target source object on the product in each dimension are obtained through mapping and matching; According to the total number of positive evaluation keywords and negative evaluation keywords of each target source object for the product in each dimension, the text evaluation value of the product in each dimension is processed and the text evaluation values are arranged in order from small to large to obtain the investigation and tracing arrangement order of the product, and the investigation and tracing arrangement order of the product is transmitted to the central controller.

9. The method for optimizing a consumer satisfaction survey based on big data according to claim 8, characterized in that: The method of obtaining a satisfaction survey evaluation value of a product based on the second valid data set and processing the product's historical sales performance data is as follows: Based on the satisfaction questionnaire data of each target source object in the second valid data set, the satisfaction questionnaire data of all target source objects are summarized to obtain a satisfaction questionnaire survey text of the product, and a number of keywords corresponding to the product are counted from the satisfaction questionnaire data, and the total number of positive evaluation keywords and the total number of negative evaluation keywords corresponding to the satisfaction questionnaire survey text of the product are obtained by mapping and matching with the positive keyword collection and the negative keyword collection preset in the database; The satisfaction survey evaluation value of the product is obtained by processing the total number of positive evaluation keywords and the total number of negative evaluation keywords corresponding to the product satisfaction questionnaire text and the historical sales performance data of the product; The historical sales performance data of the product includes the total number of historical positive evaluation keywords per capita for the product and the total number of historical negative evaluation keywords per capita for the product; The satisfaction survey evaluation value of the product is used to quantitatively analyze the second valid data set and is used as a basis for analyzing product satisfaction.

10. A consumer satisfaction survey optimization system based on big data, applying the consumer satisfaction survey optimization method based on big data according to any one of claims 1 to 9, characterized in that: include: Data acquisition module, data extraction module, data evaluation module, data screening module and label processing module; The data acquisition module is used to count the current online platforms where the product is sold, recorded as each selling platform, and obtain the basic data of each selling platform for preprocessing to obtain the satisfaction survey data extraction ratio of each selling platform; The data extraction module is used to collect the satisfaction questionnaire data sets of each selling platform, extract data according to the satisfaction survey data extraction ratio of each selling platform, and mark the extracted satisfaction questionnaire data sets as the first valid data sets; The data evaluation module is used to count the source objects corresponding to the first valid data set, and to count the consumption activity data of each source object for processing, and to determine the available evaluation value of the data of each source object; The data screening module is configured to obtain a second valid data set by screening based on the data availability evaluation value of each source object and the first valid data set; The label processing module is used to perform comprehensive processing based on the second valid data set and the historical sales performance data of the product to obtain a satisfaction survey evaluation label of the product, and to provide survey retrospective optimization prompts based on the satisfaction survey evaluation label of the product.

Images