E-commerce return data feedback method and system based on big data

Abstract

The present invention relates to the field of e-commerce, and a method and system for feedback of e-commerce return data based on big data, comprising: receiving a product return questionnaire, performing a keyword analysis operation on the questionnaire text, obtaining multiple return keywords, retrieving multiple reference product information in a product database using multiple reference numbers, obtaining a keyword feedback index using a reference evaluation set and a keyword sequence, calculating a time feedback index according to a product time mean, a reference time mean, and a user's use time, calculating a location feedback index according to user location coordinates, a product delivery location, and a reference delivery location, obtaining a merchant feedback index, calculating a data feedback score according to a keyword feedback index, a time feedback index, a location feedback index, and a merchant feedback index, displaying a product display list on a pre-built shopping user terminal, and completing feedback on e-commerce return data. The present invention can enhance the user's shopping experience and improve the user stickiness and competitiveness of an e-commerce shopping platform.

Description

Technical Field

[0001] The present invention relates to the field of e-commerce, and in particular to an e-commerce return data feedback method, system, electronic device and computer-readable storage medium based on big data. Background Art

[0002] With the rapid development of e-commerce, product returns have become an important part of consumers' purchasing experience. Many consumers will return products after purchasing them because the products do not meet their expectations or have quality problems.

[0003] Currently, when users make returns, shopping platforms usually collect return questionnaires to understand the reasons for customers’ returns, and then feedback the return questionnaires to merchants, urging them to adjust product quality or sales strategies in a timely manner.

[0004] Although the above method can realize the utilization of e-commerce return data, it does not take into account the personal needs of users when returning goods. Because most users initiate returns because the goods do not meet their expectations, the user's demand for such goods does not disappear after the return, and they still need to re-purchase other goods of the same type as the returned goods. However, at this time, the shopping platform does not respond in time to actively screen or provide users with goods that better meet the user's expectations, which may cause the goods purchased by users again to still not meet their expectations, resulting in frequent returns, which greatly affects the user's shopping experience, or the user cannot find the goods they need in time, so they switch to other shopping platforms or shopping methods, which leads to a decrease in the user stickiness of the platform. Therefore, there is an urgent need for an intelligent return data analysis method that analyzes the user's return questionnaire based on big data, and recommends goods that meet the user's needs to the user in a timely manner, thereby improving the user's shopping experience and improving the user stickiness and competitiveness of the e-commerce shopping platform. Summary of the invention

[0005] The present invention provides an e-commerce return data feedback method based on big data and a computer-readable storage medium, the main purpose of which is to enhance the user's shopping experience and improve the user stickiness and competitiveness of the e-commerce shopping platform.

[0006] To achieve the above purpose, the present invention provides an e-commerce return data feedback method based on big data, comprising:

[0007] Receiving a product return questionnaire, wherein the product return questionnaire includes: a product number and a questionnaire text, and using the product number to retrieve product information from a pre-built product database, wherein the product information includes: a product evaluation area, a product name, and a product shipping location;

[0008] Obtain the arrival time and the return time, and calculate the user usage time based on the arrival time and the return time, wherein the user usage time is the absolute difference between the arrival time and the return time, and retrieve multiple return information from the pre-built logistics database based on the product number, wherein the return information includes: the product arrival time and the product return time, and calculate the average product time based on the multiple return information;

[0009] Extracting the latest evaluation set from the product evaluation area according to a preset time range, performing a negative review extraction operation on the latest evaluation set to obtain a negative review evaluation set, wherein the negative review evaluation set includes a plurality of latest negative reviews;

[0010] Perform keyword analysis on the questionnaire text to obtain multiple return keywords, and use the multiple return keywords and the negative review set to obtain a keyword sequence;

[0011] Using the product name and a pre-built search system to obtain multiple reference numbers, and using the multiple reference numbers to retrieve multiple reference product information in a product database, wherein the reference numbers and the reference product information correspond one to one, and the reference product information includes: a reference evaluation set and a reference shipping location;

[0012] For each of the multiple reference product information, perform the following operations:

[0013] The keyword feedback index is obtained by using the reference evaluation set and the keyword sequence, the reference time mean is obtained based on the reference number corresponding to the reference product information, and the time feedback index is calculated according to the product time mean, the reference time mean and the user usage time;

[0014] Obtain the user's location coordinates, calculate the location feedback index based on the user's location coordinates, the product's shipping location and the reference shipping location, and obtain the merchant's feedback index using the reference number corresponding to the reference product information;

[0015] Calculate the data feedback score based on the keyword feedback index, time feedback index, location feedback index and merchant feedback index;

[0016] The data feedback scores are aggregated to obtain multiple data feedback scores, and the multiple reference product information are sorted in descending order according to the data feedback scores corresponding to the reference product information to obtain a product information sequence, and a product display list is constructed based on the product information sequence. The product display list is displayed on a pre-built shopping user terminal to complete the feedback of the e-commerce return data.

[0017] Optionally, calculating the average time of the product according to the multiple return information includes:

[0018] Perform the following operations for each of the multiple return information:

[0019] The commodity usage time is calculated based on the commodity arrival time and the commodity return time, where the commodity usage time is the absolute difference between the commodity arrival time and the commodity return time;

[0020] The commodity usage time is summarized to obtain multiple commodity usage times, and the commodity time average is calculated according to the multiple commodity usage times, wherein the commodity time average is the average value of the multiple commodity usage times.

[0021] Optionally, the keyword analysis operation is performed on the questionnaire text to obtain multiple return keywords, including:

[0022] Perform character removal operation on the questionnaire text to obtain Chinese character text;

[0023] Use the pre-built word segmentation tool to perform word segmentation on the Chinese text to obtain multiple Chinese character keywords;

[0024] Obtain a stop word list, wherein the stop word list includes multiple stop words, and perform the following operations on each of the multiple Chinese character keywords:

[0025] Using the Chinese character keyword to search in the stop word list, if there is no stop word identical to the Chinese character keyword in the stop word list, marking the Chinese character keyword as a primary keyword;

[0026] Summarize primary keywords to obtain a primary keyword set;

[0027] Use a word segmentation tool to perform part-of-speech tagging on each primary keyword in the primary keyword set to obtain a tagged keyword set, use the word segmentation tool to perform pronoun removal operations and conjunction removal operations on the tagged keyword set to obtain a removed keyword set, perform a part-of-speech removal operation on the removed keyword set to obtain multiple return keywords.

[0028] Optionally, the step of obtaining a keyword sequence by using a plurality of return keywords and a negative review set includes:

[0029] Perform the following operations for each of the multiple return keywords:

[0030] Perform the following operations on each of the multiple latest negative reviews:

[0031] Determine whether there is a return keyword in the latest negative review, and if there is a return keyword in the latest negative review, mark the latest negative review as a critical negative review;

[0032] Summarize the key negative reviews to obtain the key negative review set, determine the key number based on the key negative review set, and determine the standard number based on the negative review evaluation set, where the key number is the number of key negative reviews in the key negative review set, and the standard number is the number of the latest negative reviews in the negative review evaluation set;

[0033] Calculate the keyword weight based on the key number and the standard number, where the keyword weight is the ratio of the key number to the standard number;

[0034] The keyword weights are aggregated to obtain multiple keyword weights, and the multiple return keywords are sorted in descending order according to the keyword weights corresponding to the return keywords to obtain a keyword sequence.

[0035] Optionally, the obtaining the keyword feedback index by using the reference evaluation set and the keyword sequence includes:

[0036] Performing a classification operation on the reference evaluation set to obtain a reference positive evaluation set and a reference negative evaluation set, wherein the reference positive evaluation set includes multiple reference positive evaluations, and the reference negative evaluation set includes multiple reference negative evaluations;

[0037] Extract the return keywords from the keyword sequence in sequence, and perform the following operations on the extracted return keywords:

[0038] The weight of the reference positive review is calculated based on multiple reference positive reviews and return keywords, and the weight of the reference negative review is calculated based on multiple reference negative reviews and return keywords;

[0039] Summarize the weights of positive reviews for reference to obtain multiple weights of positive reviews for reference, and summarize the weights of negative reviews for reference to obtain multiple weights of negative reviews for reference;

[0040] The keyword feedback index is calculated based on the keyword sequence, multiple reference positive review weights, and multiple reference negative review weights. The calculation formula is as follows:

[0041] ;

[0042] in, and The first The reference weight of positive reviews and the reference weight of negative reviews corresponding to the return keywords are: is the keyword feedback index, is the number of return keywords in the keyword sequence, It is the position of the return keyword in the keyword sequence.

[0043] Optionally, calculating the time feedback index according to the product time average, the reference time average and the user usage time includes:

[0044] The time feedback index is calculated using the following formula:

[0045] ;

[0046] in, is the time feedback index, is the reference time mean, is the commodity time mean, Use time for users.

[0047] Optionally, calculating the location feedback index according to the user location coordinates, the product shipping location and the reference shipping location includes:

[0048] Perform coordinate positioning operation on the product shipping location to obtain the product shipping coordinates, and obtain reference shipping coordinates based on the reference shipping location;

[0049] Calculate the original delivery distance based on the user's location coordinates and the product delivery coordinates, calculate the reference delivery distance based on the user's location coordinates and the reference delivery coordinates, and calculate the product difference distance based on the product delivery coordinates and the reference delivery coordinates;

[0050] The position feedback index is calculated using the original delivery distance, reference delivery distance, and product difference distance. The calculation formula is as follows:

[0051] ;

[0052] in, is the position feedback index, is the original shipping distance, For reference shipping distance, is the product difference distance, is the hyperbolic tangent function.

[0053] Optionally, obtaining the merchant feedback index by using the reference number corresponding to the reference product information includes:

[0054] Using the reference number, the merchant transaction volume and merchant operation time are read from the product database;

[0055] Retrieving multiple reference return information in a logistics database using the reference number, and confirming a reference return quantity based on the multiple reference return information, wherein the reference return quantity is the number of reference return information in the multiple reference return information;

[0056] The merchant feedback index is calculated based on the merchant transaction volume, reference return volume and merchant operation time. The calculation formula is as follows:

[0057] ;

[0058] in, For merchant feedback index, For reference return quantity, is the merchant transaction volume, The operating time for the merchant, is the preset standard transaction volume, The default standard operation time. is a natural constant.

[0059] Optionally, the calculating of the data feedback score according to the keyword feedback index, the time feedback index, the location feedback index and the merchant feedback index includes:

[0060] The data feedback score is calculated using the following formula:

[0061] ;

[0062] in, Score the data feedback.

[0063] To achieve the above purpose, the present invention also provides an e-commerce return data feedback system based on big data, comprising:

[0064] A return information analysis module is used to receive a product return questionnaire, wherein the product return questionnaire includes: a product number and a questionnaire text, retrieve product information from a pre-built product database using the product number, wherein the product information includes: a product evaluation area, a product name and a product shipping place, obtain arrival time and return time, calculate user usage time based on the arrival time and return time, wherein the user usage time is the absolute difference between the arrival time and the return time, retrieve multiple return information from a pre-built logistics database based on the product number, wherein the return information includes: a product arrival time and a product return time, calculate the product time average based on multiple return information, extract the latest evaluation set from the product evaluation area according to a preset time range, perform a negative review extraction operation on the latest evaluation set, obtain a negative review evaluation set, wherein the negative review evaluation set includes multiple latest negative reviews, perform a keyword analysis operation on the questionnaire text, obtain multiple return keywords, and obtain a keyword sequence using multiple return keywords and the negative review evaluation set;

[0065] A reference product acquisition module is used to obtain multiple reference numbers using the product name and a pre-built search system, and retrieve multiple reference product information in the product database using the multiple reference numbers, wherein the reference numbers correspond to the reference product information one by one, and the reference product information includes: a reference evaluation set and a reference shipping place;

[0066] A product information evaluation module is used to perform the following operations on each of the multiple reference product information: obtain a keyword feedback index using a reference evaluation set and a keyword sequence, obtain a reference time mean based on a reference number corresponding to the reference product information, calculate a time feedback index based on the product time mean, the reference time mean and the user usage time, obtain user location coordinates, calculate a location feedback index based on the user location coordinates, the product shipping place and the reference shipping place, obtain a merchant feedback index using the reference number corresponding to the reference product information, and calculate a data feedback score based on the keyword feedback index, the time feedback index, the location feedback index and the merchant feedback index;

[0067] The product list display module is used to summarize the data feedback scores, obtain multiple data feedback scores, sort multiple reference product information in descending order according to the data feedback scores corresponding to the reference product information, obtain a product information sequence, build a product display list based on the product information sequence, and display the product display list on a pre-built shopping user terminal to complete the feedback of e-commerce return data.

[0068] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising:

[0069] a memory storing at least one instruction; and

[0070] The processor executes the instructions stored in the memory to implement the above-mentioned e-commerce return data feedback method based on big data.

[0071] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned e-commerce return data feedback method based on big data.

[0072] The present invention is to solve the problem described in the background technology. The present invention receives a product return questionnaire, wherein the product return questionnaire includes: a product number and a questionnaire text, and uses the product number to retrieve product information from a pre-constructed product database, wherein the product information includes: a product evaluation area, a product name, and a product shipping location. It can be seen that the embodiment of the present invention receives the product return questionnaire and uses the product number to retrieve product information from a pre-constructed product database, which is convenient for subsequent analysis of the product information, confirming the keyword sequence, and then obtaining the arrival time and the return time, and calculating the user usage time according to the arrival time and the return time, wherein the user usage time is the absolute difference between the arrival time and the return time, and the user usage time is calculated based on the product number in the pre-constructed logistics database. A plurality of return information is retrieved from the data, wherein the return information includes: the arrival time of the goods and the return time of the goods, and the average time of the goods is calculated according to the plurality of return information. It can be seen that the embodiment of the present invention can quantify the use of the goods by calculating the user's use time, thereby providing a basis for the subsequent calculation of the time feedback index, and extracting the latest evaluation set from the product evaluation area according to the preset time range, performing a negative review extraction operation on the latest evaluation set, and obtaining a negative review evaluation set, wherein the negative review evaluation set includes a plurality of the latest negative reviews, and it can be seen that the embodiment of the present invention extracts the negative review evaluation set, which is convenient for the subsequent confirmation of the weight of the return keyword according to the negative review evaluation set, performing a keyword analysis operation on the questionnaire text, and obtaining a plurality of return keywords, and using the plurality of return keywords and the negative review evaluation set It can be seen that the embodiment of the present invention successfully extracts multiple return keywords by analyzing the questionnaire text, and confirms the weight corresponding to each return keyword according to the negative review set, and then constructs a keyword sequence, uses the product name and the pre-built search system to obtain multiple reference numbers, and uses the multiple reference numbers to retrieve multiple reference product information in the product database, wherein the reference number and the reference product information correspond one to one, and the reference product information includes: a reference evaluation set and a reference shipping place. It can be seen that the embodiment of the present invention provides consumers with more similar product choices by retrieving reference product information, increases the possibility of users finding satisfactory products, and improves users' shopping experience, and performs a weighting of each reference product in the multiple reference product information. The following operations are performed for product information: the keyword feedback index is obtained by using the reference evaluation set and the keyword sequence, the reference time mean is obtained based on the reference number corresponding to the reference product information, and the time feedback index is calculated according to the product time mean, the reference time mean and the user usage time. It can be seen that the embodiment of the present invention quantifies the possibility of product problems and the severity of product problems by calculating the keyword feedback index and the time feedback index, and provides a reference for the subsequent calculation of data feedback scores. The user location coordinates are obtained, and the location feedback index is calculated according to the user location coordinates, the product shipping place and the reference shipping place. The merchant feedback index is obtained by using the reference number corresponding to the reference product information. It can be seen that the embodiment of the present invention calculates the location feedback index and the merchant feedback index.The difference between other products and the products returned by users and the credibility of the products are quantified, providing a reference for the subsequent calculation of data feedback scores. According to the keyword feedback index, time feedback index, location feedback index and merchant feedback index, the data feedback score is calculated. It can be seen that the embodiment of the present invention evaluates the return risk of the product and the degree of compliance with user expectations more comprehensively by integrating the feedback index of multiple dimensions, thereby facilitating the subsequent provision of more accurate product recommendations to consumers, improving the user's shopping experience, and improving the user stickiness and competitiveness of the e-commerce shopping platform. The data feedback scores are summarized to obtain multiple data feedback scores, and multiple reference product information is sorted in order from high to low according to the data feedback scores corresponding to the reference product information to obtain a product information sequence. A product display list is constructed based on the product information sequence, and the product display list is displayed on the pre-constructed shopping user terminal to complete the feedback of the e-commerce return data. It can be seen that the embodiment of the present invention sorts multiple reference product information by data feedback scores and displays them to users in order, thereby improving the probability of users finding products that meet their expectations, enhancing the user's shopping experience, and improving the user stickiness and competitiveness of the platform. Therefore, the present invention can improve the user's shopping experience and improve the user stickiness and competitiveness of the e-commerce shopping platform. , BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 A schematic diagram of a flow chart of an e-commerce return data feedback method based on big data provided by an embodiment of the present invention;

[0074] Figure 2 A functional module diagram of an e-commerce return data feedback system based on big data provided by an embodiment of the present invention;

[0075] Figure 3 A schematic diagram of the structure of an electronic device for implementing the big data-based e-commerce return data feedback method provided in one embodiment of the present invention.

[0076] Description of reference numerals:

[0077] 1. Electronic device; 10. Processor; 11. Memory; 12. Bus.

[0078] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0079] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0080] The embodiment of the present application provides an e-commerce return data feedback method based on big data. The execution subject of the e-commerce return data feedback method based on big data includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the e-commerce return data feedback method based on big data can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0081] Reference Figure 1 FIG. 1 is a flow chart of an e-commerce return data feedback method based on big data provided by an embodiment of the present invention. In this embodiment, the e-commerce return data feedback method based on big data includes:

[0082] S1. Receive a product return questionnaire, wherein the product return questionnaire includes: a product number and a questionnaire text, and use the product number to retrieve product information in a pre-built product database, wherein the product information includes: a product evaluation area, a product name, and a product shipping location.

[0083] It should be explained that the product return questionnaire is filled out by users of the shopping platform when they initiate a return, and the product return questionnaire includes the product number and questionnaire text. The product number refers to the number of the returned product, and the questionnaire text refers to all the words and symbols in the product return questionnaire. The shopping platform refers to the website or APP used for shopping, such as Taobao and JD.com.

[0084] It should be understood that each product on the shopping platform corresponds to a number, which is used to identify the product.

[0085] For example, Xiao Zhang is a user of a shopping platform. Because the product he purchased did not meet Xiao Zhang's expectations, Xiao Zhang initiated a return and refund on the shopping platform. When returning and refunding, the shopping platform will pop up a questionnaire for Xiao Zhang to explain the reason for the return and refund. This questionnaire is the product return questionnaire, and the product number is the number of the product that Xiao Zhang returns and refunds. If all the text and symbols filled out by Xiao Zhang in the questionnaire are: "The surface is defective and inconvenient to use", "The surface is defective and inconvenient to use" is the questionnaire text.

[0086] It is understandable that the product database is a database pre-built by the shopping platform, which is used to store product information of all products in the shopping platform, and each product information corresponds to a product number, so that the corresponding product information can be retrieved using the product number. The technology of retrieving product information from the pre-built product database using the product number is an existing technology and will not be repeated here.

[0087] It should be explained that product information includes: product evaluation area, product name and product delivery place. The product evaluation area is a collection of multiple product reviews. Product evaluation refers to the evaluation of the quality, performance, appearance, price, service and other aspects of the product by users based on their shopping experience, usually including text descriptions, pictures or videos, etc. And the types of product evaluation are divided into: good reviews, bad reviews and ordinary reviews. Product name refers to the name of the product, and product delivery place refers to the delivery address of the product.

[0088] It should be understood that when users write reviews for products on shopping platforms, there will be options (good reviews, bad reviews and ordinary reviews) or star ratings (4 stars and 5 stars are considered good reviews, 3 stars are considered ordinary reviews, and 1 star and 2 stars are considered bad reviews) on the review writing interface. Users can make their choices, so the types of product reviews are divided into: good reviews, bad reviews and ordinary reviews.

[0089] S2. Obtain the arrival time and the return time, and calculate the user usage time based on the arrival time and the return time, wherein the user usage time is the absolute difference between the arrival time and the return time. Retrieve multiple return information from a pre-built logistics database based on the product number, wherein the return information includes: product arrival time and product return time, and calculate the average product time based on multiple return information.

[0090] It should be explained that the arrival time refers to the time when the goods are signed for, and the return time refers to the time when the user initiates the return. For example, the arrival time is obtained by querying the logistics order of the returned goods when they were purchased before, and the return time is obtained by recording the time when the user initiates the return.

[0091] It is understandable that the logistics database is a database pre-built by the shopping platform, which is used to store the logistics information of all products in the shopping platform. For example, the logistics information of a certain product is: the product was collected by SF Express at 08:00:00 on December 1, the product was shipped from a business point in Nanjing at 15:00:00 on December 1, the product arrived at the Jieyang Transshipment Center in Guangdong at 10:00:00 on December 2, and the product was sent to the Shanwei Distribution Center in Guangdong at 14:00:00 on December 2. The goods were delivered to a certain business department at 20:00:00 on December 2, the goods were signed for by the user at 08:00:00 on December 3, the goods were returned by the user at 08:00:00 on December 5, and the goods were collected by SF Express at 10:00:00 on December 5... The goods were returned to a certain business point in Nanjing at 10:00:00 on December 7, that is, the arrival time of the goods was 08:00:00 on December 3, and the return time of the goods was 08:00:00 on December 5. Return information refers to the logistics information of the returned goods. The retrieval of multiple return information in the pre-constructed logistics database based on the product number refers to: retrieving all logistics information of the goods corresponding to the product number in the logistics database, and screening out multiple logistics information of the returned goods from all the logistics information, and using them as return information. And the technology of retrieving multiple return information in the pre-constructed logistics database based on the product number is a prior art, which will not be repeated here.

[0092] Specifically, the calculation of the average time of the product based on the multiple return information includes:

[0093] Perform the following operations for each of the multiple return information:

[0094] The commodity usage time is calculated based on the commodity arrival time and the commodity return time, where the commodity usage time is the absolute difference between the commodity arrival time and the commodity return time;

[0095] The commodity usage time is summarized to obtain multiple commodity usage times, and the commodity time average is calculated according to the multiple commodity usage times, wherein the commodity time average is the average value of the multiple commodity usage times.

[0096] S3. Extract the latest evaluation set from the product evaluation area according to a preset time range, perform a negative review extraction operation on the latest evaluation set, and obtain a negative review evaluation set, wherein the negative review evaluation set includes a plurality of latest negative reviews.

[0097] Exemplarily, if the preset time range is within 30 days, if today's date is December 30, all product reviews from December 1 to December 30 are extracted from the product review area and summarized to obtain the latest review set.

[0098] It is understandable that the operation of extracting negative reviews from the latest evaluation set refers to: extracting all product reviews of the type of negative reviews from the latest evaluation set and summarizing them to obtain a negative review set. The latest negative review is the product review in the negative review set.

[0099] S4. Perform keyword analysis on the questionnaire text to obtain multiple return keywords, and use the multiple return keywords and the negative review set to obtain a keyword sequence.

[0100] Specifically, the operation of performing keyword analysis on the questionnaire text to obtain multiple return keywords includes:

[0101] Perform character elimination on the questionnaire text to obtain Chinese character text;

[0102] Use a pre-built word segmentation tool to perform word segmentation on the Chinese character text to obtain multiple Chinese character keywords;

[0103] Obtain a stop word list, where the stop word list includes multiple stop words, and perform the following operations on each Chinese character keyword in the multiple Chinese character keywords:

[0104] Use the Chinese character keyword to search in the stop word list. If there is no stop word in the stop word list that is the same as the Chinese character keyword, then label the Chinese character keyword as a primary keyword;

[0105] Summarize the primary keywords to obtain a primary keyword set;

[0106] Use the word segmentation tool to perform part-of-speech tagging on each primary keyword in the primary keyword set to obtain a tagged keyword set, use the word segmentation tool to perform pronoun elimination and conjunction elimination operations on the tagged keyword set to obtain an eliminated keyword set, and perform part-of-speech removal operations on the eliminated keyword set to obtain multiple return keywords.

[0107] It is understandable that the operation of performing character elimination on the questionnaire text refers to: eliminating all symbols, numeric characters, and English characters from the questionnaire text, and only retaining the Chinese character characters in the questionnaire text.

[0108] It should be explained that the word segmentation tool is a tool implemented by Python that can split a continuous text into individual independent words. Preferably, the jieba library in Python is used as the word segmentation tool. The stop word list is a set including multiple stop words. Stop words refer to words that do not have practical meanings in a text, such as "de", "shi", "zai", etc. Optionally, the Baidu stop word list is used as the stop word list.

[0109] Exemplarily, if a certain Chinese character keyword is "crack", and there is no stop word "crack" in the stop word list, then the Chinese character keyword is labeled as a primary keyword.

[0110] It should be understood that the use of a word segmentation tool to perform part-of-speech tagging on each primary keyword in the primary keyword set to obtain a tagged keyword set means: using a word segmentation tool to identify the part-of-speech of each primary keyword in the primary keyword set, and tagging the part-of-speech after the keyword to obtain the tagged keywords, summarizing the tagged keywords, and obtaining a tagged keyword set. And the technology of using a word segmentation tool to identify the part-of-speech of each primary keyword in the primary keyword set is a prior art and will not be repeated here. The use of a word segmentation tool to perform a pronoun removal operation and a conjunction removal operation on the tagged keyword set to obtain a removed keyword set means: using a word segmentation tool to remove the tagged keywords in the tagged keyword set that are tagged with pronouns or conjunctions, and using the removed tagged keyword set as the removed keyword set. The use of a part-of-speech removal operation on the removed keyword set to obtain multiple return keywords means: removing the part-of-speech marked after each tagged keyword in the removed keyword set, summarizing the tagged keywords after the part-of-speech removal, and obtaining multiple return keywords.

[0111] Exemplarily, the primary keywords in the primary keyword set are: "I", "use", "difficult", and "but". The jieba.posseg module in the word segmentation tool is used to mark the part of speech of each primary keyword, and "I-pronoun", "use-verb", "difficult-adjective", and "but-conjunction" are obtained. The conjunctions and pronouns in these four words are removed to obtain "use-verb" and "difficult-adjective". Then the marked parts of speech in "use-verb" and "difficult-adjective" are removed to obtain "use" and "difficult", that is, "use" and "difficult" are two return keywords.

[0112] In detail, the method of obtaining a keyword sequence by using a plurality of return keywords and a negative review set includes:

[0113] Perform the following operations for each of the multiple return keywords:

[0114] Perform the following operations on each of the multiple latest negative reviews:

[0115] Determine whether there is a return keyword in the latest negative review, and if there is a return keyword in the latest negative review, mark the latest negative review as a critical negative review;

[0116] Summarize the key negative reviews to obtain the key negative review set, determine the key number based on the key negative review set, and determine the standard number based on the negative review evaluation set, where the key number is the number of key negative reviews in the key negative review set, and the standard number is the number of the latest negative reviews in the negative review evaluation set;

[0117] Calculate the keyword weight based on the key number and the standard number, where the keyword weight is the ratio of the key number to the standard number;

[0118] The keyword weights are aggregated to obtain multiple keyword weights, and the multiple return keywords are sorted in descending order according to the keyword weights corresponding to the return keywords to obtain a keyword sequence.

[0119] For example, if the return keyword is "fabric" and the latest negative review is "the fabric of this dress is too bad", since the return keyword "fabric" exists in the latest negative review, the latest negative review is identified as a critical negative review.

[0120] It should be understood that the embodiment of the present invention determines the keyword weight of the return keyword by analyzing the negative reviews in the product evaluation area. If the same words as the return keyword appear in most of the negative reviews in the product evaluation area, it means that most users encounter the same problems as the return keyword when using the product. Therefore, the return keyword has a higher keyword weight, is ranked higher in the keyword sequence, and has a higher weight in the subsequent calculation of the keyword feedback index.

[0121] S5. Use the product name and a pre-built search system to obtain multiple reference numbers, and use the multiple reference numbers to retrieve multiple reference product information in the product database, wherein the reference numbers and the reference product information correspond one to one, and the reference product information includes: a reference evaluation set and a reference shipping location.

[0122] It should be explained that the search system is a search engine pre-built by the shopping platform. When a product name is entered in the search engine, the search engine will output multiple other products related to the product name, and multiple reference numbers will be obtained by reading the numbers corresponding to the multiple other products.

[0123] For example, if the product name is pure cotton thermal underwear, enter "pure cotton thermal underwear" in the search box of Taobao, and the search engine will automatically retrieve multiple products of the product category pure cotton thermal underwear, and obtain multiple reference numbers by reading the numbers of these products. It should be noted that the number of other products retrieved by the search engine may reach thousands. In order to facilitate the subsequent analysis speed of multiple reference product information, only some of the other products retrieved by the search engine should be retained under the consideration of the computing power of the computer. Preferably, 50 reference numbers are obtained by reading the numbers corresponding to the first 50 other products output by the search engine.

[0124] It should be understood that the method of using multiple reference numbers to retrieve multiple reference product information in a product database is the same as the method of using product numbers to retrieve product information in a pre-built product database, and will not be repeated here.

[0125] S6. Perform the following operations on each of the multiple reference product information: use the reference evaluation set and the keyword sequence to obtain the keyword feedback index, obtain the reference time mean based on the reference number corresponding to the reference product information, and calculate the time feedback index based on the product time mean, the reference time mean and the user usage time.

[0126] In detail, the method of obtaining the keyword feedback index by using the reference evaluation set and the keyword sequence includes:

[0127] Performing a classification operation on the reference evaluation set to obtain a reference positive evaluation set and a reference negative evaluation set, wherein the reference positive evaluation set includes multiple reference positive evaluations, and the reference negative evaluation set includes multiple reference negative evaluations;

[0128] Extract the return keywords from the keyword sequence in sequence, and perform the following operations on the extracted return keywords:

[0129] The weight of the reference positive review is calculated based on multiple reference positive reviews and return keywords, and the weight of the reference negative review is calculated based on multiple reference negative reviews and return keywords;

[0130] Summarize the weights of positive reviews for reference to obtain multiple weights of positive reviews for reference, and summarize the weights of negative reviews for reference to obtain multiple weights of negative reviews for reference;

[0131] The keyword feedback index is calculated based on the keyword sequence, multiple reference positive review weights, and multiple reference negative review weights. The calculation formula is as follows:

[0132] ;

[0133] in, and The first The reference weight of positive reviews and the reference weight of negative reviews corresponding to the return keywords are: is the keyword feedback index, is the number of return keywords in the keyword sequence, It is the position of the return keyword in the keyword sequence.

[0134] It should be explained that the classification operation performed on the reference evaluation set to obtain the reference good evaluation set and the reference bad evaluation set means: extracting all the bad product evaluations from the reference evaluation set and summarizing them to obtain the reference bad evaluation set, extracting all the good product evaluations from the reference evaluation set and summarizing them to obtain the reference good evaluation set. The reference bad evaluations are product evaluations in the reference bad evaluation set. The reference good evaluations are product evaluations in the reference good evaluation set.

[0135] It can be understood that the method of calculating the reference positive review weight based on multiple reference positive reviews and return keywords and the method of calculating the reference negative review weight based on multiple reference negative reviews and return keywords are the same as the method of obtaining keyword weights using key negative review sets and return keywords, and will not be repeated here.

[0136] It should be understood that the keyword feedback index reflects the possibility that the product corresponding to the reference product information and the product returned by the user have the same problems by evaluating the correlation between the return keywords and the reference positive reviews in the reference positive review set and the reference negative reviews in the reference negative review set. The higher the keyword feedback index, the lower the possibility that the product corresponding to the reference product information and the product returned by the user have the same problems. Therefore, the product corresponding to the reference product information should be recommended to the user.

[0137] It is understandable that the method of obtaining the reference time average based on the reference number corresponding to the reference product information is the same as the method of obtaining the product time average using the product number, and will not be repeated here.

[0138] In detail, the time feedback index is calculated based on the product time average, the reference time average and the user usage time, including:

[0139] The time feedback index is calculated using the following formula:

[0140] ;

[0141] in, is the time feedback index, is the reference time mean, is the commodity time mean, Use time for users.

[0142] It should be understood that the more obvious and unacceptable the problems of a product are, the quicker the user will initiate a return and the shorter the average reference time will be. Therefore, the time feedback index reflects the obviousness or severity of the problems of the product corresponding to the reference product information. The higher the time feedback index, the less obvious or severe the problems of the product corresponding to the reference product information, and therefore the product corresponding to the reference product information should be recommended to users.

[0143] S7. Obtain the user location coordinates, calculate the location feedback index according to the user location coordinates, the product shipping location and the reference shipping location, and obtain the merchant feedback index using the reference number corresponding to the reference product information.

[0144] Optionally, the longitude and latitude coordinates of the location of the user's mobile phone are located using a global positioning system, and the longitude and latitude coordinates are used as the user's location coordinates.

[0145] In detail, the location feedback index is calculated based on the user location coordinates, the product shipping location and the reference shipping location, including:

[0146] Perform coordinate positioning operation on the product shipping location to obtain the product shipping coordinates, and obtain reference shipping coordinates based on the reference shipping location;

[0147] Calculate the original delivery distance based on the user's location coordinates and the product delivery coordinates, calculate the reference delivery distance based on the user's location coordinates and the reference delivery coordinates, and calculate the product difference distance based on the product delivery coordinates and the reference delivery coordinates;

[0148] The position feedback index is calculated using the original delivery distance, reference delivery distance, and product difference distance. The calculation formula is as follows:

[0149] ;

[0150] in, is the position feedback index, is the original shipping distance, For reference shipping distance, is the product difference distance, is the hyperbolic tangent function.

[0151] For example, the place of shipment of the goods is a business point in Nanjing, so the location of the business point is found in Baidu Maps, and the longitude and latitude coordinates of the location are read using the Baidu Maps's coordinate picking system, and the longitude and latitude coordinates are the goods shipment coordinates. The method of obtaining the reference shipment coordinates based on the reference shipment place is the same as the method of performing a coordinate positioning operation on the goods shipment place to obtain the goods shipment coordinates, and will not be repeated here.

[0152] It is understandable that the original shipping distance refers to the distance in reality between the position corresponding to the user location coordinates and the position corresponding to the product shipping coordinates, and the technology for calculating the original shipping distance based on the user location coordinates and the product shipping coordinates is a prior art and will not be described in detail here. The method for calculating the reference shipping distance based on the user location coordinates and the reference shipping coordinates and the method for calculating the product difference distance based on the product shipping coordinates and the reference shipping coordinates are the same as the method for calculating the original shipping distance based on the user location coordinates and the product shipping coordinates and will not be described in detail here.

[0153] It should be understood that the farther the shipping place of the product corresponding to the reference product information is from the shipping place of the product returned by the user, the greater the difference between the manufacturing environment and raw materials of the product corresponding to the reference product information and the product returned by the user may be. Therefore, the position feedback index reflects the degree of difference between the product corresponding to the reference product information and the product returned by the user. The higher the position feedback index, the higher the degree of difference between the product corresponding to the reference product information and the product returned by the user, and the lower the possibility that the product corresponding to the reference product information and the product returned by the user have the same problem. Therefore, the product corresponding to the reference product information should be recommended to the user.

[0154] In detail, the obtaining of the merchant feedback index by using the reference number corresponding to the reference product information includes:

[0155] Using the reference number, the merchant transaction volume and merchant operation time are read from the product database;

[0156] Retrieving multiple reference return information in a logistics database using the reference number, and confirming a reference return quantity based on the multiple reference return information, wherein the reference return quantity is the number of reference return information in the multiple reference return information;

[0157] The merchant feedback index is calculated based on the merchant transaction volume, reference return volume and merchant operation time. The calculation formula is as follows:

[0158] ;

[0159] in, For merchant feedback index, For reference return quantity, is the merchant transaction volume, The operating time for the merchant, is the preset standard transaction volume, The default standard operation time. is a natural constant.

[0160] It should be explained that the merchant transaction volume refers to the total number of all transaction orders of the product corresponding to the reference number, and the merchant operation time refers to the number of days from the date of listing of the product corresponding to the reference number to the current date. For example, the merchant time of the product is October 20, 2024, and the current time is December 20, 2024, so the merchant operation time is 61 days. Optionally, the standard transaction volume is set to 10,000 orders, and the marked operation time is set to 30 days.

[0161] It should be understood that each commodity corresponding to the merchant transaction volume and merchant operation time corresponds to a commodity number, and the merchant transaction volume and merchant operation time corresponding to all commodities are stored in the commodity database. Therefore, the merchant transaction volume and merchant operation time can be retrieved from the commodity database using the reference number, and the technology of retrieving the merchant transaction volume and merchant operation time from the commodity database using the reference number is existing technology.

[0162] It is understandable that the method of retrieving multiple reference return information in the logistics database using the reference number is the same as the method of retrieving multiple return information in the pre-built logistics database based on the product number, and will not be repeated here.

[0163] It should be understood that the longer a product is on the shelves and the more transaction orders it receives, the more popular the product is with users and the higher the credibility of the product. Therefore, the merchant feedback index reflects the credibility or popularity of the product corresponding to the reference product information. The higher the merchant feedback index, the higher the credibility or popularity of the product corresponding to the reference product information, and therefore the product corresponding to the reference product information should be recommended to users.

[0164] S8. Calculate the data feedback score based on the keyword feedback index, time feedback index, location feedback index and merchant feedback index.

[0165] In detail, the data feedback score is calculated according to the keyword feedback index, time feedback index, location feedback index and merchant feedback index, including:

[0166] The data feedback score is calculated using the following formula:

[0167] ;

[0168] in, Score the data feedback.

[0169] It should be understood that the data feedback score is a value that comprehensively considers the possibility of the same problem occurring in the product, the severity of the problem occurring in the product, the degree of difference between the product and the product returned by the user, and the credibility of the product. It reflects the degree to which the product meets user expectations. The higher the data feedback score, the more the product meets user expectations.

[0170] S9. Summarize the data feedback scores to obtain multiple data feedback scores, sort the multiple reference product information in descending order according to the data feedback scores corresponding to the reference product information to obtain a product information sequence, build a product display list based on the product information sequence, display the product display list on a pre-built shopping user terminal, and complete the feedback of the e-commerce return data.

[0171] In detail, the step of constructing a product display list based on the product information sequence includes:

[0172] Perform the following operations on each reference product information in the product information sequence:

[0173] Retrieving a product display page from a product database using a reference number corresponding to the reference product information;

[0174] The product display pages are aggregated to obtain a plurality of product display pages, and the plurality of product display pages are sorted according to the order of the reference product information corresponding to the product display pages in the product information sequence to obtain a product display list.

[0175] For example, a product display page is a page that is pre-created by the merchant of a product when the merchant lists the product on the shopping platform, which contains a thumbnail of the product and some information (such as the product name and product introduction). In addition, a shopping platform APP used by users usually displays multiple product display pages on one interface to facilitate user selection. When the user clicks on the product display page on the shopping user end, the shopping platform will automatically jump to the product details page, which is convenient for the user to browse the pictures and details of the product on the product details page and make a purchase. The product details page is a page that contains multiple pictures and detailed information of the product.

[0176] It should be understood that each product display page corresponds to a product number, and the product display pages corresponding to all products are stored in the product database. Therefore, the product display page can be retrieved from the product database using the reference number, and the technology of retrieving the product display page from the product database using the reference number is prior art.

[0177] It is understandable that the shopping user terminal refers to the APP of the shopping platform on the mobile phone. For example, when the customer returns the product, the next time the shopping platform APP is opened, the product display list will be displayed on the interface of the shopping platform APP, so that the user can quickly choose a more satisfactory product according to the product display list.

[0178] The present invention is to solve the problem described in the background technology. The present invention receives a product return questionnaire, wherein the product return questionnaire includes: a product number and a questionnaire text, and uses the product number to retrieve product information from a pre-constructed product database, wherein the product information includes: a product evaluation area, a product name, and a product shipping location. It can be seen that the embodiment of the present invention receives the product return questionnaire and uses the product number to retrieve product information from a pre-constructed product database, which is convenient for subsequent analysis of the product information, confirming the keyword sequence, and then obtaining the arrival time and the return time, and calculating the user usage time according to the arrival time and the return time, wherein the user usage time is the absolute difference between the arrival time and the return time, and the user usage time is calculated based on the product number in the pre-constructed logistics database. A plurality of return information is retrieved from the data, wherein the return information includes: the arrival time of the goods and the return time of the goods, and the average time of the goods is calculated according to the plurality of return information. It can be seen that the embodiment of the present invention can quantify the use of the goods by calculating the user's use time, thereby providing a basis for the subsequent calculation of the time feedback index, and extracting the latest evaluation set from the product evaluation area according to the preset time range, performing a negative review extraction operation on the latest evaluation set, and obtaining a negative review evaluation set, wherein the negative review evaluation set includes a plurality of the latest negative reviews, and it can be seen that the embodiment of the present invention extracts the negative review evaluation set, which is convenient for the subsequent confirmation of the weight of the return keyword according to the negative review evaluation set, performing a keyword analysis operation on the questionnaire text, and obtaining a plurality of return keywords, and using the plurality of return keywords and the negative review evaluation set It can be seen that the embodiment of the present invention successfully extracts multiple return keywords by analyzing the questionnaire text, and confirms the weight corresponding to each return keyword according to the negative review set, and then constructs a keyword sequence, uses the product name and the pre-built search system to obtain multiple reference numbers, and uses the multiple reference numbers to retrieve multiple reference product information in the product database, wherein the reference number and the reference product information correspond one to one, and the reference product information includes: a reference evaluation set and a reference shipping place. It can be seen that the embodiment of the present invention provides consumers with more similar product choices by retrieving reference product information, increases the possibility of users finding satisfactory products, and improves users' shopping experience, and performs a weighting of each reference product in the multiple reference product information. The following operations are performed for product information: the keyword feedback index is obtained by using the reference evaluation set and the keyword sequence, the reference time mean is obtained based on the reference number corresponding to the reference product information, and the time feedback index is calculated according to the product time mean, the reference time mean and the user usage time. It can be seen that the embodiment of the present invention quantifies the possibility of product problems and the severity of product problems by calculating the keyword feedback index and the time feedback index, and provides a reference for the subsequent calculation of data feedback scores. The user location coordinates are obtained, and the location feedback index is calculated according to the user location coordinates, the product shipping place and the reference shipping place. The merchant feedback index is obtained by using the reference number corresponding to the reference product information. It can be seen that the embodiment of the present invention calculates the location feedback index and the merchant feedback index.The difference between other products and the products returned by users and the credibility of the products are quantified, providing a reference for the subsequent calculation of data feedback scores. According to the keyword feedback index, time feedback index, location feedback index and merchant feedback index, the data feedback score is calculated. It can be seen that the embodiment of the present invention evaluates the return risk of the product and the degree of compliance with user expectations more comprehensively by integrating the feedback index of multiple dimensions, thereby facilitating the subsequent provision of more accurate product recommendations to consumers, improving the user's shopping experience, and improving the user stickiness and competitiveness of the e-commerce shopping platform. The data feedback scores are summarized to obtain multiple data feedback scores, and multiple reference product information is sorted in order from high to low according to the data feedback scores corresponding to the reference product information to obtain a product information sequence. A product display list is constructed based on the product information sequence, and the product display list is displayed on the pre-constructed shopping user terminal to complete the feedback of the e-commerce return data. It can be seen that the embodiment of the present invention sorts multiple reference product information by data feedback scores and displays them to users in order, thereby improving the probability of users finding products that meet their expectations, enhancing the user's shopping experience, and improving the user stickiness and competitiveness of the platform. Therefore, the present invention can improve the user's shopping experience and improve the user stickiness and competitiveness of the e-commerce shopping platform. ,

[0179] like Figure 2 , which is a functional module diagram of an e-commerce return data feedback system based on big data provided by one embodiment of the present invention.

[0180] The e-commerce return data feedback system 100 based on big data of the present invention can be installed in an electronic device. According to the functions implemented, the e-commerce return data feedback system 100 based on big data can include a return information analysis module 101, a reference product acquisition module 102, a product information evaluation module 103 and a product list display module 104. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0181] The return information analysis module 101 is used to receive a product return questionnaire, wherein the product return questionnaire includes: a product number and a questionnaire text, retrieve product information from a pre-constructed product database using the product number, wherein the product information includes: a product evaluation area, a product name and a product shipping place, obtain arrival time and return time, calculate user usage time based on the arrival time and return time, wherein the user usage time is the absolute difference between the arrival time and the return time, retrieve multiple return information from a pre-constructed logistics database based on the product number, wherein the return information includes: a product arrival time and a product return time, calculate the product time average based on multiple return information, extract the latest evaluation set from the product evaluation area according to a preset time range, perform a negative review extraction operation on the latest evaluation set to obtain a negative review evaluation set, wherein the negative review evaluation set includes multiple latest negative reviews, perform a keyword analysis operation on the questionnaire text to obtain multiple return keywords, and obtain a keyword sequence using the multiple return keywords and the negative review evaluation set;

[0182] The reference product acquisition module 102 is used to obtain multiple reference numbers using the product name and a pre-built search system, and retrieve multiple reference product information in the product database using the multiple reference numbers, wherein the reference numbers correspond to the reference product information one by one, and the reference product information includes: a reference evaluation set and a reference shipping place;

[0183] The product information evaluation module 103 is used to perform the following operations on each of the multiple reference product information: obtain a keyword feedback index using a reference evaluation set and a keyword sequence, obtain a reference time mean based on a reference number corresponding to the reference product information, calculate a time feedback index based on the product time mean, the reference time mean and the user usage time, obtain a user location coordinate, calculate a location feedback index based on the user location coordinate, the product shipping place and the reference shipping place, obtain a merchant feedback index using the reference number corresponding to the reference product information, and calculate a data feedback score based on the keyword feedback index, the time feedback index, the location feedback index and the merchant feedback index;

[0184] The product list display module 104 is used to summarize the data feedback scores to obtain multiple data feedback scores, sort the multiple reference product information in descending order according to the data feedback scores corresponding to the reference product information to obtain a product information sequence, build a product display list based on the product information sequence, and display the product display list on a pre-built shopping user terminal to complete the feedback of the e-commerce return data.

[0185] In detail, each module in the big data-based e-commerce return data feedback system 100 in the embodiment of the present invention is used in the same manner as above. Figure 1The technical means are the same as the e-commerce return data feedback method based on big data described in, and can produce the same technical effects, so I will not go into details here.

[0186] like Figure 3 , is a schematic diagram of the structure of an electronic device for implementing an e-commerce return data feedback method based on big data provided by an embodiment of the present invention.

[0187] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as an e-commerce return data feedback method program based on big data.

[0188] Among them, the memory 11 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (for example: SD or DX memory, etc.), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as a mobile hard disk of the electronic device 1. In other embodiments, the memory 11 can also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 1. Further, the memory 11 also includes an internal storage unit of the electronic device 1 and an external storage device. The memory 11 can not only be used to store application software and various types of data installed in the electronic device 1, such as the code of the e-commerce return data feedback method program based on big data, but also can be used to temporarily store data that has been output or is to be output.

[0189] The processor 10 may be composed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips, etc. The processor 10 is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device, and executes or executes programs or modules stored in the memory 11 (such as an e-commerce return data feedback method program based on big data, etc.), and calls data stored in the memory 11 to execute various functions of the electronic device 1 and process data.

[0190] The bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 may be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0191] Figure 3 Only an electronic device with components is shown, and those skilled in the art will understand that Figure 3 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0192] For example, although not shown, the electronic device 1 may also include a power source (such as a battery) for supplying power to various components. Preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0193] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0194] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visual user interface.

[0195] The e-commerce return data feedback method program based on big data stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can achieve:

[0196] Receiving a product return questionnaire, wherein the product return questionnaire includes: a product number and a questionnaire text, and using the product number to retrieve product information from a pre-built product database, wherein the product information includes: a product evaluation area, a product name, and a product shipping location;

[0197] Obtain the arrival time and the return time, and calculate the user usage time based on the arrival time and the return time, wherein the user usage time is the absolute difference between the arrival time and the return time, and retrieve multiple return information from the pre-built logistics database based on the product number, wherein the return information includes: the product arrival time and the product return time, and calculate the average product time based on the multiple return information;

[0198] Extracting the latest evaluation set from the product evaluation area according to a preset time range, performing a negative review extraction operation on the latest evaluation set to obtain a negative review evaluation set, wherein the negative review evaluation set includes a plurality of latest negative reviews;

[0199] Perform keyword analysis on the questionnaire text to obtain multiple return keywords, and use the multiple return keywords and the negative review set to obtain a keyword sequence;

[0200] Using the product name and a pre-built search system to obtain multiple reference numbers, and using the multiple reference numbers to retrieve multiple reference product information in a product database, wherein the reference numbers and the reference product information correspond one to one, and the reference product information includes: a reference evaluation set and a reference shipping location;

[0201] For each of the multiple reference product information, perform the following operations:

[0202] The keyword feedback index is obtained by using the reference evaluation set and the keyword sequence, the reference time mean is obtained based on the reference number corresponding to the reference product information, and the time feedback index is calculated according to the product time mean, the reference time mean and the user usage time;

[0203] Obtain the user's location coordinates, calculate the location feedback index based on the user's location coordinates, the product's shipping location and the reference shipping location, and obtain the merchant's feedback index using the reference number corresponding to the reference product information;

[0204] Calculate the data feedback score based on the keyword feedback index, time feedback index, location feedback index and merchant feedback index;

[0205] The data feedback scores are aggregated to obtain multiple data feedback scores, and the multiple reference product information are sorted in descending order according to the data feedback scores corresponding to the reference product information to obtain a product information sequence, and a product display list is constructed based on the product information sequence. The product display list is displayed on a pre-built shopping user terminal to complete the feedback of the e-commerce return data.

[0206] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to Figures 1 to 3 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0207] Furthermore, if the module / unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0208] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, the computer program can implement:

[0209] Receiving a product return questionnaire, wherein the product return questionnaire includes: a product number and a questionnaire text, and using the product number to retrieve product information from a pre-built product database, wherein the product information includes: a product evaluation area, a product name, and a product shipping location;

[0210] Obtain the arrival time and the return time, and calculate the user usage time based on the arrival time and the return time, wherein the user usage time is the absolute difference between the arrival time and the return time, and retrieve multiple return information from the pre-built logistics database based on the product number, wherein the return information includes: the product arrival time and the product return time, and calculate the average product time based on the multiple return information;

[0211] Extracting the latest evaluation set from the product evaluation area according to a preset time range, performing a negative review extraction operation on the latest evaluation set to obtain a negative review evaluation set, wherein the negative review evaluation set includes a plurality of latest negative reviews;

[0212] Perform keyword analysis on the questionnaire text to obtain multiple return keywords, and use the multiple return keywords and the negative review set to obtain a keyword sequence;

[0213] Using the product name and a pre-built search system to obtain multiple reference numbers, and using the multiple reference numbers to retrieve multiple reference product information in a product database, wherein the reference numbers and the reference product information correspond one to one, and the reference product information includes: a reference evaluation set and a reference shipping location;

[0214] For each of the multiple reference product information, perform the following operations:

[0215] The keyword feedback index is obtained by using the reference evaluation set and the keyword sequence, the reference time mean is obtained based on the reference number corresponding to the reference product information, and the time feedback index is calculated according to the product time mean, the reference time mean and the user usage time;

[0216] Obtain the user's location coordinates, calculate the location feedback index based on the user's location coordinates, the product's shipping location and the reference shipping location, and obtain the merchant's feedback index using the reference number corresponding to the reference product information;

[0217] Calculate the data feedback score based on the keyword feedback index, time feedback index, location feedback index and merchant feedback index;

[0218] The data feedback scores are aggregated to obtain multiple data feedback scores, and the multiple reference product information are sorted in descending order according to the data feedback scores corresponding to the reference product information to obtain a product information sequence, and a product display list is constructed based on the product information sequence. The product display list is displayed on a pre-built shopping user terminal to complete the feedback of the e-commerce return data.

[0219] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, and actual implementation may have other division methods.

[0220] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0221] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0222] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0223] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for e-commerce return data feedback based on big data, characterized in that: The method comprises: Receiving a product return questionnaire, wherein the product return questionnaire includes: a product number and a questionnaire text, and using the product number to retrieve product information from a pre-built product database, wherein the product information includes: a product evaluation area, a product name, and a product shipping location; Obtain the arrival time and the return time, and calculate the user usage time based on the arrival time and the return time, wherein the user usage time is the absolute difference between the arrival time and the return time, and retrieve multiple return information from the pre-built logistics database based on the product number, wherein the return information includes: the product arrival time and the product return time, and calculate the average product time based on the multiple return information; Extracting the latest evaluation set from the product evaluation area according to a preset time range, performing a negative review extraction operation on the latest evaluation set to obtain a negative review evaluation set, wherein the negative review evaluation set includes a plurality of latest negative reviews; Perform keyword analysis on the questionnaire text to obtain multiple return keywords, and use the multiple return keywords and the negative review set to obtain a keyword sequence; Using the product name and a pre-built search system to obtain multiple reference numbers, and using the multiple reference numbers to retrieve multiple reference product information in a product database, wherein the reference numbers and the reference product information correspond one to one, and the reference product information includes: a reference evaluation set and a reference shipping location; For each of the multiple reference product information, perform the following operations: The keyword feedback index is obtained by using the reference evaluation set and the keyword sequence, the reference time mean is obtained based on the reference number corresponding to the reference product information, and the time feedback index is calculated according to the product time mean, the reference time mean and the user usage time; Obtain the user's location coordinates, calculate the location feedback index based on the user's location coordinates, the product's shipping location and the reference shipping location, and obtain the merchant's feedback index using the reference number corresponding to the reference product information; Calculate the data feedback score based on the keyword feedback index, time feedback index, location feedback index and merchant feedback index; The data feedback scores are aggregated to obtain multiple data feedback scores, and the multiple reference product information are sorted in descending order according to the data feedback scores corresponding to the reference product information to obtain a product information sequence, and a product display list is constructed based on the product information sequence. The product display list is displayed on a pre-built shopping user terminal to complete the feedback of the e-commerce return data.

2. The e-commerce return data feedback method based on big data as claimed in claim 1, characterized in that: The calculating the average time of the goods according to the multiple return information includes: Perform the following operations for each of the multiple return information: The commodity usage time is calculated based on the commodity arrival time and the commodity return time, where the commodity usage time is the absolute difference between the commodity arrival time and the commodity return time; The commodity usage time is summarized to obtain multiple commodity usage times, and the commodity time average is calculated according to the multiple commodity usage times, wherein the commodity time average is the average value of the multiple commodity usage times.

3. The e-commerce return data feedback method based on big data as claimed in claim 2, characterized in that: The keyword analysis operation is performed on the questionnaire text to obtain multiple return keywords, including: Perform character removal operation on the questionnaire text to obtain Chinese character text; Use the pre-built word segmentation tool to perform word segmentation on the Chinese text to obtain multiple Chinese character keywords; Obtain a stop word list, wherein the stop word list includes multiple stop words, and perform the following operations on each of the multiple Chinese character keywords: Using the Chinese character keyword to search in the stop word list, if there is no stop word identical to the Chinese character keyword in the stop word list, marking the Chinese character keyword as a primary keyword; Summarize primary keywords to obtain a primary keyword set; Use a word segmentation tool to perform part-of-speech tagging on each primary keyword in the primary keyword set to obtain a tagged keyword set, use the word segmentation tool to perform pronoun removal operations and conjunction removal operations on the tagged keyword set to obtain a removed keyword set, perform a part-of-speech removal operation on the removed keyword set to obtain multiple return keywords.

4. The e-commerce return data feedback method based on big data as claimed in claim 3, characterized in that: The method of obtaining a keyword sequence by using a plurality of return keywords and a negative review set includes: Perform the following operations for each of the multiple return keywords: Perform the following operations on each of the multiple latest negative reviews: Determine whether there is a return keyword in the latest negative review, and if there is a return keyword in the latest negative review, mark the latest negative review as a critical negative review; Summarize the key negative reviews to obtain the key negative review set, determine the key number based on the key negative review set, and determine the standard number based on the negative review evaluation set, where the key number is the number of key negative reviews in the key negative review set, and the standard number is the number of the latest negative reviews in the negative review evaluation set; Calculate the keyword weight based on the key number and the standard number, where the keyword weight is the ratio of the key number to the standard number; The keyword weights are aggregated to obtain multiple keyword weights, and the multiple return keywords are sorted in descending order according to the keyword weights corresponding to the return keywords to obtain a keyword sequence.

5. The e-commerce return data feedback method based on big data as claimed in claim 4, characterized in that: The method of obtaining a keyword feedback index by using a reference evaluation set and a keyword sequence includes: Performing a classification operation on the reference evaluation set to obtain a reference positive evaluation set and a reference negative evaluation set, wherein the reference positive evaluation set includes multiple reference positive evaluations, and the reference negative evaluation set includes multiple reference negative evaluations; Extract the return keywords from the keyword sequence in sequence, and perform the following operations on the extracted return keywords: The weight of the reference positive review is calculated based on multiple reference positive reviews and return keywords, and the weight of the reference negative review is calculated based on multiple reference negative reviews and return keywords; Summarize the weights of positive reviews for reference to obtain multiple weights of positive reviews for reference, and summarize the weights of negative reviews for reference to obtain multiple weights of negative reviews for reference; The keyword feedback index is calculated based on the keyword sequence, multiple reference positive review weights, and multiple reference negative review weights. The calculation formula is as follows: ； in, and The first The reference weight of positive reviews and the reference weight of negative reviews corresponding to the return keywords are: is the keyword feedback index, is the number of return keywords in the keyword sequence, It is the position of the return keyword in the keyword sequence.

6. The e-commerce return data feedback method based on big data as claimed in claim 5, characterized in that: The time feedback index is calculated based on the product time average, the reference time average and the user usage time, including: The time feedback index is calculated using the following formula: ； in, is the time feedback index, is the reference time mean, is the commodity time mean, Use time for users.

7. The e-commerce return data feedback method based on big data as claimed in claim 6, characterized in that: The calculating of the location feedback index according to the user location coordinates, the product shipping location and the reference shipping location includes: Perform coordinate positioning operation on the product shipping location to obtain the product shipping coordinates, and obtain reference shipping coordinates based on the reference shipping location; Calculate the original delivery distance based on the user's location coordinates and the product delivery coordinates, calculate the reference delivery distance based on the user's location coordinates and the reference delivery coordinates, and calculate the product difference distance based on the product delivery coordinates and the reference delivery coordinates; The position feedback index is calculated using the original delivery distance, reference delivery distance, and product difference distance. The calculation formula is as follows: ； in, is the position feedback index, is the original shipping distance, For reference shipping distance, is the product difference distance, is the hyperbolic tangent function.

8. The e-commerce return data feedback method based on big data as claimed in claim 7, characterized in that: The obtaining of the merchant feedback index by using the reference number corresponding to the reference product information includes: Using the reference number, the merchant transaction volume and merchant operation time are read from the product database; Retrieving multiple reference return information in a logistics database using the reference number, and confirming a reference return quantity based on the multiple reference return information, wherein the reference return quantity is the number of reference return information in the multiple reference return information; The merchant feedback index is calculated based on the merchant transaction volume, reference return volume and merchant operation time. The calculation formula is as follows: ； in, For merchant feedback index, For reference return quantity, is the merchant transaction volume, The operating time for the merchant, is the preset standard transaction volume, The default standard operation time. is a natural constant.

9. The method for e-commerce return data feedback based on big data as claimed in claim 8, characterized in that: The data feedback score is calculated based on the keyword feedback index, time feedback index, location feedback index and merchant feedback index, including: The data feedback score is calculated using the following formula: ； in, Score the data feedback.

10. An e-commerce return data feedback system based on big data, characterized in that: The system comprises: A return information analysis module is used to receive a product return questionnaire, wherein the product return questionnaire includes: a product number and a questionnaire text, retrieve product information from a pre-built product database using the product number, wherein the product information includes: a product evaluation area, a product name and a product shipping place, obtain arrival time and return time, calculate user usage time based on the arrival time and return time, wherein the user usage time is the absolute difference between the arrival time and the return time, retrieve multiple return information from a pre-built logistics database based on the product number, wherein the return information includes: a product arrival time and a product return time, calculate the product time average based on multiple return information, extract the latest evaluation set from the product evaluation area according to a preset time range, perform a negative review extraction operation on the latest evaluation set, obtain a negative review evaluation set, wherein the negative review evaluation set includes multiple latest negative reviews, perform a keyword analysis operation on the questionnaire text, obtain multiple return keywords, and obtain a keyword sequence using multiple return keywords and the negative review evaluation set; A reference product acquisition module is used to obtain multiple reference numbers using the product name and a pre-built search system, and retrieve multiple reference product information in the product database using the multiple reference numbers, wherein the reference numbers correspond to the reference product information one by one, and the reference product information includes: a reference evaluation set and a reference shipping place; A product information evaluation module is used to perform the following operations on each of the multiple reference product information: obtain a keyword feedback index using a reference evaluation set and a keyword sequence, obtain a reference time mean based on a reference number corresponding to the reference product information, calculate a time feedback index based on the product time mean, the reference time mean and the user usage time, obtain user location coordinates, calculate a location feedback index based on the user location coordinates, the product shipping place and the reference shipping place, obtain a merchant feedback index using the reference number corresponding to the reference product information, and calculate a data feedback score based on the keyword feedback index, the time feedback index, the location feedback index and the merchant feedback index; The product list display module is used to summarize the data feedback scores, obtain multiple data feedback scores, sort multiple reference product information in descending order according to the data feedback scores corresponding to the reference product information, obtain a product information sequence, build a product display list based on the product information sequence, and display the product display list on a pre-built shopping user terminal to complete the feedback of e-commerce return data.

Images