Clothing sales data intelligent analysis system and method

Abstract

The invention discloses a clothing sales data intelligent analysis system and method, and relates to the technical field of electric digital data processing. The intelligent clothing sales data analysis system comprises a clothing sales data processing module, an imitation analysis module, a return analysis module and a real-time monitoring module. According to the invention, the clothing information of the clothing product is collected to obtain the clothing sales data, the clothing sales data of the clothing product and the competitive product is detected to obtain the imitation risk coefficient, and then the interactive behavior of the user and the clothing sales data is analyzed to obtain the malicious order returning threshold. And finally, clothing stock is automatically supplemented and orders are managed by monitoring the clothing product stock level and the sales speed in real time, so that the effect of improving the clothing sales data security in the intelligent analysis process is achieved, and the problem of low clothing sales data security in the intelligent analysis process in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic digital data processing, and in particular to a clothing sales data intelligent analysis system and method. Background Art

[0002] As consumers' pursuit of personalized, quality and fashionable clothing continues to increase, the clothing industry has gradually entered a stage of transformation and upgrading. At the same time, with the development of the Internet, the proportion of online sales channels has continued to increase, and new e-commerce models such as live streaming and instant retail have a significant driving effect on the growth of online consumption. Clothing sales data reflects the overall situation of clothing sales and is an important basis for formulating sales strategies, optimizing inventory management and improving customer experience. Especially driven by e-commerce, China's clothing region has maintained a high-speed growth trend. Through the analysis of sales data, companies can gain an in-depth understanding of regional trends and consumer behavior, and grasp the changes in regional demand and consumer preferences. Sales data provides accurate operating data and key indicators to help companies understand sales, inventory, production efficiency and other aspects, so as to carry out refined operational planning and decision-making. By analyzing sales data and customer behavior, companies can formulate more effective regional marketing strategies, such as limited-time discounts, new product launches, membership systems, etc., to improve sales efficiency and customer satisfaction. With the development of big data and artificial intelligence technology, big data technology and artificial intelligence algorithms are used to conduct in-depth analysis and mining of sales data, while providing targeted decision support. Reports and charts can be customized according to the specific needs of the company to meet the personalized needs of different users.

[0003] In the prior art, artificial intelligence-based data analysis technology is used to capture features based on consumers' historical purchase records and browsing records, and the correlation features of the semantic understanding features of consumers' historical purchase records and the semantic understanding features of browsing records are mined to comprehensively recommend consumers' preferred clothing types, thereby realizing personalized clothing sales.

[0004] For example, the invention patent announcement with announcement number: CN117349344B announces a method and system for intelligent collection of product sales data based on big data, including: extracting data representation vectors from multiple sales training data sets marked with the same product classification indication information in the sales training database through a target sales data analysis model to obtain a training representation vector cluster under the same product classification indication information; pre-deploying multiple preset sub-cluster centers for the training representation vector cluster, and determining the preset sub-cluster centers corresponding to each of the multiple training representation vectors in the training representation vector cluster; updating the multiple preset sub-cluster centers through the commonality measurement results between the multiple training representation vectors and the corresponding preset sub-cluster centers to obtain no less than one target sub-cluster center under the same product classification indication information; determining the model training error through the target sub-cluster centers under multiple product classification indication information and the training representation vectors corresponding to each target sub-cluster center, and training the target sales data analysis model through the model training error.

[0005] For example, a method and system for clothing matching is announced in the invention patent with announcement number: CN108920828B, including: obtaining product data of clothing through a first terminal, the product data including color data, style data, size data and occasion data; the first terminal obtains personal data of the user, the personal data including image data; the first terminal matches target clothing for the user in the clothing according to the personal data and product data, the target clothing including clothing items and combinations of clothing items; wherein the first terminal performs data statistics based on the personal data of the user obtained, and confirms the user's demand; performs statistics according to the occasion data in the personal data, and determines the product scene in the product attributes of the target clothing; confirms the category and quantity of items required for the occasion; and according to the matching requirements of the occasion, obtains the combination of different items of the target clothing, and generates a wardrobe report, and the wardrobe report includes the item name and quantity of the target clothing required for different occasions.

[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems: In the existing technology, data analysis tools and technologies are used to conduct in-depth mining and analysis of data. The analysis can include consumer behavior analysis, regional trend prediction, supply chain optimization and other aspects. By analyzing consumer purchase history, browsing history and other data, it is possible to predict consumer future needs and provide a basis for personalized recommendations. At the same time, by integrating data such as purchase quantity and sales quantity, it is possible to predict future sales trends and help companies optimize inventory and supply chain management. However, since many clothing companies lack strict data security management systems and with the popularization of e-commerce and mobile Internet, users may maliciously return goods during the clothing sales process, competing products may steal product promotional images, and even malicious reviews may occur. There is a problem of low security of clothing sales data in the intelligent analysis process. Summary of the invention

[0007] The embodiments of the present application solve the problem of low security of clothing sales data in the intelligent analysis process in the prior art by providing a clothing sales data intelligent analysis system and method, thereby improving the security of clothing sales data in the intelligent analysis process.

[0008] The clothing sales data intelligent analysis system provided by the embodiment of the present application includes a clothing sales data processing module, an imitation analysis module, a return order analysis module and a real-time monitoring module; wherein the clothing sales data processing module is used to obtain clothing sales data by collecting clothing information of clothing products, the clothing information is used to record information of clothing products, and the clothing sales data represents the data of clothing products after preprocessing of numerical data corresponding to the clothing information; the imitation analysis module is used to detect clothing sales data of clothing products and competing products to obtain a counterfeit risk coefficient, and the counterfeit risk coefficient is used to measure the degree of counterfeit risk of clothing products; the return order analysis module is used to obtain a malicious return order threshold by analyzing the interaction behavior between users and clothing sales data, and the malicious return order threshold is used to identify the malicious return order behavior of users during the sales process of clothing products; the real-time monitoring module is used to automatically replenish clothing inventory and manage orders by real-time monitoring of clothing product inventory levels and sales speeds.

[0009] Furthermore, the specific method for obtaining the imitation risk coefficient is as follows: obtain the number of clothing products and corresponding competitive products, and number the clothing products and competitive products; obtain a content similarity evaluation value by analyzing the text similarity between the clothing products and competitive products; extract text keywords of the clothing products and competitive products and count the number of text keyword overlaps to obtain a keyword repetition rate; compare the similarity between the text grammatical structure and the syntactic structure of the clothing products and competitive products to obtain structural similarity; obtain a weight coefficient of imitation data from a preset database, the imitation data including a content similarity evaluation value, a keyword repetition rate and a structural similarity, and the weight coefficient including a first weight coefficient, a second weight coefficient and a third weight coefficient; normalize the imitation data, and obtain the imitation risk coefficient in combination with the weight coefficient.

[0010] Furthermore, the imitation risk coefficient is specifically calculated using the following formula: ; In the formula, represents the imitation risk coefficient, Indicates the number of the clothing product. , represents the total number of clothing products, Indicates Content similarity evaluation value of a set of clothing products, Indicates The keyword repetition rate of a set of clothing products, Indicates The structural similarity of a set of clothing products, represents the first weight coefficient, represents the second weight coefficient, Represents the third weight coefficient.

[0011] Furthermore, the specific method for obtaining the malicious order return threshold is as follows: within a preset time period, the average number of order returns and the average number of orders for each customer are counted to obtain the corresponding order return frequency and order frequency; within a preset time period, the amount of each refund and the corresponding purchase amount of each user are counted, and the ratio of the total refund amount of the user to the total purchase amount is analyzed to obtain the refund ratio; the correction factor of the refund ratio is obtained from a preset database; the refund ratio threshold is obtained and the malicious order return data is normalized to obtain the malicious order return threshold, and the malicious order return data includes order frequency, order return frequency and refund ratio.

[0012] Furthermore, the calculation formula of the malicious return order threshold is as follows: ; In the formula, Indicates the malicious chargeback threshold for a preset time period. Indicates the frequency of return orders in the preset time period. Indicates the order frequency for a preset time period, Indicates the refund ratio. Indicates the refund ratio threshold. A correction factor indicating the refund ratio.

[0013] Furthermore, the intelligent analysis method for clothing sales data includes the following steps: obtaining clothing sales data by collecting clothing information of clothing products; detecting clothing sales data of clothing products and competing products to obtain imitation risk coefficients, wherein the imitation risk coefficients are used to measure the degree of imitation risk of clothing products; obtaining malicious order return thresholds by analyzing the interactive behavior between users and clothing sales data, wherein the malicious order return thresholds are used to identify malicious order return behaviors of users during the sales process of clothing products; and automatically replenishing clothing inventory and managing orders by real-time monitoring of clothing product inventory levels and sales speeds.

[0014] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. The clothing sales data is obtained by collecting clothing information of clothing products, and the clothing sales data of clothing products and competing products are detected to obtain the imitation risk coefficient. Then, the malicious order return threshold is obtained by analyzing the interaction between users and clothing sales data. Finally, the clothing inventory is automatically replenished and orders are managed by real-time monitoring of clothing product inventory levels and sales speeds, thereby improving the security of clothing sales data in the intelligent analysis process, and effectively solving the problem of low security of clothing sales data in the intelligent analysis process in the prior art.

[0015] 2. By obtaining the number of clothing products and corresponding competitive products, and then analyzing the text similarity between clothing products and competitive products to obtain the content similarity evaluation value, extracting the text keywords of clothing products and competitive products and counting the number of text keyword overlaps to obtain the keyword repetition rate, and then comparing the similarity of the text grammatical structure and syntactic structure of clothing products and competitive products to obtain the structural similarity, and finally combining the weight coefficient to obtain the imitation risk coefficient, thereby realizing the numerical evaluation of imitation risk, and then realizing a more accurate evaluation of imitation risk.

[0016] 3. By counting the average number of returns and the average number of orders for each customer within a preset time period, the corresponding return frequency and order frequency are obtained. Then, the refund amount and the corresponding purchase amount of each user are counted within the preset time period, and the ratio of the total refund amount to the total purchase amount of the user is analyzed to obtain the refund ratio. Then, the correction factor of the refund ratio is obtained from the preset database. Finally, the refund ratio threshold is obtained and the malicious return data is normalized to obtain the malicious return threshold, thereby realizing the numerical evaluation of malicious return orders, and then realizing a more accurate evaluation of malicious return orders. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of the structure of the clothing sales data intelligent analysis system provided in the embodiment of the present application; Figure 2 A statistical chart showing changes in the malicious return order threshold provided in an embodiment of the present application; Figure 3 A flowchart of real-time monitoring provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] The embodiments of the present application solve the problem of low security of clothing sales data in the prior art by providing a clothing sales data intelligent analysis system and method. The clothing sales data is obtained by collecting clothing information of clothing products, and the clothing sales data of clothing products and competing products are detected to obtain the imitation risk coefficient. Then, the malicious order return threshold is obtained by analyzing the interaction between users and clothing sales data. Finally, the clothing inventory is automatically replenished and orders are managed by real-time monitoring of the inventory level and sales speed of clothing products, thereby improving the security of clothing sales data in the intelligent analysis process.

[0019] The technical solution in the embodiment of the present application is to solve the problem of low security of clothing sales data in the above-mentioned intelligent analysis process. The overall idea is as follows: By detecting the clothing sales data of clothing products and competing products to obtain the imitation risk coefficient, and then obtaining the malicious return order threshold by analyzing the interaction between users and clothing sales data, the security of clothing sales data in the intelligent analysis process is improved.

[0020] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0021] like Figure 1As shown, it is a structural schematic diagram of the clothing sales data intelligent analysis system provided by the embodiment of the present application. The clothing sales data intelligent analysis system provided by the embodiment of the present application includes a clothing sales data processing module, an imitation analysis module, a return order analysis module and a real-time monitoring module; wherein the clothing sales data processing module is used to obtain clothing sales data by collecting clothing information of clothing products, the clothing information is used to record the information of clothing products, and the clothing sales data represents the data of clothing products after the numerical data corresponding to the clothing information is preprocessed; the imitation analysis module is used to detect the clothing sales data of clothing products and competing products to obtain the imitation risk coefficient, and the imitation risk coefficient is used to measure the degree of imitation risk of clothing products; the return order analysis module is used to obtain the malicious return order threshold by analyzing the interaction behavior between users and clothing sales data, and the malicious return order threshold is used to identify the malicious return order behavior of users during the sales process of clothing products; the real-time monitoring module is used to automatically replenish clothing inventory and manage orders by real-time monitoring the inventory level and sales speed of clothing products.

[0022] In this embodiment, when the user's order return behavior exceeds the set malicious order return threshold, the system will determine that the user has the possibility of malicious order return, and take corresponding measures to prevent and handle the situation. The system will automatically trigger the early warning mechanism to remind the merchant to pay attention to the user's order return behavior; at the same time, the merchant needs to further review the user's order return behavior that has been warned. If it is confirmed that the user has malicious order return behavior, the merchant can take corresponding measures to handle it, such as restricting the user's purchasing authority, requiring the user to bear the return shipping fee, etc.; finally, the merchant needs to regularly provide feedback and optimize the use of the malicious order return threshold. According to the actual results and user feedback, the merchant can adjust and optimize the threshold to improve its accuracy and effectiveness; in short, the malicious order return threshold is of great significance in the sales process of clothing products, which can help merchants more accurately identify malicious order return behavior and take corresponding measures to prevent and handle it, thereby improving transaction security.

[0023] Furthermore, the clothing sales data processing module includes a digitization unit, a data conversion unit, a visualization unit and a data update unit; the digitization unit is used to convert the collected clothing information into digital form to obtain clothing sales data; the data conversion unit is used to convert the acquired clothing sales data into a preset format and classify it; the visualization unit is used to visualize the clothing sales data and obtain a data analysis report based on it, and the data analysis report includes a counterfeit risk factor and a malicious return order threshold; the data update unit is used to regularly clean up expired clothing sales data and back up the clothing sales data at the same time.

[0024] In this embodiment, the data unit is the basis of the entire process. It converts the collected clothing information into binary code through an analog-to-digital converter, and further converts it into digital form to obtain clothing sales data, so that the originally complex and diverse information becomes quantifiable and analyzable, providing support for subsequent data analysis and decision-making; the data conversion unit converts the fields and values ​​in the data into specific fields and values ​​according to preset mapping rules, such as mapping the product number to the product name, or mapping the color code to the specific color name, and summarizes or classifies the data to achieve data normalization, so that data from different sources and in different formats can be uniformly processed and analyzed. Through data conversion, enterprises can understand the information behind the data more clearly, providing more support for subsequent data analysis and decision-making. Accurate and reliable data support is provided; the visualization display unit uses the visualization analysis platform to obtain the characteristics of clothing sales data, and designs a bar chart based on the characteristics of clothing sales data to visualize the clothing sales data, making the data analysis results more intuitive and easy to understand. Through the data analysis report, enterprises can have a more comprehensive understanding of their own operations and market conditions, and provide more powerful support for decision-making. The counterfeit risk coefficient and malicious return order threshold in the data analysis report represent the counterfeit risk of clothing products and the possibility of malicious return order behavior by users, respectively. Through these two indicators; the data update unit can avoid the interference of expired data on decision-making. At the same time, backup data can ensure the security of data and prevent data loss or damage; the accuracy of data processing is improved.

[0025] Furthermore, the specific method for obtaining the imitation risk coefficient is as follows: obtain the number of clothing products and corresponding competitive products, and number the clothing products and competitive products; obtain a content similarity evaluation value by analyzing the text similarity between the clothing products and competitive products, and the content similarity evaluation value is used to measure the text similarity between the clothing products and competitive products; extract the text keywords of the clothing products and competitive products and count the number of text keyword overlaps to obtain the keyword repetition rate, and the keyword repetition rate is used to measure the repetition degree of the keywords of the clothing products and competitive products; compare the similarity between the text grammatical structure and the syntactic structure of the clothing products and competitive products to obtain the structural similarity, Structural similarity is used to measure the similarity between grammatical structure and syntactic structure; the weight coefficient of the imitation data is obtained from the preset database, the imitation data includes the content similarity evaluation value, the keyword repetition rate and the structural similarity, the weight coefficient includes the first weight coefficient, the second weight coefficient and the third weight coefficient, the first weight coefficient is used to measure the influence of the content similarity evaluation value on the imitation risk coefficient, the second weight coefficient is used to measure the influence of the keyword repetition rate on the imitation risk coefficient, and the third weight coefficient is used to measure the influence of the structural similarity on the imitation risk coefficient; the imitation data is normalized, and the imitation risk coefficient is obtained in combination with the weight coefficient.

[0026] In this embodiment, the weight coefficient is obtained from a preset database, thereby improving the accuracy of the clothing product imitation risk analysis.

[0027] In a specific embodiment, the first weight coefficient obtains a fitting curve by fitting the relationship between product text similarity and imitation risk by using the content similarity evaluation value and the imitation risk coefficient, and obtains the first weight coefficient corresponding to the input content similarity evaluation value according to the fitting curve.

[0028] In a specific embodiment, the second weight coefficient obtains a fitting curve by fitting the relationship between the number of times of text keyword overlap and the imitation risk through the keyword repetition rate and the imitation risk coefficient, and obtains the second weight coefficient corresponding to the input keyword repetition rate according to the fitting curve; In a specific embodiment, the third weight coefficient obtains a fitting curve by fitting the relationship between text structure similarity and imitation risk through structural similarity and imitation risk coefficient, and obtains the third weight coefficient corresponding to the input structural similarity according to the fitting curve.

[0029] Furthermore, the imitation risk coefficient is specifically calculated using the following formula: ; In the formula, represents the imitation risk coefficient, Indicates the number of the clothing product. , represents the total number of clothing products, Indicates Content similarity evaluation value of a set of clothing products, Indicates The keyword repetition rate of a set of clothing products, Indicates The structural similarity of a set of clothing products, represents the first weight coefficient, represents the second weight coefficient, Represents the third weight coefficient.

[0030] In this embodiment, the content similarity evaluation value of clothing products is positively correlated with the imitation risk coefficient, the keyword repetition rate of clothing products is positively correlated with the imitation risk coefficient, and the structural similarity of clothing products is positively correlated with the imitation risk coefficient; a comprehensive assessment of the imitation risk of clothing products is achieved, effectively solving the problem of low security of clothing sales data.

[0031] Furthermore, a specific method for obtaining a content similarity evaluation value is as follows: constructing a clothing description text table based on consistent product words in the collected description texts of clothing products, the consistent product words indicating product information that has also appeared in the texts of clothing products and competing products, and the clothing description text table is used to count the consistent product words in the description texts of clothing products and competing products; counting the number of times consistent product words appear in clothing product texts based on the clothing description text table and sorting them to obtain a first sequence, the first sequence indicating the number of times consistent product words appear in clothing product texts; counting the number of times consistent product words appear in competing product texts based on the clothing description text table and sorting them to obtain a second sequence, and obtaining a content similarity evaluation value in combination with the first sequence, the second sequence indicating the number of times consistent product words appear in competing product texts.

[0032] In this embodiment, the content similarity evaluation value is specifically calculated using the following formula: ; In the formula, Indicates Content similarity evaluation value of a set of clothing products, Indicates the number of the clothing product. , represents the total number of clothing products, A number indicating the product term, , Indicates the total number of product words, The first sequence elements, The second sequence elements; for the convenience of analysis, define , is the first sequence coefficient, , is the second sequence coefficient, and is defined , is the total coefficient of the sequence, then the calculation formula for the content similarity evaluation value is , assuming is 1, and the statistical table of changes in the content similarity evaluation value is shown in Table 1: Table 1 Statistics of changes in content similarity evaluation values

[0033] Among them, the first sequence coefficient is negatively correlated with the content similarity evaluation value, the second sequence coefficient is negatively correlated with the content similarity evaluation value, and the total sequence coefficient is positively correlated with the content similarity evaluation value; a comprehensive evaluation of the content similarity evaluation value is achieved, effectively solving the problem of high risk of content imitation.

[0034] Furthermore, the specific steps for obtaining structural similarity are as follows: extract the frequency and distribution of nouns and adjectives in the text of clothing products and competing products to obtain part-of-speech features, and the part-of-speech features are used to measure the similarity of the frequency and distribution of nouns and adjectives in the text of clothing products and competing products; extract structural features by performing syntactic analysis on the text of clothing products and competing products, and syntactic analysis is used to reveal the relationship between the subject and modifiers in the text sentence, and the structural features are used to measure the syntactic similarity in the text of clothing products and competing products; use a graph matching algorithm to compare the structural features of the corresponding texts and combine the part-of-speech features to obtain structural similarity.

[0035] In this embodiment, a syntactic analyzer can be used to extract the frequency and distribution of nouns and adjectives in the text of clothing products and competing products to obtain part-of-speech features. The results of the syntactic analysis are usually presented in the form of a parse tree or a dependency graph. The syntactic analysis reveals the internal structure and relationships of the sentence, which helps to understand the deep meaning and logical structure of the text, thereby improving the security of the text structure.

[0036] Furthermore, a specific method for obtaining the malicious order return threshold is as follows: within a preset time period, the average number of order returns for each customer is counted to obtain the order return frequency, and the order return frequency is used to measure the abnormality of user order return; within a preset time period, the average number of orders for each customer is counted to obtain the order frequency, and the order frequency is used to measure the abnormality of user order placement; within a preset time period, the refund amount and the corresponding purchase amount of each user are counted, and the ratio of the total refund amount of the user to the total purchase amount is analyzed to obtain the refund ratio, and the refund ratio is used to quantify the possibility of users maliciously obtaining benefits through purchases and refunds; a correction factor for the refund ratio is obtained from a preset database, and the correction factor for the refund ratio is used to measure the degree of influence of the refund ratio on the malicious order return threshold; the refund ratio threshold is obtained and the malicious order return data is normalized to obtain the malicious order return threshold, and the malicious order return data includes the order frequency, the order return frequency and the refund ratio.

[0037] In this embodiment, a refund ratio threshold is obtained from a database; a correction factor for the refund ratio is obtained by fitting the refund ratio and the corresponding malicious order return threshold to obtain a fitting curve based on the relationship between the possibility of users maliciously obtaining profits through purchases and refunds during the sales process of clothing products and malicious order return behavior, and the correction factor for the refund ratio corresponding to the input refund ratio is obtained based on the fitting curve; thereby achieving a more accurate analysis of malicious refunds.

[0038] Furthermore, the calculation formula for the malicious return order threshold is as follows: ; In the formula, Indicates the malicious chargeback threshold for a preset time period. Indicates the frequency of return orders in the preset time period. Indicates the order frequency for a preset time period, Indicates the refund ratio, Indicates the refund ratio threshold. A correction factor indicating the refund ratio.

[0039] In this embodiment, for the convenience of analysis, we define , is the frequency coefficient of order cancellations, defined as , is the refund ratio coefficient, then the calculation formula for the malicious refund threshold is: , where the value range of the order cancellation frequency coefficient is , the refund ratio coefficient range is ,like Figure 2 As shown, there is a statistical chart of the changes in the malicious order return threshold provided in the example of this application. It can be seen from the figure that the order return frequency coefficient is positively correlated with the malicious order return threshold, and the refund ratio coefficient is positively correlated with the malicious order return threshold, which realizes a comprehensive evaluation of malicious order returns, correspondingly ensures the accuracy and reliability of clothing sales data, and can achieve more accurate analysis.

[0040] Further, such as Figure 3 As shown, it is a flowchart of real-time monitoring provided by an embodiment of the present application, and the specific steps of real-time monitoring are as follows: using a barcode recognition method to monitor and locate the barcode labels of clothing inventory in real time; obtaining the changing rules of inventory levels and sales speeds by real-time locating inventory levels and monitoring sales speeds; setting inventory warning lines according to the changing rules of inventory levels and sales speeds, and judging in real time whether the inventory level is lower than the inventory warning line, and automatically issuing an early warning notification if it is, otherwise continuing to sell, and the inventory warning line is used to prevent production and sales interruptions caused by insufficient inventory; obtaining replenishment suggestion data in combination with the statistical chart variation coefficient and sales speed, and the replenishment suggestion data is used to automatically update and replenish orders.

[0041] In this embodiment, historical sales data and inventory data are first analyzed to understand the fluctuation of inventory levels over time, and then the inventory level range is obtained in combination with sales demand and inventory cost; within this range, the corresponding minimum inventory warning line and maximum inventory warning line are obtained according to the minimum and maximum values ​​of the inventory level range. When the inventory level drops below the minimum inventory warning line, replenishment measures need to be taken immediately to avoid out-of-stock. At the same time, when the inventory level exceeds the maximum inventory warning line, it is necessary to adjust the procurement plan or take promotional measures to reduce the inventory level; according to the law of inventory level changes and the law of sales speed changes, the system can set an inventory warning line. When the inventory level is lower than the warning line, the system will automatically issue an early warning notification to remind managers to replenish in time, which can prevent production and sales interruptions caused by insufficient inventory and ensure the continuity and stability of the supply chain. The entire system realizes the automated and intelligent management of clothing inventory, reduces manual intervention and errors, improves management efficiency, and realizes the automated management of clothing product inventory.

[0042] The embodiment of the present application provides an intelligent analysis method for clothing sales data, which includes the following steps: obtaining clothing sales data by collecting clothing information of clothing products, the clothing information is used to record information of clothing products, and the clothing sales data represents data of clothing products after numerical data corresponding to the clothing information is preprocessed; detecting clothing sales data of clothing products and competing products to obtain imitation risk coefficients, and the imitation risk coefficients are used to measure the degree of imitation risk of clothing products; obtaining malicious order return thresholds by analyzing the interactive behavior of users and clothing sales data, and the malicious order return thresholds are used to identify malicious order return behaviors of users during the sales process of clothing products; and automatically replenishing clothing inventory and managing orders by real-time monitoring of clothing product inventory levels and sales speeds.

[0043] In this embodiment, obtaining the counterfeit risk coefficient can help enterprises better understand the risk level of their products being counterfeited, so as to take corresponding preventive measures to ensure that the interests of the enterprise are not harmed. Setting a malicious order return threshold can help enterprises better identify malicious order return behavior and take special measures for orders exceeding the threshold, such as refusing refunds, deducting credit points, etc., so as to maintain transaction order and consumer rights; at the same time, to improve the security of clothing sales data, it is also necessary to use encryption algorithms to encrypt sensitive data to ensure the security of data during transmission and storage, set strict access rights and identity authentication mechanisms to ensure that only authorized personnel can access and modify data, prevent internal personnel from abusing data or external personnel from illegally intruding, and achieve improved security of clothing sales data management.

[0044] To summarize, the embodiment of the present application obtains clothing sales data by collecting clothing information of clothing products, detects clothing sales data of clothing products and competing products to obtain imitation risk coefficients, and then obtains malicious order return thresholds by analyzing the interaction between users and clothing sales data. Finally, clothing inventory is automatically replenished and orders are managed by real-time monitoring of clothing product inventory levels and sales speeds, thereby improving the security of clothing sales data in the intelligent analysis process, and effectively solving the problem of low security of clothing sales data in the intelligent analysis process in the prior art.

[0045] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0046] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0047] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0048] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1The steps for the functions specified in one or more boxes.

[0049] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0050] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. The clothing sales data intelligent analysis system is characterized by: It includes clothing sales data processing module, imitation analysis module, return order analysis module and real-time monitoring module; Wherein, the clothing sales data processing module is used to obtain clothing sales data by collecting clothing information of clothing products; The imitation analysis module is used to detect clothing sales data of clothing products and competing products to obtain imitation risk coefficients, and the imitation risk coefficients are used to measure the degree of imitation risk of clothing products; The chargeback analysis module is used to obtain a malicious chargeback threshold by analyzing the interaction between the user and the clothing sales data, and the malicious chargeback threshold is used to identify the malicious chargeback behavior of the user during the clothing product sales process; The real-time monitoring module is used to automatically replenish clothing inventory and manage orders by monitoring clothing product inventory levels and sales speeds in real time.

2. The clothing sales data intelligent analysis system according to claim 1, characterized in that: The clothing sales data processing module includes a data unit, a data conversion unit, a visual display unit and a data update unit; The data conversion unit is used to convert the collected clothing information into digital form to obtain clothing sales data; The data conversion unit is used to convert the acquired clothing sales data into a preset format and classify it; The visualization display unit is used to visualize the clothing sales data and obtain a data analysis report based on it; The data updating unit is used to regularly clean up expired clothing sales data and back up clothing sales data.

3. The clothing sales data intelligent analysis system according to claim 1, characterized in that: The specific method for obtaining the imitation risk coefficient is as follows: Obtain the clothing quantity of the clothing product and the corresponding competing product, and number the clothing product and the competing product; Obtain content similarity evaluation values ​​by analyzing the text similarity between clothing products and competing products; Extract the text keywords of clothing products and competing products and count the number of times the text keywords overlap to obtain the keyword repetition rate; Compare the similarity between the grammatical structure and syntactic structure of the text of clothing products and competing products to obtain structural similarity; Obtaining weight coefficients of imitation data from a preset database, the imitation data including a content similarity evaluation value, a keyword repetition rate and a structural similarity, the weight coefficients including a first weight coefficient, a second weight coefficient and a third weight coefficient; The imitation data is normalized and combined with the weight coefficient to obtain the imitation risk coefficient.

4. The clothing sales data intelligent analysis system as claimed in claim 3, characterized in that: The imitation risk coefficient is specifically calculated using the following formula: ； In the formula, represents the imitation risk coefficient, Indicates the number of the clothing product. , represents the total number of clothing products, Indicates Content similarity evaluation value of a set of clothing products, Indicates The keyword repetition rate of a set of clothing products, Indicates The structural similarity of a set of clothing products, represents the first weight coefficient, represents the second weight coefficient, Represents the third weight coefficient.

5. The clothing sales data intelligent analysis system as claimed in claim 3, characterized in that: The specific method for obtaining the content similarity evaluation value is as follows: Constructing a clothing description text table according to consistent product words in the collected description texts of clothing products; According to the clothing description text table, the number of times the consistent product words appear in the clothing product text is counted and sorted to obtain the first sequence; According to the clothing description text table, the number of times the consistent product words appear in the competing product texts is counted and sorted to obtain a second sequence, and combined with the first sequence to obtain a content similarity evaluation value.

6. The clothing sales data intelligent analysis system as claimed in claim 3, characterized in that: The specific steps for obtaining the structural similarity are as follows: Extract the frequency and distribution of nouns and adjectives in the text of clothing products and competing products to obtain part-of-speech features; Structural features are extracted by syntactic analysis of the texts of clothing products and competing products; A graph matching algorithm is used to compare the structural features of the corresponding texts and combine the part-of-speech features to obtain structural similarity.

7. The clothing sales data intelligent analysis system according to claim 1, characterized in that: The specific method for obtaining the malicious chargeback threshold is as follows: Count the average number of return orders and the average number of orders for each customer within a preset time period to obtain the corresponding return order frequency and order frequency; Count the refund amount and corresponding purchase amount of each user within a preset time period, and analyze the ratio of the total refund amount to the total purchase amount of the user to obtain the refund ratio; Obtain the correction factor of the refund ratio from the preset database; A refund ratio threshold is obtained and malicious return order data is normalized to obtain a malicious return order threshold, wherein the malicious return order data includes order frequency, return order frequency and refund ratio.

8. The clothing sales data intelligent analysis system as claimed in claim 7, characterized in that: The calculation formula of the malicious chargeback threshold is as follows: ； In the formula, Indicates the malicious chargeback threshold for a preset time period. Indicates the frequency of return orders in the preset time period. Indicates the order frequency for a preset time period, Indicates the refund ratio. Indicates the refund ratio threshold. A correction factor indicating the refund ratio.

9. The clothing sales data intelligent analysis system according to claim 1, characterized in that: The specific steps of the real-time monitoring are as follows: Use barcode recognition methods to monitor and locate barcode tags of clothing inventory in real time; By locating inventory levels in real time and monitoring sales speed, we can obtain the changing patterns of inventory levels and sales speed; Set inventory warning lines based on the changing patterns of inventory levels and sales speeds, and determine in real time whether the inventory level is below the inventory warning line. If so, an early warning notification will be automatically issued; otherwise, sales will continue; Combine the coefficient of variation and sales velocity in the statistical chart to obtain replenishment recommendation data.

10. An intelligent analysis method for clothing sales data, characterized in that: The following steps are involved: Obtain clothing sales data by collecting clothing information of clothing products; Detecting clothing sales data of clothing products and competing products to obtain a counterfeit risk coefficient, wherein the counterfeit risk coefficient is used to measure the counterfeit risk level of the clothing products; Obtaining a malicious order return threshold by analyzing the interaction between users and clothing sales data, wherein the malicious order return threshold is used to identify malicious order return behaviors of users during clothing product sales; Automatically replenish apparel inventory and manage orders by monitoring apparel product stock levels and sales velocity in real time.

Images