Commodity information management method and system based on big data analysis

Abstract

The invention belongs to the technical field of e-commerce, and provides a commodity information management method and system based on big data analysis, and the method comprises the steps: collecting the multi-dimensional data of commodities on all e-commerce platforms and social media in real time through an automatic crawler program; the collected data are preprocessed; performing deep analysis on the preprocessed data by applying a machine learning algorithm, constructing a commodity feature model, identifying commodity relevance, and predicting a market trend; on the basis of an analysis result, fine classification is automatically carried out on the commodities; through the automatic data acquisition and processing flow, the information processing speed is remarkably increased, human errors are reduced, the data quality is improved, meanwhile, the commodity features and the market trend are accurately predicted by using the deep learning technology and the machine learning algorithm, a scientific basis is provided for operation decision making, and by dynamically adjusting the inventory level, the product quality is improved. The inventory overstock risk is reduced, the overall efficiency of logistics and supply chains is improved, and the operation cost is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of e-commerce, and specifically relates to a commodity information management method and system based on big data analysis. Background Art

[0002] With the rapid development of the e-commerce industry, the amount of commodity information has increased dramatically. The traditional manual management model faces huge challenges in processing massive data, achieving precision marketing and efficient inventory management. Traditional commodity information management systems often rely on manual input and simple data analysis, which makes it difficult to adapt to rapidly changing market demands and consumer preferences.

[0003] To this end, those skilled in the art have proposed a commodity information management method and system based on big data analysis to solve the problems raised by the background technology. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a commodity information management method and system based on big data analysis, which realizes intelligent classification, accurate recommendation, inventory optimization and market trend prediction of commodity information, so as to improve the operational efficiency and customer satisfaction of e-commerce enterprises, and solve the huge challenges faced by the manual management mode in the existing technology in processing massive data, realizing accurate marketing and efficient inventory management. Traditional commodity information management systems often rely on manual input and simple data analysis, and are difficult to adapt to rapidly changing market demands and consumer preferences.

[0005] A commodity information management method based on big data analysis includes the following steps:

[0006] S1. Through automated crawler programs, real-time collection of multi-dimensional data of products on major e-commerce platforms and social media, including prices, sales, user reviews and pictures;

[0007] S2. Preprocess the collected data, including cleaning, deduplication, and standardization, and use natural language processing technology to analyze user comments and extract sentiment tendencies and keywords;

[0008] S3. Apply machine learning algorithms to conduct in-depth analysis of preprocessed data, build commodity feature models, identify commodity relevance, and predict market trends;

[0009] S4. Based on the analysis results, the products are automatically classified and combined with user behavior data to provide personalized product recommendations for different user groups;

[0010] S5. Analyze sales data and inventory status, use forecasting models to dynamically adjust inventory levels, and optimize logistics and supply chain management;

[0011] S6. Provide an intuitive and easy-to-use background management system to generate visual data analysis reports and support quick operational decision-making.

[0012] Preferably, the step S1 is specifically described as follows:

[0013] S1.1. Determine the collection target:

[0014] Identify the data sources that need to be collected, such as which e-commerce platforms (such as Taobao, JD.com, etc.) and social media (such as Weibo, Xiaohongshu, etc.);

[0015] Determine the type of data that needs to be collected, such as product prices, sales, user reviews, pictures, etc.;

[0016] S1.2. Design crawler architecture:

[0017] Design the overall architecture of the crawler, including the number of crawlers, deployment method (single machine or multi-machine distributed), data storage method, etc.;

[0018] Determine the crawler technology stack to be used, such as Python's Scrapy, Beautiful Soup, and other tools;

[0019] S1.3. Write a crawler program:

[0020] Write code to implement network requests and simulate user browsing behavior to obtain web page content;

[0021] Use regular expressions, XPath, CSS selectors and other methods to extract target data from web pages;

[0022] Write exception handling mechanisms to ensure that the crawler can run stably when encountering network problems, anti-crawler mechanisms, etc.

[0023] S1.4, set up scheduled tasks;

[0024] Configure the crawler to run as a scheduled task to ensure that the latest data can be collected in real time or regularly;

[0025] You can use Linux's Cron job or Python's APScheduler and other libraries to implement the timing function.

[0026] Preferably, in step S2, the data preprocessing includes:

[0027] 1) Data cleaning:

[0028] Check whether there are missing values ​​in the data set, fill them according to business logic (such as using mean, median, mode, or inferring from other fields) or delete them directly;

[0029] Outlier processing: Identify and process outliers, such as data with prices that are clearly outside the normal range, and choose to delete or adjust them based on business logic;

[0030] Format unification: Ensure that the data format is unified, such as date, time, number, etc.;

[0031] 2) Deduplication:

[0032] Identify and delete duplicate data records by comparing key fields (such as product ID, SKU, etc.);

[0033] You can use the database deduplication function or implement it programmatically;

[0034] 3) Standardization: Standardize data so that data from different sources and in different formats can be processed and analyzed uniformly, including unified units (such as converting prices from US dollars to RMB) and unified classification standards (such as mapping product categories on different platforms to a unified classification system).

[0035] Preferably, the step S3 is specifically described as follows:

[0036] S3.1. Ensure that the pre-processed data is ready, including the cleaned, deduplicated, and standardized product information data, and divide the data into training sets, validation sets, and test sets for model training and evaluation;

[0037] S3.2. Select features useful for building product feature models from the preprocessed data, such as price, sales volume, user evaluation sentiment score, keyword frequency, and then transform the selected features, such as encoding categorical variables, normalizing or standardizing numerical variables, etc., to improve model performance;

[0038] S3.3. Select appropriate machine learning algorithms, such as decision trees, random forests, gradient boosted tree (GBDT), and neural networks, based on specific tasks (such as classification, regression, clustering, etc.) and data characteristics. Use the training set data to train the selected machine learning algorithm, adjust the model parameters to optimize model performance, and then use the validation set data to validate the trained model and evaluate the generalization ability of the model.

[0039] S3.4. Based on the trained machine learning model, a product feature model is constructed. The model can capture the key features of the product, such as price sensitivity, user preference, market acceptance, etc. Through further analysis of the product feature model, the market performance and user feedback of the product can be deeply understood. The feature similarity or association rule mining algorithm (such as April algorithm, FP-Growth algorithm, etc.) in the product feature model is used to identify the correlation between different products. Correlation analysis helps to discover potential product combination or matching sales opportunities and improve sales conversion rate.

[0040] S3.5. Based on historical sales data, user behavior data, and market change factors (such as holidays, promotional activities, policy changes, etc.), use forecasting models such as time series analysis, regression analysis, or deep learning to predict market trends. Market trend forecasting can help companies adjust inventory, formulate marketing strategies, and plan production plans in advance.

[0041] Preferably, the step S4 is specifically described as follows:

[0042] 1) Refined classification of goods:

[0043] Analyze product characteristics: Use the product characteristic model built by the big data analysis module to conduct in-depth analysis of the multi-dimensional data of the product and extract the key characteristics of the product;

[0044] Cluster analysis: Use clustering algorithms (such as K-means, hierarchical clustering, etc.) to group products so that products with similar characteristics are classified into one category, thus achieving refined classification of products;

[0045] Classification adjustment and optimization: According to market changes and user feedback, we continuously adjust and optimize commodity classification rules to ensure the accuracy and timeliness of classification;

[0046] 2) User behavior data analysis:

[0047] Collect user behavior data: Collect user behavior data through the interaction data between users and e-commerce platforms (such as browsing history, purchase history, search keywords, etc.);

[0048] User portrait construction: Use user behavior data to build user portraits, including user interest preferences, purchasing power, consumption habits and other characteristics;

[0049] User group segmentation: Based on the characteristics of user portraits, clustering or classification algorithms are used to divide users into different groups for personalized recommendations;

[0050] 3) Personalized product recommendations:

[0051] Matching algorithm: Based on product classification and user group division, collaborative filtering, content-based recommendation or hybrid recommendation algorithm are used to recommend suitable products to users;

[0052] Real-time update: As user behavior continues to change and product information is updated, the recommendation algorithm is adjusted in real time to ensure the accuracy and timeliness of recommendations;

[0053] Recommendation result display: Display recommendation results to users in an intuitive and easy-to-use way, such as recommendation lists, product detail page recommendations, etc., to improve user experience and sales conversion rate;

[0054] 4) Feedback and optimization:

[0055] User feedback collection: Collect user feedback on recommendation results through user clicks, purchases, comments, etc.

[0056] Recommendation effect evaluation: Use evaluation indicators (such as click-through rate, conversion rate, user satisfaction, etc.) to evaluate the recommendation effect;

[0057] Optimize recommendation strategies: Based on evaluation results and user feedback, continuously optimize recommendation algorithms and strategies to improve recommendation effects and user satisfaction.

[0058] Preferably, the step S5 is specifically described as follows:

[0059] S5.1. Sales data and inventory status analysis:

[0060] Data collection: Regularly collect sales data (such as sales volume, sales revenue, sales speed, etc.) and inventory status data (such as inventory volume, inventory turnover rate, out-of-stock rate, etc.) of goods;

[0061] Data analysis: Use statistical analysis methods to process the collected data to identify key information such as sales trends, seasonal changes, hot and slow-selling products;

[0062] S5.2. Prediction model construction and application:

[0063] Model selection: Select appropriate forecasting models based on product characteristics and business needs, such as time series analysis, regression analysis, machine learning models, etc.

[0064] Model training: Use historical sales data and inventory status data to train the forecasting model to ensure that the model can accurately predict future sales trends and inventory requirements;

[0065] S5.3. Dynamic adjustment of inventory levels

[0066] Inventory strategy formulation: formulate inventory adjustment strategies based on forecast results and inventory targets (such as safety stock, inventory turnover rate, etc.);

[0067] Inventory adjustment execution: Dynamically adjust inventory levels according to inventory adjustment strategies, including replenishment, promotion, transfer and other measures;

[0068] Monitoring and adjustment: Monitor inventory status and sales in real time, and flexibly adjust inventory adjustment strategies based on actual conditions:

[0069] S5.4. Logistics and supply chain management optimization:

[0070] Supply chain collaboration: Establish close collaborative relationships with suppliers, logistics service providers and other partners to ensure efficient operation of the supply chain;

[0071] Logistics optimization: Optimize logistics routes, distribution plans and warehouse layouts based on inventory adjustments and sales forecast results to reduce logistics costs and improve distribution efficiency;

[0072] Risk management: Identify and evaluate potential risks in the supply chain, formulate countermeasures, and ensure the stability and reliability of the supply chain;

[0073] S5.5. Continuous optimization and improvement:

[0074] Effect evaluation: Regularly evaluate the effects of inventory management and supply chain optimization, including indicators such as inventory cost, out-of-stock rate, and customer satisfaction;

[0075] Feedback loop: Based on evaluation results and user feedback, we continuously optimize forecasting models, inventory adjustment strategies, and supply chain management processes to form a closed loop of continuous improvement.

[0076] Preferably, the step S6 is specifically described as follows:

[0077] 1)Backend management system design:

[0078] User interface design: Design an intuitive and concise user interface to ensure that operators can quickly get started and use the system efficiently;

[0079] Functional module division: According to business needs, the backend management system is divided into different functional modules, such as data management, analysis reports, system configuration, etc.;

[0080] Permission management: Set up a reasonable permission management mechanism to ensure that users at different levels can only access and operate data and functions within their permission scope;

[0081] 2) Data visualization design:

[0082] Data selection: Select key indicators and data to be displayed from the data warehouse based on operational needs;

[0083] Chart design: Use data visualization tools (such as Tableau, Power BI, etc.) to design intuitive and easy-to-understand charts, such as line charts, bar charts, pie charts, scatter charts, etc.;

[0084] Interaction design: Design user interaction functions such as filtering, sorting, zooming, etc. so that users can freely explore and mine data;

[0085] 3) Data analysis report generation:

[0086] Report template design: Design data analysis report templates based on business needs, including report title, summary, data charts, analysis conclusions, etc.;

[0087] Data filling and verification: fill the pre-processed and analyzed data into the report template and perform data verification to ensure the accuracy and completeness of the data;

[0088] Report export and sharing: Provides report export function, supports exporting reports to PDF, Excel and other formats, convenient for users to share and archive;

[0089] 4) Operational decision support:

[0090] Data analysis: Through the generated data analysis reports, in-depth analysis of business data can be carried out to discover business rules and problems;

[0091] Decision-making suggestions: Based on data analysis results, provide operations personnel with targeted decision-making suggestions, such as inventory management strategies, product promotion plans, etc.;

[0092] Decision execution and monitoring: Support operations personnel to perform relevant operations according to decision suggestions and monitor business data changes in real time to evaluate and adjust the decision effects;

[0093] 5) System maintenance and upgrade:

[0094] Performance monitoring: Regularly monitor the performance indicators of the backend management system, such as response time, number of concurrent users, etc., to ensure stable operation of the system;

[0095] Data update: Regularly update the data and models in the background management system to maintain the timeliness and accuracy of the data;

[0096] System upgrade: Based on user feedback and technological development, the backend management system is continuously optimized and upgraded to improve the system's ease of use and functionality.

[0097] A commodity information management system based on big data analysis, comprising:

[0098] Data collection module, used to automatically capture product information;

[0099] Data preprocessing module, used for data cleaning, deduplication, standardization and user comment analysis;

[0100] Big data analysis module, used to deeply analyze data, build models and predict trends;

[0101] Intelligent classification and recommendation system module, used for product classification and personalized recommendations;

[0102] Inventory and supply chain optimization module, used for inventory management and supply chain optimization;

[0103] User interface module, providing background management and data analysis reports.

[0104] 9. A processor configured to execute the above-mentioned commodity information management method based on big data analysis.

[0105] 10. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the above-mentioned commodity information management method based on big data analysis.

[0106] Compared with the prior art, the present invention has the following beneficial effects:

[0107] 1. The present invention significantly speeds up information processing, reduces human errors and improves data quality by automating the data collection and processing process.

[0108] 2. The present invention uses deep learning technology and machine learning algorithms to accurately predict product characteristics and market trends, providing a scientific basis for operational decisions.

[0109] 3. The present invention reduces the risk of inventory backlogs by dynamically adjusting inventory levels, improves the overall efficiency of logistics and supply chains, and reduces operating costs.

[0110] 4. The personalized recommendation strategy of the present invention provides customized product recommendations based on user behavior, improves user satisfaction and sales conversion rate, and enhances the market competitiveness of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0111] Figure 1 This is a flow chart of the commodity information management method based on big data analysis of the present invention;

[0112] Figure 2 This is a framework diagram of the commodity information management system based on big data analysis of the present invention. DETAILED DESCRIPTION

[0113] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0114] like Figure 1 As shown:

[0115] Embodiment: The present invention provides a commodity information management method based on big data analysis, comprising the following steps:

[0116] S1. Collect multi-dimensional data of products on major e-commerce platforms and social media in real time through automated crawler programs. The specific description is as follows:

[0117] S1.1. First, clarify the type of data that needs to be collected, such as product prices, sales, user reviews, and pictures;

[0118] S1.2, then design the overall architecture of the crawler, including the number of crawlers, deployment method, and data storage method;

[0119] S1.3. Write code to implement network requests, simulate user browsing behavior to obtain web page content, and use regular expressions, XPath, CSS selectors and other methods to extract target data from web pages;

[0120] S1.4. Configure the crawler to run as a scheduled task to ensure that the latest data can be collected in real time or regularly;

[0121] S2. Preprocessing the collected data; the data preprocessing includes:

[0122] Data cleaning: Check if there are missing values ​​in the data set, fill them in or delete them directly according to business logic; identify and handle outliers, such as data with prices that are obviously beyond the normal range, which can be deleted or adjusted according to business logic; format unification: ensure that the format of data is unified, such as date, time, number, etc.;

[0123] Deduplication: Identify and delete duplicate data records by comparing key fields; this can be achieved using the database deduplication function or programming;

[0124] Standardization: Standardize data so that data from different sources and in different formats can be processed and analyzed uniformly;

[0125] S3. Apply machine learning algorithms to conduct in-depth analysis on preprocessed data, build commodity feature models, identify commodity relevance, and predict market trends. The specific description is as follows:

[0126] S3.1. Ensure that the pre-processed data is ready, including the cleaned, deduplicated, and standardized product information data, and divide the data into training sets, validation sets, and test sets for model training and evaluation;

[0127] S3.2, selecting features useful for building a product feature model from the preprocessed data, and then transforming the selected features;

[0128] S3.3. Select an appropriate machine learning algorithm based on the specific task and data characteristics, use the training set data to train the selected machine learning algorithm, adjust the model parameters to optimize the model performance, and then use the validation set data to validate the trained model;

[0129] S3.4. Build a product feature model based on the trained machine learning model, and use the feature similarity or association rule mining algorithm in the product feature model to identify the association between different products;

[0130] S3.5. Predict market trends based on historical sales data, user behavior data and market change factors;

[0131] S4. Based on the analysis results, the products are automatically classified in fine detail, and personalized product recommendations are provided for different user groups in combination with user behavior data. The specific description is as follows:

[0132] S4.1. Use the commodity feature model built by the big data analysis module to conduct in-depth analysis of the multi-dimensional data of commodities, extract the key features of commodities, and then use clustering algorithms to group commodities so that commodities with similar features are classified into one category, thus achieving refined classification of commodities;

[0133] S4.2. Collect user behavior data through the interaction data between users and e-commerce platforms, and use user behavior data to build user portraits, including user interest preferences, purchasing power, consumption habits and other characteristics;

[0134] S4.3. Based on product classification and user group segmentation, collaborative filtering, content-based recommendation or hybrid recommendation algorithms are used to recommend appropriate products to users;

[0135] S4.4. Collect user feedback on recommendation results through user clicks, purchases, comments, etc.

[0136] S5. Analyze sales data and inventory status, use forecasting models to dynamically adjust inventory levels, and optimize logistics and supply chain management; the specific description is as follows:

[0137] S5.1. Use statistical analysis methods to process the collected data and identify key information on sales trends, seasonal changes, and hot and slow-selling products;

[0138] S5.2. Use historical sales data and inventory status data to train the forecasting model to ensure that the model can accurately predict future sales trends and inventory requirements;

[0139] S5.3. Formulate inventory adjustment strategies based on forecast results and inventory targets, and dynamically adjust inventory levels based on the inventory adjustment strategies;

[0140] S5.4. Optimize logistics routes, distribution plans and warehouse layout based on inventory adjustments and sales forecast results;

[0141] S5.5. Based on the evaluation results and user feedback, continuously optimize the forecasting model, inventory adjustment strategy and supply chain management process to form a closed loop of continuous improvement;

[0142] S6. Provides an intuitive and easy-to-use backend management system to generate visual data analysis reports and support quick operational decision-making. The specific description is as follows:

[0143] S6.1. Divide the backend management system into different functional modules according to business needs;

[0144] S6.2. Select key indicators and data to be displayed from the data warehouse according to operational needs;

[0145] S6.3. Design a template for data analysis report based on business needs, then fill the pre-processed and analyzed data into the report template and perform data verification;

[0146] S6.4. Provide targeted decision-making suggestions to operators based on data analysis results;

[0147] S6.5. Regularly monitor the performance indicators of the background management system.

[0148] From the above, it can be seen that this method significantly speeds up information processing, reduces human errors, and improves data quality by automating data collection and processing procedures. At the same time, by utilizing deep learning technology and machine learning algorithms, it accurately predicts product characteristics and market trends, providing a scientific basis for operational decisions. By dynamically adjusting inventory levels, it reduces the risk of inventory backlogs, improves the overall efficiency of logistics and supply chains, and reduces operating costs. The personalized recommendation strategy of the present invention provides customized product recommendations based on user behavior, improves user satisfaction and sales conversion rate, and enhances the company's market competitiveness.

[0149] The performance indicators of the existing method (i.e., the traditional commodity information management method based on big data analysis to process commodity information) and the commodity information management method based on big data analysis provided in this embodiment are compared, and the comparison results are shown in Table 1;

[0150] Table 1- Performance index comparison results

[0151]

[0152]

[0153] It can be seen from Table 1 that compared with the existing methods, the commodity information management technology based on big data analysis provided by this embodiment has significant advantages over current conventional technologies in many aspects, especially in data processing speed, forecasting ability, inventory management and personalized recommendations, etc., which can better meet the operational needs of e-commerce companies and improve overall operational efficiency and user experience.

[0154] like Figure 2 As shown:

[0155] A commodity information management system based on big data analysis, comprising:

[0156] Data collection module, used to automatically capture product information;

[0157] Data preprocessing module, used for data cleaning, deduplication, standardization and user comment analysis;

[0158] Big data analysis module, used to deeply analyze data, build models and predict trends;

[0159] Intelligent classification and recommendation system module, used for product classification and personalized recommendations;

[0160] Inventory and supply chain optimization module, used for inventory management and supply chain optimization;

[0161] User interface module, providing background management and data analysis reports.

[0162] In summary, the present invention provides e-commerce companies with an efficient and intelligent product information management solution by integrating big data analysis and artificial intelligence technology, effectively solves the pain points in the traditional management model, and promotes the further development of the e-commerce industry.

[0163] The embodiment of the present application provides an electronic device, which is applicable to the above-mentioned commodity information management method based on big data analysis, including:

[0164] Memory, used to protect computer programs and data;

[0165] Processor, used to run system programs.

[0166] The embodiment of the present application provides a computer storage medium, which is applicable to the above-mentioned commodity information management method based on big data analysis, and performs hierarchical confidentiality management on the above-mentioned system and data in accordance with confidentiality management requirements.

[0167] Those skilled in the art will appreciate that the embodiments of the present application may be provided as a system or a computer program product. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt 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.) that contain computer-usable program code.

[0168] The embodiments of the present invention are provided for the purpose of illustration and description. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations of the present invention. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A commodity information management method based on big data analysis, characterized in that: The following steps are involved: S1. Collect multi-dimensional data of products on major e-commerce platforms and social media in real time through automated crawler programs; S2, preprocessing the collected data; S3. Apply machine learning algorithms to conduct in-depth analysis of preprocessed data, build commodity feature models, identify commodity relevance, and predict market trends; S4. Based on the analysis results, the products are automatically classified and combined with user behavior data to provide personalized product recommendations for different user groups; S5. Analyze sales data and inventory status, use forecasting models to dynamically adjust inventory levels, and optimize logistics and supply chain management; S6. Provide an intuitive and easy-to-use background management system to generate visual data analysis reports and support quick operational decision-making.

2. A commodity information management method based on big data analysis as claimed in claim 1, characterized in that: The specific description of step S1 is as follows: S1.

1. First, clarify the type of data that needs to be collected, such as product prices, sales, user reviews, and pictures; S1.2, then design the overall architecture of the crawler, including the number of crawlers, deployment method, and data storage method; S1.

3. Write code to implement network requests, simulate user browsing behavior to obtain web page content, and use regular expressions, XPath, CSS selectors and other methods to extract target data from web pages; S1.

4. Configure the crawler to run as a scheduled task to ensure that the latest data can be collected in real time or regularly.

3. A commodity information management method based on big data analysis as claimed in claim 1, characterized in that: In step S2, the data preprocessing includes: Data cleaning: Check if there are missing values ​​in the data set, fill them in or delete them directly according to business logic; identify and handle outliers, such as data with prices that are obviously beyond the normal range, which can be deleted or adjusted according to business logic; format unification: ensure that the format of data is unified, such as date, time, number, etc.; Deduplication: Identify and delete duplicate data records by comparing key fields; this can be achieved using the database deduplication function or programming; Standardization: Standardize data so that data from different sources and in different formats can be processed and analyzed uniformly.

4. A commodity information management method and system based on big data analysis as claimed in claim 1, characterized in that: The specific description of step S3 is as follows: S3.

1. Ensure that the pre-processed data is ready, including the cleaned, deduplicated, and standardized product information data, and divide the data into training sets, validation sets, and test sets for model training and evaluation; S3.2, selecting features useful for building a product feature model from the preprocessed data, and then transforming the selected features; S3.

3. Select an appropriate machine learning algorithm based on the specific task and data characteristics, use the training set data to train the selected machine learning algorithm, adjust the model parameters to optimize the model performance, and then use the validation set data to validate the trained model; S3.

4. Build a product feature model based on the trained machine learning model, and use the feature similarity or association rule mining algorithm in the product feature model to identify the association between different products; S3.

5. Predict market trends based on historical sales data, user behavior data and market change factors.

5. The commodity information management method based on big data analysis as claimed in claim 1, characterized in that: The specific description of step S4 is as follows: S4.

1. Use the commodity feature model built by the big data analysis module to conduct in-depth analysis of the multi-dimensional data of commodities, extract the key features of commodities, and then use clustering algorithms to group commodities so that commodities with similar features are classified into one category, thus achieving refined classification of commodities; S4.

2. Collect user behavior data through the interaction data between users and e-commerce platforms, and use user behavior data to build user portraits, including user interest preferences, purchasing power, consumption habits and other characteristics; S4.

3. Based on product classification and user group segmentation, collaborative filtering, content-based recommendation or hybrid recommendation algorithms are used to recommend appropriate products to users; S4.

4. Collect user feedback on recommendation results through user clicks, purchases, comments, etc.

6. A commodity information management method based on big data analysis as claimed in claim 1, characterized in that: The step S5 is specifically described as follows: S5.

1. Use statistical analysis methods to process the collected data and identify key information on sales trends, seasonal changes, and hot and slow-selling products; S5.

2. Use historical sales data and inventory status data to train the forecasting model to ensure that the model can accurately predict future sales trends and inventory requirements; S5.

3. Formulate inventory adjustment strategies based on forecast results and inventory targets, and dynamically adjust inventory levels based on the inventory adjustment strategies; S5.

4. Optimize logistics routes, distribution plans and warehouse layout based on inventory adjustments and sales forecast results; S5.

5. Based on the evaluation results and user feedback, continuously optimize the forecasting model, inventory adjustment strategy and supply chain management process to form a closed loop of continuous improvement.

7. A commodity information management method based on big data analysis as claimed in claim 1, characterized in that: The step S6 is specifically described as follows: S6.

1. Divide the backend management system into different functional modules according to business needs; S6.

2. Select key indicators and data to be displayed from the data warehouse according to operational needs; S6.

3. Design a template for data analysis report based on business needs, then fill the pre-processed and analyzed data into the report template and perform data verification; S6.

4. Provide targeted decision-making suggestions to operators based on data analysis results; S6.

5. Regularly monitor the performance indicators of the background management system.

8. A commodity information management system based on big data analysis as described in any one of claims 1 to 7, characterized in that: include: Data collection module, used to automatically capture product information; Data preprocessing module, used for data cleaning, deduplication, standardization and user comment analysis; Big data analysis module, used to deeply analyze data, build models and predict trends; Intelligent classification and recommendation system module, used for product classification and personalized recommendations; Inventory and supply chain optimization module, used for inventory management and supply chain optimization; User interface module, providing background management and data analysis reports.

9. A processor, characterized in that: The method is configured to execute the commodity information management method based on big data analysis according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the commodity information management method based on big data analysis described in any one of claims 1 to 7 is implemented.

Images