A real-time monitoring method and system for sales data based on data analysis
By analyzing and forecasting sales data in real time, identifying key factors, establishing sales forecasting models, and dynamically adjusting strategies, the problem of existing technologies being unable to capture market trends and consumer preferences in a timely manner has been solved. This has enabled more accurate business decisions and inventory management, and improved customer satisfaction and market competitiveness.
Patent Information
- Application Number
- CN202411176214.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-26
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-08-26
AI Technical Summary
Existing real-time sales data monitoring methods cannot capture changes in consumer behavior and market trends in a timely manner, making it difficult for companies to make accurate business decisions, manage inventory effectively, increase the risk of inventory backlog and obsolescence, and make it impossible to adjust product features or services in a timely manner to improve customer satisfaction and loyalty.
By acquiring sales data related to electronic products, preprocessing it, and then using factor analysis algorithms to identify key factors, a sales forecasting model is established. Sales trends are evaluated in real time, and trend analysis algorithms are used to identify sales change patterns for early warning. The parameter configuration of the sales forecasting model is optimized, and sales strategies are dynamically adjusted to cope with market changes.
It enables businesses to quickly identify market trends and consumer preferences, make more accurate business decisions, optimize resource allocation, reduce inventory backlog, improve customer satisfaction and loyalty, and enhance market adaptability and operational efficiency.
Smart Images

Figure CN118674493B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data analysis, and in particular to a method and system for real-time monitoring of sales data based on data analysis. Background Art
[0002] Electronic products are products that use electricity as the basis of their work, mainly including: watches, smart phones, telephones, televisions, video players (VCD, SVCD, DVD), video recorders, camcorders, radios, tape recorders, combination speakers, laser players (CD), computers, game consoles, mobile communication products, etc. With the development of the times, people's demand for electronic products is increasing, and the frequency of electronic product updates is getting faster and faster. In the field of electronic products, rapid technological innovation and the diversity of consumer demand require companies to be able to respond to market changes in real time. Traditional sales management systems face several core challenges: the inability to capture consumer behavior in a timely manner, rapid changes in market trends, and short product iteration cycles. These challenges make it difficult for companies to efficiently manage sales processes and inventory, which in turn affects profitability and market response speed.
[0003] The existing real-time monitoring methods of sales data are not convenient for real-time analysis of sales data. Enterprises cannot quickly identify market trends and consumer preferences, and thus cannot make more accurate business decisions. It is also not convenient for enterprises to more accurately predict future sales trends and effectively manage inventory, which increases the risk of inventory backlog and obsolescence, and cannot ensure that inventory is synchronized with market demand. At the same time, it is not convenient to analyze sales trends, and thus it is not convenient to better understand customer needs and to timely adjust product features or services to improve customer satisfaction and loyalty.
[0004] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention
[0005] In view of the deficiencies in the prior art, the present invention proposes a real-time monitoring method and system for sales data based on data analysis, which solves the problems raised in the above background technology that the existing system is not convenient for enterprises to quickly identify market trends and consumer preferences through real-time analysis of sales data, and thus cannot make more accurate business decisions; it is not convenient for enterprises to more accurately predict future sales trends, cannot effectively manage inventory, increases the risk of inventory backlog and obsolescence, and cannot ensure that inventory is synchronized with market demand; at the same time, it is not convenient to analyze sales trends, and thus it is not convenient to better understand customer needs and it is not convenient to timely adjust product features or services to improve customer satisfaction and loyalty.
[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions:
[0007] According to one aspect of the present invention, a method for real-time monitoring of sales data based on data analysis is provided, and the method for real-time monitoring of sales data based on data analysis comprises the following steps:
[0008] S1. Acquire sales data related to electronic products, and pre-process the sales data to obtain sales volume characteristic data;
[0009] S2. Analyze the obtained sales volume characteristic data using a factor analysis algorithm to obtain key factors affecting the sales volume of electronic products;
[0010] S3. Establish a sales forecasting model based on key factors, and use the sales forecasting model to evaluate the sales trend of electronic products in real time;
[0011] S4, analyzing the obtained electronic product sales trend using a trend analysis algorithm, identifying a sales change pattern of the electronic products, and issuing an early warning reminder based on the identified sales change pattern of the electronic products;
[0012] Among them, based on key factors, establishing a sales forecasting model and evaluating the sales trend of electronic products in real time through the sales forecasting model include the following steps:
[0013] S31. Convert the obtained key factors into risk assessment cycle;
[0014] S32, replacing the risk assessment cycle into the cycle index adjustment algorithm, dividing the obtained key factors into cycle indicators and trend indicators, and constructing and training a sales forecasting model through the cycle indicators and trend indicators to predict future sales trends;
[0015] S33. Using error indicators as evaluation criteria for the accuracy of sales forecast models;
[0016] S34. Based on the trend indicators, use the forecast model optimization algorithm to adjust and optimize the parameters of the sales forecast model and find the best parameter configuration;
[0017] S35, applying the optimal parameter configuration to the constructed sales forecasting model, using time series forecasting technology to forecast trend data, and obtaining sales trends at future moments;
[0018] S36. Convert the obtained future sales trend into actual sales risk assessment, construct a final sales forecasting model, and use the final sales forecasting model to make real-time forecasts of future electronic product sales trends.
[0019] Furthermore, obtaining sales data related to electronic products and preprocessing the sales data to obtain sales volume characteristic data includes the following steps:
[0020] S11, obtaining duplicate data, missing values and outliers of sales data related to electronic products, and performing denoising, filtering and smoothing on the duplicate data, missing values and outliers;
[0021] S12, connecting the unprocessed data in the sales data to generate a new data table, associating different data tables through external key values, generating a complete data table, and obtaining an accurate data set;
[0022] S13. Extract features from the accurate data set and obtain sales volume feature data.
[0023] Further, using the factor analysis algorithm to analyze the obtained sales volume characteristic data to obtain the key factors affecting the sales volume of electronic products includes the following steps:
[0024] S21. Identify the key factors that have the greatest impact on sales volume from sales volume feature data through factor analysis algorithms;
[0025] S22. Determine the indicators for analyzing the sales efficiency of electronic products, and set the efficiency threshold and corresponding improvement technology;
[0026] S23. If the sales efficiency caused by the key factors is lower than the threshold, the key factors are adjusted using the set improvement technology;
[0027] S24, establishing an optimization model for key factors in the electronic product sales environment, and determining the objective function and constraint conditions of the optimization model;
[0028] S25. Apply the factor analysis algorithm to the optimization model to find the best solution that meets the constraints and use it as a basis for determining the key factors that affect the sales volume of electronic products.
[0029] Furthermore, applying the factor analysis algorithm to the optimization model to find the best solution that meets the constraints and as a basis for determining the key factors affecting the sales volume of electronic products includes the following steps:
[0030] S251, processing the identified key factors affecting the sales of electronic products to meet preset conditions and obtain processed key factor data;
[0031] S252, initializing the parameters of the factor analysis algorithm and setting the initial electronic product sales volume optimization improvement technology;
[0032] S253. Evaluate the specific impact of each improved technology on sales volume through the established objective function, and update the status according to the results;
[0033] S254. Optimize, evaluate and adjust the improvement technology of key factors affecting the sales of electronic products, and continuously update the status according to the preset conditions;
[0034] S255. If the preset sales efficiency reaches the preset threshold, determine the best improvement technology as the key factor affecting the sales volume of the electronic product; if the sales efficiency does not reach the preset threshold, continue to optimize and adjust the improvement technology.
[0035] Further, based on the trend indicators, the parameters of the sales forecast model are adjusted and optimized using the forecast model optimization algorithm, and finding the best parameter configuration includes the following steps:
[0036] S341, initializing key parameters of the prediction model optimization algorithm;
[0037] S342, randomly initializing the parameters of the sales forecasting model, and calculating the initial performance index of each parameter configuration;
[0038] S343. Calculate the relative importance of the parameters by analyzing the impact of each parameter on the forecast result, and determine the priority of adjusting the sales forecast model parameters according to the relative importance;
[0039] S344. Update the sales forecast model parameter configuration according to the relative importance and trend indicators, and randomly perturb the optimal parameter configuration;
[0040] S345. Re-evaluate the accuracy and effectiveness of the sales forecast model based on the updated parameter configuration;
[0041] S346. If the maximum number of iterations is reached, the optimization is stopped; otherwise, the optimization is continued and the parameter configuration is adjusted;
[0042] S347. When the optimization process is completed, the best parameter configuration and the optimized sales forecasting model are output.
[0043] Furthermore, the formula for updating the sales forecast model parameter configuration is:
[0044] ;
[0045] In the formula, Indicates the current parameter configuration;
[0046] Represented as the optimal parameter configuration;
[0047] It is expressed as the attractiveness coefficient;
[0048] Expressed as step size factor;
[0049] Represented as a random number.
[0050] Further, using a trend analysis algorithm to analyze the obtained electronic product sales trend, identifying the electronic product sales change pattern, and issuing an early warning reminder based on the identified electronic product sales change pattern includes the following steps:
[0051] S41. Use heuristic algorithms to identify key sales trend factors that affect the sales trend of electronic products;
[0052] S42, defining the transition probability of changes in key sales trend factors and setting the maximum number of iterations to continuously analyze and update sales trends;
[0053] S43, randomly generating random numbers for all key sales trend factors, and if the random numbers are greater than the transfer probability, globally optimizing the key sales trend factors, otherwise locally optimizing;
[0054] S44, check whether the preset number of iterations has been reached, if so, proceed to step S45, otherwise, return to step S43 to continue iterating;
[0055] S45. Output the global optimal sales trend model after evaluation and optimization, adjust the sales of electronic products according to the global optimal sales trend model, and implement early warning to respond to future market changes in advance.
[0056] Further, using the heuristic algorithm to identify key sales trend factors affecting the sales trend of electronic products includes the following steps:
[0057] S411, randomly initializing a set of values of key sales trend factors in the defined search space, and setting parameters of the heuristic algorithm;
[0058] S412, randomly assigning an initial value to each key sales trend factor, and finding the current optimal solution through fitness evaluation;
[0059] S413, adjusting the value of each key sales trend factor, simulating the change frequency, change amount and set value change of the key sales trend factor;
[0060] S414, generating random numbers, determining whether to randomly perturb the current optimal sales trend factor according to the frequency of change, and exploring new potential solutions;
[0061] S415, generate another random number, if it is less than the current change frequency, accept the new solution generated in step S414;
[0062] S416. Sort the fitness values of all key sales trend factors to identify the current optimal solution and optimal value;
[0063] S417, repeating steps S412 to S415 until the set maximum number of iterations is reached, and outputting the global optimal solution and optimal value as key sales trend factors affecting the sales trend of electronic products.
[0064] Furthermore, the formula for simulating the change frequency, change amount and set value change of key sales trend factors is:
[0065] ;
[0066] ;
[0067] ;
[0068] In the formula, Indicates z The frequency of changes in the key sales trend factors at the current moment;
[0069] , Respectively represent the maximum and minimum values of the frequency of change;
[0070] G represents the coefficient of inertia;
[0071] Indicates z Key sales trend factors t The set value at the moment;
[0072] Indicates z Key sales trend factors t The amount of change at a moment;
[0073] Represents a uniformly distributed random number;
[0074] Indicates the optimal solution for setting value.
[0075] According to another aspect of the present invention, a real-time monitoring system for sales data based on data analysis is also provided. The real-time monitoring system for sales data based on data analysis comprises:
[0076] A data acquisition module is used to acquire sales data related to electronic products and pre-process the sales data to obtain sales volume characteristic data;
[0077] A data analysis module is used to analyze the obtained sales volume characteristic data using a factor analysis algorithm to obtain key factors affecting the sales volume of electronic products;
[0078] A model building module is used to build a sales forecasting model based on key factors and evaluate the sales trend of electronic products in real time through the sales forecasting model;
[0079] A trend analysis module, used to analyze the obtained electronic product sales trend using a trend analysis algorithm, identify the electronic product sales change pattern, and issue an early warning reminder based on the identified electronic product sales change pattern;
[0080] The data acquisition module is connected to the model building module through the data analysis module, and the model building module is connected to the trend analysis module.
[0081] The beneficial effects of the present invention are:
[0082] 1. Through real-time analysis of sales data, the present invention enables enterprises to quickly identify market trends and consumer preferences, thereby making more accurate business decisions. Real-time evaluation helps adjust marketing strategies and product supply chains and optimize resource allocation. Through sales forecasting models, enterprises can more accurately predict future sales trends and effectively manage inventory, which helps reduce inventory backlogs and reduce obsolescence risks, ensuring that inventory is synchronized with market demand. In addition, real-time monitoring of sales data and trend analysis helps to better understand customer needs, and timely adjust product features or services to improve customer satisfaction and loyalty, thereby enhancing the market adaptability and flexibility of the enterprise, and significantly improving operational efficiency and customer service quality.
[0083] 2. The present invention identifies the key factors that have the greatest impact on sales volume through factor analysis algorithms, helping companies to more accurately understand the driving force behind sales, thereby making targeted strategic adjustments. By setting efficiency thresholds and optimization models, the sales process is made more efficient. When sales efficiency is lower than expected, improvement measures are taken quickly to reduce the risks caused by insufficient efficiency, continuously evaluate and adjust key sales factors, achieve continuous optimization, and ensure that sales strategies always keep pace with market demand, thereby effectively improving sales efficiency and market competitiveness.
[0084] 3. The present invention uses trend indicators and periodic evaluation adjustments to ensure that the sales forecast model can reflect market changes in real time, enabling enterprises to respond quickly to fluctuations in market demand. The introduced error indicator serves as a standard for evaluating the sales forecast model, helping to optimize the algorithm to reduce forecast errors and improve the accuracy and efficiency of decision-making. In addition, the parameter optimization process ensures the accuracy and reliability of the sales forecast model. By continuously adjusting and optimizing parameters, the sales forecast model can more accurately predict future sales trends, thereby providing enterprises with feasible decision-making support.
[0085] 4. The present invention initializes key sales trend factors in the defined search space, uses heuristic algorithms to continuously adjust and optimize these factors, accurately identifies the factors that have the greatest impact on sales volume, and dynamically adjusts sales strategies to cope with rapid changes in market demand by simulating the change frequency, change amount, and set value of key sales trend factors. It also implements early warnings based on analysis results, responds to potential market risks and sales challenges in a timely manner, and helps companies prepare and make adjustments in advance. Through the prediction model optimization algorithm, model parameters can be adjusted and optimized to improve the accuracy and efficiency of prediction. This provides companies with more accurate future sales trend forecasts, and accurate trend analysis and timely strategy adjustments enable companies to more effectively cope with competition, increase market share and customer satisfaction, and ultimately enhance the overall competitiveness of the company. BRIEF DESCRIPTION OF THE DRAWINGS
[0086] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0087] Figure 1 is a flow chart of a method for real-time monitoring of sales data based on data analysis according to an embodiment of the present invention;
[0088] Figure 2 It is a principle block diagram of a sales data real-time monitoring system based on data analysis according to an embodiment of the present invention.
[0089] In the figure:
[0090] 1. Data acquisition module; 2. Data analysis module; 3. Model building module; 4. Trend analysis module. DETAILED DESCRIPTION
[0091] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.
[0092] In the description of the present invention, unless otherwise specified, the meaning of "plurality" is two or more. In addition, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.
[0093] According to an embodiment of the present invention, a method and system for real-time monitoring of sales data based on data analysis are provided.
[0094] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to the real-time monitoring method of sales data based on data analysis according to an embodiment of the present invention, the real-time monitoring method of sales data based on data analysis includes the following steps:
[0095] S1. Acquire sales data related to electronic products, and pre-process the sales data to obtain sales volume characteristic data;
[0096] Specific sales data related to electronic products include:
[0097] Sales quantity data: sales quantity of different products, including data broken down by day, week, and month.
[0098] Sales data: The total sales of products can help analyze the market performance and profitability of products.
[0099] Pricing information: A record of the product's selling price, including historical price changes and prices during special offers or discounts.
[0100] Customer purchasing behavior data: including customer purchase frequency, purchase time, purchase channel (online or offline), etc.
[0101] Inventory data: Inventory levels of products, including inventory by region or point of sale.
[0102] Market feedback and user reviews: Customer reviews, ratings and feedback on products. This information can be used to analyze the product's market acceptance and consumer satisfaction.
[0103] Return and exchange data: product return rate and exchange status, used to analyze product quality issues or consumer satisfaction.
[0104] Promotional activity data: detailed information on various promotional activities, including promotion time, promotion type (such as discount, buy one get one free, etc.), and evaluation of the effectiveness of promotional activities.
[0105] Competitor data: Sales data of similar products of major competitors, including price, promotion activities and market share.
[0106] S2. Analyze the obtained sales volume characteristic data using a factor analysis algorithm to obtain key factors affecting the sales volume of electronic products;
[0107] Specifically, the key factors affecting the sales volume of electronic products include:
[0108] Product price: The pricing strategy of a product has a significant impact on sales. Consumers who are highly price sensitive may change their purchasing decisions when prices change.
[0109] Product features and innovation: Product features such as functionality, design, and technological innovation are important factors that influence consumers' purchasing decisions.
[0110] Market competition: Competitors' product prices, features and marketing activities can have an impact on the sales volume of this product.
[0111] Advertising and Promotion: Effective advertising and promotion can increase product visibility and appeal, thereby increasing sales.
[0112] Consumer reviews and word of mouth: Positive user reviews and good word of mouth can boost the confidence of potential buyers and promote sales.
[0113] Economic environment: Macroeconomic conditions, such as consumer income levels and economic growth rates, also affect the sales of electronic products.
[0114] Seasonality and Holiday Factors: Sales of certain electronics products may be affected by seasons and holidays, such as sales surges during the holiday season.
[0115] Channel and distribution strategy: Effective management of sales channels and optimization of distribution strategies are also key to sales success.
[0116] Social Trends and Technological Advancements: Sociocultural trends and technological advancements can change consumer demands and thus affect the sales of specific types of electronic products.
[0117] S3. Establish a sales forecasting model based on key factors, and use the sales forecasting model to evaluate the sales trend of electronic products in real time;
[0118] Specifically, the sales trend of electronic products refers to the changes and development patterns of key indicators such as sales volume and sales revenue of electronic products within a certain period of time, including:
[0119] Time Series Trend: A long-term trend in sales data that increases or decreases over time. For example, a product may have an increasing sales trend as its market acceptance increases over time.
[0120] Seasonality: Periodic fluctuations due to seasonal changes, holidays, or special events. For example, sales of electronics may increase during the holiday shopping season or back-to-school season.
[0121] Market Saturation: As a market becomes increasingly saturated, the rate of introduction of new products may slow and sales of older products may begin to decline.
[0122] Competitive dynamics: Competitors’ activities such as new product launches, price adjustments, etc. can also affect sales trends.
[0123] Changes in consumer behavior: Technological advances or changes in consumer preferences, such as an increased preference for environmentally friendly or high-tech products, can also change sales trends.
[0124] External economic factors: Changes in the economic environment, such as recession or growth, inflation or changes in currency value, can also affect consumer purchasing power and sales trends.
[0125] S4, analyzing the obtained electronic product sales trend using a trend analysis algorithm, identifying a sales change pattern of the electronic products, and issuing an early warning reminder based on the identified sales change pattern of the electronic products;
[0126] Specifically, the sales change patterns of electronic products refer to the various regular changes reflected in the sales data. These change patterns help companies understand how and when sales will change so that they can make timely adjustments and responses, including:
[0127] Trend change pattern: refers to the long-term upward or downward trend of sales over time. For example, the sales of a certain type of smart device may show a continuous upward trend due to rapid technological updates.
[0128] Cyclical patterns: Sales fluctuations that are associated with specific time periods (e.g., seasons, holidays). For example, sales of video games and toys may increase significantly around Christmas.
[0129] Abnormal change patterns: refers to sales activities that do not conform to normal expectations, such as sudden sales peaks or troughs caused by promotional activities or external events (such as economic crisis, health crisis).
[0130] Saturation and Decline Pattern: As market saturation increases, sales of certain products begin to decline, a pattern that indicates the product may have reached the peak of its life cycle.
[0131] Among them, based on key factors, establishing a sales forecasting model and evaluating the sales trend of electronic products in real time through the sales forecasting model include the following steps:
[0132] S31. Convert the obtained key factors into risk assessment cycle;
[0133] S32, replacing the risk assessment cycle into the cycle index adjustment algorithm, dividing the obtained key factors into cycle indicators and trend indicators, and constructing and training a sales forecasting model through the cycle indicators and trend indicators to predict future sales trends;
[0134] S33. Use the error index (i.e., absolute percentage error) as the evaluation criterion for the accuracy of the sales forecast model;
[0135] S34. Based on the trend indicators, use the forecast model optimization algorithm to adjust and optimize the parameters of the sales forecast model and find the best parameter configuration;
[0136] S35. Apply the optimal parameter configuration to the constructed sales forecasting model, use the time series forecasting technology (i.e., the second-order adaptive coefficient method) to forecast the trend data, and obtain the sales trend at the future moment;
[0137] S36. Convert the obtained future sales trend into actual sales risk assessment, construct a final sales forecasting model, and use the final sales forecasting model to make real-time forecasts of future electronic product sales trends.
[0138] It should be explained that, taking the prediction of smartphone sales trends as an example, in the smartphone market, key factors such as the introduction of new technologies, competitor price adjustments, and seasonal purchasing patterns are identified. The impact of these key factors is periodized using historical data statistical analysis. For example, the impact period of the introduction of new technologies is defined as the first three months after the product is released. The identified cyclical purchasing patterns (such as the Christmas shopping season) and technological trends (such as the popularization of 5G networks) are used as cycle indicators and trend indicators, respectively. A cyclical index adjustment algorithm, such as seasonally decomposed time series (STL), is used to separate and emphasize the role of these indicators in the forecasting model. STL) is used to separate and emphasize the role of these indicators in the forecasting model. The model accuracy is evaluated by calculating the mean square error (MSE) and absolute percentage error (MAPE) between the predicted value and the actual sales data. Based on trend indicators, such as the market acceptance speed of 5G network popularization, the forecasting model parameters are adjusted using optimization algorithms such as the firefly algorithm, genetic algorithm, or gradient descent method to find the best parameter configuration. The optimized parameter configuration is applied to time series forecasting models such as ARIMA or LSTM to predict the sales trend of smartphones in the next three months. The forecast results are converted into risk assessment. For example, the forecast results show that a future sales decline may lead to inventory backlogs, thereby triggering a risk warning.
[0139] Preferably, acquiring sales data related to electronic products and preprocessing the sales data to obtain sales volume characteristic data comprises the following steps:
[0140] S11, obtaining duplicate data, missing values and outliers of sales data related to electronic products, and performing denoising, filtering and smoothing on the duplicate data, missing values and outliers;
[0141] S12, connecting the unprocessed data in the sales data to generate a new data table, associating different data tables through external key values, generating a complete data table, and obtaining an accurate data set;
[0142] S13. Extract features from the accurate data set and obtain sales volume feature data.
[0143] It should be explained that, taking smartphone sales data as an example, the sales data of smartphones are classified by date and time period (such as morning and evening every day), and the sales quantity in each time period is calculated. This helps to identify the peak sales period of the day and detect duplicate data, missing values and outliers. Duplicate data will be deleted to avoid repeated counting; missing values can be filled by interpolation methods, such as using the average sales number of the previous and next time points; outlier detection can be implemented by analyzing the statistical distribution of sales quantity, and outliers can be replaced or deleted. The sales data in the morning and evening are combined to generate a complete data table. This table will include an overview of sales for each day, help identify which days or time periods have better sales performance, and extract key features such as the average sales quantity per day, the maximum sales quantity, and the fluctuation range from the complete data set. These features are very valuable for analyzing sales trends and formulating sales strategies. The extracted features are standardized to eliminate the impact of data of different magnitudes and ensure the consistency and comparability of the data. This step is particularly important for subsequent data analysis and machine learning models.
[0144] Preferably, using a factor analysis algorithm to analyze the obtained sales volume characteristic data to obtain key factors affecting the sales volume of electronic products includes the following steps:
[0145] S21. Identify the key factors that have the greatest impact on sales volume from sales volume feature data through factor analysis algorithms;
[0146] S22. Determine the indicators for analyzing the sales efficiency of electronic products, and set the efficiency threshold and corresponding improvement technology;
[0147] S23. If the sales efficiency caused by the key factors is lower than the threshold, the key factors are adjusted using the set improvement technology;
[0148] S24, establishing an optimization model for key factors in the electronic product sales environment, and determining the objective function and constraint conditions of the optimization model;
[0149] S25. Apply the factor analysis algorithm to the optimization model to find the best solution that meets the constraints and use it as a basis for determining the key factors that affect the sales volume of electronic products.
[0150] It should be explained that, taking smartphone sales as an example, key data such as price, promotional activities, market trends, competitor behavior, etc. are collected from the sales feature data set, and factor analysis algorithms such as principal component analysis (PCA) and fish swarm algorithm are applied to identify the factors that most significantly affect sales. Key indicators such as sales conversion rate and inventory turnover rate are defined to measure sales efficiency. Based on historical data and industry standards, efficiency thresholds for each indicator are set to determine which factors cause sales efficiency to be lower than the threshold, such as changing pricing strategies, increasing promotional activities, etc., and adjusting inefficient factors that affect sales efficiency. An optimization model is constructed, including objective functions (such as maximizing sales profits) and constraints (such as budgets and resource constraints). The optimization objectives of the optimization model and the constraints that need to be considered in actual operations are clarified, the key factors obtained from the factor analysis are input into the optimization model, and mathematical optimization methods (such as linear programming and integer programming) are used to find the optimal solution that meets the constraints.
[0151] Specifically, the factor analysis algorithm is the fish school algorithm, which is an optimization algorithm based on swarm intelligence. It is inspired by the foraging, clustering and following behaviors of fish schools in nature. The algorithm imitates the collective behavior of fish schools in finding food and avoiding danger.
[0152] Preferably, applying the factor analysis algorithm to the optimization model to find the best solution that meets the constraints and using it as a basis for determining the key factors affecting the sales volume of electronic products includes the following steps:
[0153] S251. Process (whitening) the identified key factors affecting the sales of electronic products to meet preset conditions and obtain processed key factor data;
[0154] S252, initializing the parameters of the factor analysis algorithm (i.e., initializing the parameters of the fish school algorithm), and setting the initial electronic product sales volume optimization improved technology (i.e., artificial fish);
[0155] S253. Evaluate the specific impact of each improved technology on sales volume through the established objective function, and update the status according to the results;
[0156] S254. Optimize (i.e., foraging behavior in fish school algorithm), evaluate and adjust (i.e., clustering behavior in fish school algorithm) the improved technology of key factors affecting the sales of electronic products, and continuously update the status according to preset conditions (i.e., tail chasing behavior in fish school algorithm);
[0157] S255. If the preset sales efficiency reaches the preset threshold, determine the best improvement technology as the key factor affecting the sales volume of electronic products. If the sales efficiency does not reach the preset threshold, continue to optimize and adjust the improvement technology (i.e., improvement strategy or improvement measure).
[0158] It should be explained that, taking smartphone sales as an example, after identifying the key factors affecting smartphone sales, whitening processing is performed, that is, the data is standardized to eliminate the correlation between factors and make each factor independent. Data transformation techniques, such as Z-score standardization, are used to ensure that each factor has zero mean and unit variance. Parameters in the fish school algorithm are set, such as the number of fish, field of view, step length, etc., and initial sales improvement techniques (artificial fish) are defined, such as price adjustment, increased promotion, etc. An objective function is defined to evaluate the impact of different improvement techniques on sales. According to the evaluation results of the objective function, the state and position of the artificial fish are updated to simulate the fish looking for an area with a higher objective function value within the perception range. If there is no better area around, the fish will approach other fish and collectively look for the optimal solution. The fish school will follow the fish that finds the best solution and move in that direction as a whole. If the preset sales efficiency threshold is reached, the current best improvement technology is accepted. If the threshold is not reached, the iterative optimization process continues.
[0159] Preferably, based on the trend indicator, adjusting and optimizing the parameters of the sales forecast model using a forecast model optimization algorithm and finding the best parameter configuration includes the following steps:
[0160] S341, initializing key parameters of the prediction model optimization algorithm;
[0161] It needs to be explained that the application example is: choose an algorithm suitable for parameter optimization, such as genetic algorithm, firefly algorithm or particle swarm optimization (PSO), and initialize its key parameters such as population size, number of iterations, crossover rate and mutation rate (genetic algorithm) or particle speed and position (PSO).
[0162] Theoretical basis: Initialization parameter setting is the basis of the optimization process. Correct parameter setting can effectively improve the optimization efficiency and success rate.
[0163] S342, randomly initializing the parameters of the sales forecasting model, and calculating the initial performance index of each parameter configuration;
[0164] It should be explained that the application example is: a set of initial parameter configurations are randomly generated in the parameter space, each parameter configuration represents a possible solution. The evaluation value of the sales forecast model under each parameter configuration is calculated and used as the basis for evaluating the quality of the parameter configuration.
[0165] Theoretical basis: Random initialization helps explore the parameter space, ensures the diversity of solutions, and prevents the algorithm from converging to the local optimal solution too early.
[0166] S343. Calculate the relative importance of the parameters by analyzing the impact of each parameter on the forecast result, and determine the priority of adjusting the sales forecast model parameters according to the relative importance;
[0167] It should be explained that the application example is: using methods such as sensitivity analysis or influence assessment to calculate the influence of each parameter on the output (evaluation value) of the sales forecast model to determine which parameters are more critical.
[0168] Rationale: Relative importance assessment can help determine which parameters should be adjusted first during the optimization process to more effectively improve sales forecasting model performance.
[0169] S344. Update the sales forecast model parameter configuration according to the relative importance and trend indicators, and randomly perturb the optimal parameter configuration;
[0170] It is necessary to explain that the application example is: according to the relative importance of each parameter and the observed trend, the parameter value is adjusted. This step can be combined with random perturbations to avoid the prediction model optimization algorithm from falling into the local optimum.
[0171] Theoretical basis: Parameter updating and perturbation are key strategies to maintain optimization dynamics and avoid local optimality.
[0172] S345. Re-evaluate the accuracy and effectiveness of the sales forecast model based on the updated parameter configuration;
[0173] It should be explained that the application example is: using the updated parameter configuration, re-running the sales forecast model, and using pre-set evaluation indicators (such as MAPE) to evaluate the accuracy and effectiveness of the sales forecast model.
[0174] S346. If the maximum number of iterations is reached, the optimization is stopped; otherwise, the optimization is continued and the parameter configuration is adjusted;
[0175] It should be explained that the application example is: repeatedly executing the parameter update and sales forecast model evaluation process until the set maximum number of iterations is reached or the sales forecast model performance is no longer significantly improved.
[0176] Theoretical basis: The iterative process helps to fine-tune the parameters and gradually approach the optimal solution.
[0177] S347. When the optimization process is completed, the best parameter configuration and the optimized sales forecasting model are output.
[0178] It should be explained that in the application example, after the optimization process is completed, the parameter configuration that achieves the best performance index and the corresponding sales forecast model version are output, and this version will be used for actual index evaluation.
[0179] Theoretical basis: The final output sales forecast model should have high accuracy and reliability, and be able to support actual inspection indicator safety management decisions.
[0180] Specifically, the prediction model optimization algorithm is the firefly algorithm, which contains two elements, namely brightness and attractiveness. Brightness reflects the pros and cons of the firefly's position and determines its movement direction, and attractiveness determines the distance the firefly moves. By continuously updating brightness and attractiveness, the target optimization is achieved.
[0181] Preferably, the formula for updating the sales forecast model parameter configuration is:
[0182] ;
[0183] In the formula, Indicates the current parameter configuration;
[0184] Represented as the optimal parameter configuration;
[0185] It is expressed as the attractiveness coefficient;
[0186] Specifically, the attractiveness coefficient determines the degree to which the current parameter configuration moves toward a better parameter configuration during parameter update.
[0187] Expressed as step size factor;
[0188] Specifically, the step factor is a [0, 1] constant, which is used to determine the scale of random perturbations and help the prediction model optimization algorithm escape from the local optimal solution.
[0189] Represented as a random number.
[0190] Specifically, is a random number uniformly distributed on [0, 1], which is used to introduce randomness and increase the diversity of the search.
[0191] It should be explained that in the formula for updating the sales forecast model parameter configuration, rand -0.5 is to introduce randomness, which can help the algorithm avoid falling into the local optimal solution and enable more extensive exploration in the search space.
[0192] It needs to be explained that rand is a uniformly distributed random number in the range [0, 1], and its expected value is 0.5. rand When 0.5 is subtracted from randThe range of -0.5 becomes [-0.5, 0.5], and its expected value becomes 0. After this processing, the center position of the random number is near 0, which allows the algorithm to have the opportunity to move towards the "good" solution during the search process, as well as the opportunity to move away from the current position and explore other possible solutions. In the firefly algorithm, the step size factor and rand -0.5 multiplication means that the algorithm will introduce a certain degree of random exploration in each update step, which helps the algorithm to jump out of the local optimal solution and find the global optimal solution. rand The expected value of -0.5 is 0. In the long run, this randomness will not have a negative impact on the convergence trend of the algorithm.
[0193] Preferably, analyzing the obtained electronic product sales trend using a trend analysis algorithm, identifying the electronic product sales change pattern, and issuing an early warning reminder based on the identified electronic product sales change pattern includes the following steps:
[0194] S41. Use heuristic algorithms to identify key sales trend factors that affect the sales trend of electronic products;
[0195] Specifically, the inspiration algorithm is the bat algorithm, which is an optimization algorithm based on swarm intelligence and is inspired by the echolocation behavior of bats. Bats emit sound waves when flying and hunting, and locate obstacles and prey by receiving the reflected sound waves. In the bat algorithm, this behavior is abstracted to find the optimal solution in a complex search space. The key components of the bat algorithm include:
[0196] 1) Sound wave emission frequency:
[0197] Each bat uses different frequencies, speeds, and loudness to detect targets and obstacles during the search process, which helps to dynamically adjust the search range and refine the potential solution area.
[0198] 2) Echo loudness and pulse emission rate:
[0199] The initial loudness value is usually high, and as a better solution is found, the loudness decreases to reduce the search range and focus around the optimal solution.
[0200] The pulse firing rate increases as the target is approached, which helps to update the position more frequently to approach the global optimal solution.
[0201] 4) Speed and position update:
[0202] The bat adjusts its flight speed and direction based on its frequency, and the position and velocity update formulas in the algorithm are used to simulate this behavior.
[0203] S42, defining the transition probability of changes in key sales trend factors and setting the maximum number of iterations to continuously analyze and update sales trends;
[0204] S43, randomly generating random numbers for all key sales trend factors, and if the random numbers are greater than the transfer probability, globally optimizing the key sales trend factors, otherwise locally optimizing;
[0205] S44, check whether the preset number of iterations has been reached, if so, proceed to step S45, otherwise, return to step S43 to continue iterating;
[0206] S45. Output the global optimal sales trend model after evaluation and optimization, adjust the sales of electronic products according to the global optimal sales trend model, and implement early warning to respond to future market changes in advance.
[0207] It should be explained that, taking smartphone sales as an example, suitable heuristic algorithms are selected, such as genetic algorithms, bat algorithms or simulated annealing, to explore key trend factors in sales data. Various possible combinations of factors are explored through algorithms to evaluate their impact on sales trends. Transition probabilities are defined for each key factor. These probabilities determine the possibility of factors transitioning between different states. A maximum number of iterations is set to systematically analyze and update data. A random number is generated for each factor. If the random number is greater than the set transition probability, global optimization is performed; otherwise, local optimization is performed to evaluate whether the current number of iterations has reached the preset maximum number. Based on the iteration results, the evaluated and optimized global optimal sales trend model is output. Based on this model, the smartphone market strategy is adjusted, and necessary early warning systems are implemented to cope with future market changes. The optimization results are actually applied to sales strategies, which can respond to market changes in advance and optimize sales performance.
[0208] Specifically, the trend analysis algorithm is an improvement of the pollen algorithm based on the bat algorithm. In view of the defects of the pollen algorithm, such as easy to fall into the local optimum, slow convergence speed, and low convergence accuracy, the bat algorithm is introduced into the pollen algorithm to optimize the initial position of the pollen gametes and establish a pollen algorithm based on the bat algorithm. The bat algorithm generates local new solutions near the optimal solution through random flight, which can enable the pollen algorithm to have a larger search range at the initial moment and strengthen the overall local search ability of the algorithm. Therefore, the bat algorithm is used to optimize the initial value of the pollen algorithm to improve the quality of the pollen algorithm initialization population.
[0209] Specifically, the pollen algorithm, also known as the Flower Pollination Algorithm (FPA), is an optimization algorithm that simulates the pollination process of flowers in nature. It mainly imitates the global and local search behavior of pollen through pollination mechanisms in nature (such as wind or insect pollination). The key features of the pollen algorithm include:
[0210] 1) Global pollination and local pollination:
[0211] Global pollination: Simulating pollination over long distances by long-distance dispersal mechanisms such as insects or wind. This process often involves random walks between pollen grains, which facilitates global exploration in the search space.
[0212] Local pollination: simulates pollen spreading within a smaller area through itself (self-pollination) or pollen from its neighbors (close-range pollination). This process exploits local information in the search space and helps fine-tune the solution.
[0213] 2) Switching probability:
[0214] There is a switch probability in the algorithm to decide whether to perform global or local pollination. This probability is usually related to the size and complexity of the problem.
[0215] 3) Fitness function:
[0216] After each pollination, the fitness of the new solution is calculated to determine whether to adopt the new solution. A solution with a higher fitness usually means it is closer to the optimal solution.
[0217] Preferably, using a heuristic algorithm to identify key sales trend factors affecting the sales trend of electronic products includes the following steps:
[0218] S411, randomly initializing a set of values of key sales trend factors in the defined search space, and setting parameters of the heuristic algorithm;
[0219] S412, randomly assigning an initial value to each key sales trend factor, and finding the current optimal solution through fitness evaluation;
[0220] S413, adjusting the value of each key sales trend factor, simulating the change of the change frequency (i.e., the pulse frequency in the bat algorithm), the change amount (i.e., the speed in the bat algorithm), and the set value (i.e., the position in the bat algorithm) of the key sales trend factor (i.e., the bat in the bat algorithm);
[0221] S414, generating random numbers, determining whether to randomly perturb the current optimal sales trend factor according to the frequency of change, and exploring new potential solutions;
[0222] S415, generate another random number, if it is less than the current change frequency, accept the new solution generated in step S414;
[0223] S416. Sort the fitness values of all key sales trend factors to identify the current optimal solution and optimal value;
[0224] S417, repeating steps S412 to S415 until the set maximum number of iterations is reached, and outputting the global optimal solution and optimal value as key sales trend factors affecting the sales trend of electronic products.
[0225] It should be explained that, taking smartphone sales as an example, in the defined search space, the values of key influencing factors of smartphone sales (such as price, marketing budget, market penetration, etc.) are randomly initialized, and the basic parameters of the bat algorithm, including pulse frequency, speed, sound wave loudness, etc., are set. Assign initial random values to each influencing factor, and use fitness functions (such as sales volume, customer satisfaction) to evaluate the performance of each parameter combination and determine the current optimal solution. According to the update formula of the bat algorithm, simulate the change frequency, change amount and adjustment of the set value of the key influencing factors. These adjustments simulate the process of actual bats adjusting their flight behavior to optimize hunting efficiency. Generate a random number. If it is greater than the current change frequency, randomly perturb the current optimal influencing factor to generate a new solution. Generate another random number. If it is less than the current change frequency, decide whether to accept this new solution based on the fitness evaluation results. Sort the fitness values of all influencing factors, find the best performing solution and the corresponding optimal value, until the set maximum number of iterations or other predefined stop conditions (such as no significant improvement in fitness) are reached. The key to the bat algorithm is that it can find the optimal solution by simulating the hunting behavior of bats in nature (using sound waves to locate and adjust flight patterns). This algorithm is particularly suitable for solving complex optimization problems with multiple local optimal solutions, because it combines global search and local exploration strategies, which can effectively avoid falling into local optimal solutions and missing better global solutions. This process can not only maximize the market performance of smartphones, but also improve the effectiveness of the overall sales strategy through continuous parameter optimization.
[0226] Preferably, the formula for simulating the change frequency, change amount and change of the set value of the key sales trend factor is:
[0227] ;
[0228] ;
[0229] ;
[0230] In the formula, Indicates zThe frequency of changes in the key sales trend factors at the current moment;
[0231] , Respectively represent the maximum and minimum values of the frequency of change;
[0232] G represents the coefficient of inertia;
[0233] Indicates z Key sales trend factors t The set value at the moment;
[0234] Indicates z Key sales trend factors t The amount of change at a moment;
[0235] Represents a uniformly distributed random number;
[0236] Indicates the optimal solution for setting value.
[0237] According to another embodiment of the present invention, Figure 2 As shown, a real-time monitoring system for sales data based on data analysis is also provided, and the real-time monitoring system for sales data based on data analysis includes:
[0238] Data acquisition module 1, used to acquire sales data related to electronic products, and pre-process the sales data to obtain sales volume characteristic data;
[0239] Data analysis module 2, used to analyze the obtained sales volume characteristic data using factor analysis algorithm to obtain key factors affecting the sales volume of electronic products;
[0240] Model building module 3, used to build a sales forecasting model based on key factors, and to evaluate the sales trend of electronic products in real time through the sales forecasting model;
[0241] The trend analysis module 4 is used to analyze the obtained electronic product sales trend using a trend analysis algorithm, identify the electronic product sales change pattern, and issue an early warning reminder based on the identified electronic product sales change pattern;
[0242] The data acquisition module 1 is connected to the model building module 3 through the data analysis module 2 , and the model building module 3 is connected to the trend analysis module 4 .
[0243] In summary, with the help of the above technical solution of the present invention, the present invention identifies the key factors that have the greatest impact on sales volume through a factor analysis algorithm, helps enterprises to more accurately understand the driving force behind sales, and thus make targeted strategic adjustments, and by setting efficiency thresholds and optimization models, the sales process is made more efficient. When sales efficiency is lower than expected, improvement measures are quickly taken to reduce the risks caused by insufficient efficiency, and key sales factors are continuously evaluated and adjusted to achieve continuous optimization, ensuring that sales strategies always keep pace with market demand, thereby effectively improving sales efficiency and market competitiveness. The present invention uses trend indicators and periodic evaluation adjustments to ensure that the sales forecast model can reflect market changes in real time, so that enterprises can quickly respond to fluctuations in market demand. The introduced error indicators are used as the standard for evaluating the sales forecast model, helping the optimization algorithm to reduce prediction errors and improve the accuracy and efficiency of decision-making. In addition, the parameter optimization process ensures the accuracy and reliability of the sales forecast model. By continuously adjusting and optimizing parameters, the sales forecast model can more accurately predict future sales trends, thereby providing feasible decision support for enterprises. The present invention initializes key sales trend factors in the defined search space, uses heuristic algorithms to continuously adjust and optimize these factors, accurately identifies the factors that have the greatest impact on sales volume, and dynamically adjusts sales strategies to cope with rapid changes in market demand by simulating the change frequency, change amount, and set value of key sales trend factors. It also implements early warnings based on analysis results, responds to potential market risks and sales challenges in a timely manner, and helps companies prepare and make adjustments in advance. Through the prediction model optimization algorithm, model parameters can be adjusted and optimized to improve the accuracy and efficiency of prediction. This provides companies with more accurate future sales trend forecasts, and accurate trend analysis and timely strategy adjustments enable companies to more effectively cope with competition, increase market share and customer satisfaction, and ultimately enhance the overall competitiveness of the company.
[0244] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A real-time monitoring method for sales data based on data analysis, It is characterized in that The sales data real-time monitoring method based on data analysis includes the following steps: S1. Obtain sales data of electronic products, and pre-process the sales data to obtain sales volume characteristic data; S2. Analyze the obtained sales volume characteristic data using a factor analysis algorithm to obtain key factors affecting the sales volume of electronic products; S3. Establish a sales forecasting model based on key factors, and use the sales forecasting model to evaluate the sales trend of electronic products in real time; S4, analyzing the obtained electronic product sales trend using a trend analysis algorithm, identifying a sales change pattern of the electronic products, and issuing an early warning reminder based on the identified sales change pattern of the electronic products; The step of establishing a sales forecasting model based on key factors and evaluating the sales trend of electronic products in real time through the sales forecasting model includes the following steps: S31. Convert the obtained key factors into risk assessment cycle; S32, replacing the risk assessment cycle into the cycle index adjustment algorithm, dividing the obtained key factors into cycle indicators and trend indicators, and constructing and training a sales forecasting model through the cycle indicators and trend indicators to predict future sales trends; S33. Using error indicators as evaluation criteria for the accuracy of sales forecast models; S34. Based on the trend indicators, use the forecast model optimization algorithm to adjust and optimize the parameters of the sales forecast model and find the best parameter configuration; S35, applying the optimal parameter configuration to the constructed sales forecasting model, using time series forecasting technology to forecast trend data, and obtaining sales trends at future moments; S36, converting the obtained future sales trend into an actual sales risk assessment, constructing a final sales forecasting model, and making a real-time forecast of the future sales trend of electronic products through the final sales forecasting model; The method of adjusting and optimizing the parameters of the sales forecast model based on trend indicators using a forecast model optimization algorithm and finding the best parameter configuration includes the following steps: S341, initializing key parameters of the prediction model optimization algorithm; S342, randomly initializing the parameters of the sales forecasting model, and calculating the initial performance index of each parameter configuration; S343. Calculate the relative importance of the parameters by analyzing the impact of each parameter on the forecast result, and determine the priority of adjusting the sales forecast model parameters according to the relative importance; S344. Update the sales forecast model parameter configuration according to the relative importance and trend indicators, and randomly perturb the optimal parameter configuration; S345. Re-evaluate the accuracy and effectiveness of the sales forecast model based on the updated parameter configuration; S346. If the maximum number of iterations is reached, the optimization is stopped; otherwise, the optimization is continued and the parameter configuration is adjusted; S347. When the optimization process is completed, the optimal parameter configuration and the optimized sales forecast model are output; The method of analyzing the obtained electronic product sales trend by using a trend analysis algorithm, identifying the electronic product sales change pattern, and issuing an early warning reminder according to the identified electronic product sales change pattern includes the following steps: S41. Use heuristic algorithms to identify key sales trend factors that affect the sales trend of electronic products; S42, defining the transition probability of changes in key sales trend factors and setting the maximum number of iterations to continuously analyze and update sales trends; S43, randomly generating random numbers for all key sales trend factors, and if the random numbers are greater than the transfer probability, globally optimizing the key sales trend factors, otherwise locally optimizing; S44, check whether the preset number of iterations has been reached, if so, proceed to step S45, otherwise, return to step S43 to continue iterating; S45. Output the global optimal sales trend model after evaluation and optimization, adjust the sales of electronic products according to the global optimal sales trend model, and implement early warning to respond to future market changes in advance; The method of using a heuristic algorithm to identify key sales trend factors affecting the sales trend of electronic products includes the following steps: S411, randomly initializing a set of values of key sales trend factors in the defined search space, and setting parameters of the heuristic algorithm; S412, randomly assigning an initial value to each key sales trend factor, and finding the current optimal solution through fitness evaluation; S413, adjusting the value of each key sales trend factor, simulating the change frequency, change amount and set value of the key sales trend factor; S414, generating random numbers, determining whether to randomly perturb the current optimal sales trend factor according to the frequency of change, and exploring new potential solutions; S415, generate another random number, if it is less than the current change frequency, accept the new solution generated in step S414; S416. Sort the fitness values of all key sales trend factors to identify the current optimal solution and optimal value; S417, repeating steps S412 to S415 until the set maximum number of iterations is reached, and outputting the global optimal solution and optimal value as key sales trend factors affecting the sales trend of electronic products.
2. A method for real-time monitoring of sales data based on data analysis according to claim 1, It is characterized in that The step of acquiring sales data related to electronic products and preprocessing the sales data to obtain sales volume characteristic data comprises the following steps: S11, obtaining duplicate data, missing values and outliers of sales data related to electronic products, and performing denoising, filtering and smoothing on the duplicate data, missing values and outliers; S12, connecting the unprocessed data in the sales data to generate a new data table, associating different data tables through external key values, generating a complete data table, and obtaining an accurate data set; S13. Extract features from the accurate data set and obtain sales volume feature data.
3. A method for real-time monitoring of sales data based on data analysis according to claim 1, It is characterized in that The method of analyzing the obtained sales volume characteristic data by using a factor analysis algorithm to obtain key factors affecting the sales volume of electronic products comprises the following steps: S21. Identify the key factors that have the greatest impact on sales volume from sales volume feature data through factor analysis algorithms; S22. Determine the indicators for analyzing the sales efficiency of electronic products, and set the efficiency threshold and corresponding improvement technology; S23. If the sales efficiency caused by the key factors is lower than the threshold, the key factors are adjusted using the set improvement technology; S24, establishing an optimization model for key factors in the electronic product sales environment, and determining the objective function and constraint conditions of the optimization model; S25. Apply the factor analysis algorithm to the optimization model to find the best solution that meets the constraints and use it as a basis for determining the key factors that affect the sales volume of electronic products.
4. A method for real-time monitoring of sales data based on data analysis according to claim 3, It is characterized in that The method of applying the factor analysis algorithm to the optimization model to find the best solution that meets the constraint conditions and use it as a basis for determining the key factors that affect the sales volume of electronic products includes the following steps: S251, processing the identified key factors affecting the sales of electronic products to meet preset conditions and obtain processed key factor data; S252, initializing the parameters of the factor analysis algorithm and setting the initial electronic product sales volume optimization improvement technology; S253. Evaluate the specific impact of each improved technology on sales volume through the established objective function, and update the status according to the results; S254. Optimize, evaluate and adjust the improvement technology of key factors affecting the sales of electronic products, and continuously update the status according to the preset conditions; S255. If the preset sales efficiency reaches the preset threshold, determine the best improvement technology as the key factor affecting the sales volume of the electronic product; if the sales efficiency does not reach the preset threshold, continue to optimize and adjust the improvement technology.
5. A method for real-time monitoring of sales data based on data analysis according to claim 1, It is characterized in that The formula for updating the sales forecast model parameter configuration is: ; In the formula, Indicates the current parameter configuration; Represented as the optimal parameter configuration; It is expressed as the attractiveness coefficient; Expressed as step size factor; Represented as a random number.
6. A method for real-time monitoring of sales data based on data analysis according to claim 1, It is characterized in that The formula for simulating the change frequency, change amount and change of the set value of the key sales trend factor is: ; ; ; In the formula, Indicates z The frequency of changes in the key sales trend factors at the current moment; , Respectively represent the maximum and minimum values of the frequency of change; G represents the coefficient of inertia; Indicates z Key sales trend factors t The set value at the moment; Indicates z Key sales trend factors t The amount of change at a moment; Represents a uniformly distributed random number; Indicates the optimal solution for setting value.
7. A sales data real-time monitoring system based on data analysis, used to implement the sales data real-time monitoring method based on data analysis according to any one of claims 1 to 6, It is characterized in that The real-time sales data monitoring system based on data analysis includes: The data acquisition module is used to acquire the sales data of electronic products and pre-process the sales data to obtain sales volume characteristic data; A data analysis module is used to analyze the obtained sales volume characteristic data using a factor analysis algorithm to obtain key factors affecting the sales volume of electronic products; A model building module is used to build a sales forecasting model based on key factors and evaluate the sales trend of electronic products in real time through the sales forecasting model; A trend analysis module, used to analyze the obtained electronic product sales trend using a trend analysis algorithm, identify the electronic product sales change pattern, and issue an early warning reminder based on the identified electronic product sales change pattern; The data acquisition module is connected to the model building module through the data analysis module, and the model building module is connected to the trend analysis module.
