Data analysis method and system and medium

Abstract

The invention discloses a data analysis method and system and a medium thereof, and relates to the field of data analysis, and the method comprises the following specific steps: collecting data to be analyzed, drawing a time histogram according to classified and summarized data, displaying data information by using the time histogram, controlling the time histogram, and performing display control on the time histogram. Analyzing the data according to the time histogram, generating a data analysis report, performing data association based on the market basket, performing association analysis on the data, and generating a data analysis report. According to the method, a data analysis method is utilized, a time histogram display mode is adopted, time is taken as an X axis, and a numerical value is taken as a Y axis, so that the change trend of the sales data is clear at a glance, decision makers can quickly understand information behind the data, line drawing processing is performed on the time histogram, the change trend of the sales data can be clear at a glance, and the decision makers can quickly understand information behind the data. And a decision maker can quickly understand information behind the data.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to a data analysis method, system and medium thereof. Background Art

[0002] Sales data analysis is a vital part of corporate management and decision-making. By analyzing sales data, companies can gain an in-depth understanding of market trends, customer behavior, product performance, etc., thereby formulating more effective sales strategies and marketing plans.

[0003] With the advent of the big data era, the amount of data generated by enterprises in their daily operations has grown exponentially. Data analysis has become an important tool for corporate decision-making, especially the analysis of sales data, which is of great significance for optimizing operations, improving sales performance and formulating market strategies. Traditional data analysis methods often have some limitations, such as low data processing efficiency, lack of intuitive analysis results, and lack of real-time performance, which makes it impossible for enterprises to respond quickly to market changes.

[0004] Currently, many companies rely on static reports and simple statistical tools for sales data analysis, which makes it difficult for decision makers to extract valuable information from complex data. In addition, existing sales data analysis tools are usually unable to effectively integrate multi-dimensional data, resulting in one-sided and incomplete analysis results.

[0005] Therefore, a data analysis method, system and medium thereof are proposed to solve or alleviate the above problems. Summary of the Invention

[0006] The object of the present invention is to provide a data analysis method, system and medium thereof to solve the problems raised in the above background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a data analysis method, system and medium thereof, comprising the following specific steps: S1: Collect the data to be analyzed and classify and summarize the data; S2: Draw a time bar chart based on the classified and summarized data, and use the time bar chart to display data information, wherein the time bar chart uses time as the X-axis and value as the Y-axis; S3: Manages and controls time histograms, including merging, drawing lines, comparing, and modifying different types of time histograms. Before drawing lines, a visibility algorithm is applied to convert the time series data into a complex network diagram to identify and filter out noisy data points. The visibility algorithm constructs a network based on the visibility edges of adjacent data points, with edge weights dynamically calculated based on the Y-value difference and X-time interval between data points. S4: Display and control the time bar graph and store the time bar graph data; S5: Analyze the data according to the time histogram and generate a data analysis report, wherein the analysis includes calculating the data fluctuation anomaly index based on the node degree centrality of the complex network graph; S6: Perform data association based on the market basket, perform association analysis on the data, and generate a data analysis report. The association analysis uses the market basket weighted association formula: weight = (promotion day impact factor × volatility) / (1 + time decay coefficient); S7: Comprehensive analysis and generation of data analysis report.

[0008] Preferably, the data collected in S1 includes time, sales, sales cost and region. The data is classified and summarized based on the cycle time, sales, sales cost and region as constraints, and a classification storage folder is established to store the classified data, and data labels are established based on the data in the classification storage folder.

[0009] Preferably, the time bar chart drawn in S2 is based on time as the X-axis and value as the Y-axis. The data labels in S1 are input into the time bar chart, and the value of each time point is displayed using a bar chart. The height of the column represents the size of the value, and a classified time bar chart is generated. The classified time bar chart represents the sales time bar chart and sales cost time bar chart of a single period and a single region.

[0010] Preferably, the fusion of different types of time histograms in S3 is performed by fusing classified time histograms, and the sales time histogram and sales cost time histogram of a single region in a single cycle are fused to generate a single-cycle single-region profit time histogram, a multi-cycle single-region profit time histogram, a single-cycle multi-region profit time histogram, a multi-cycle multi-region profit time histogram, a multi-cycle multi-region profit time histogram, a multi-cycle single-region sales time histogram, a single-cycle multi-region sales time histogram, a multi-cycle multi-region sales time histogram, a multi-cycle single-region sales time histogram, a multi-cycle multi-region sales time histogram, a multi-cycle single-region sales cost time histogram, a single-cycle multi-region sales cost time histogram, and a multi-cycle multi-region sales cost time histogram. Drawing lines for different types of time histograms in S3 is the drawing of straight lines and curved lines on the time histogram, and the drawing of the straight lines includes the highest line, the average line and the lowest line dynamically adjusted based on the abnormality index of the complex network diagram, wherein the average line calculation excludes data points with an abnormality index greater than 1.

[0011] Preferably, the display control of the time bar chart in S4 is to perform visual design on the time bar chart, including the design of clear axes and the design of data visualization. The design of clear axes is to set clear labels for the time X-axis and the data Y-axis to indicate the units and meanings, and to set reasonable scale intervals for the time X-axis and the data Y-axis. The design of data visualization is to set the column width so that the width corresponds to the time X-axis and adjacent columns have a visual interval, and then fill the column with color, use different colors to distinguish data series and categories, add data labels to the bar chart, display specific values ​​at the top of the column, and set graphic labels on the time bar chart to call the time bar chart according to the graphic labels, and establish a database to store the time bar chart data information.

[0012] Preferably, performing association analysis on the data in S6 includes the following steps: S6.1: Data collection, collecting data corresponding to the analysis data based on the market basket, and processing and classifying the data; S6.2: Draw a time histogram of the collected market basket data and fuse it to generate a time histogram corresponding to the analyzed data; S6.3: Draw lines on the market basket time histogram and compare it with the time histogram of the analyzed data to determine the differences between the highest line, lowest line, average line, and curved line between the market basket time histogram and the analyzed data time histogram. Remove the promotion day data from the time histogram and calculate the volatility using the profit volatility calculation formula. Add promotion day data to the time histogram at intervals. Add promotion day data to the histogram at different time intervals and calculate the volatility values ​​for each. Analyze the impact of promotion day on profit volatility and reversely calibrate the promotion day impact factor using the volatility value. Adding promotion day data includes at least three iterative calibrations: the first step is to add basic promotion day data, the second step is to add data from the 7 days before the promotion day, and the third step is to add data from the 7 days after the promotion day. After each iteration, the volatility is recalculated and the differences are compared to calibrate the impact factor. Furthermore, when adding promotion day data, daily data must be recollected from the market basket to ensure the accuracy of profit volatility calculation. S6.4: Generate a data analysis report based on the results of the market basket time histogram and the analyzed data time histogram.

[0013] Preferably, the comprehensive analysis in S7 is a combination of the data analysis report in S5 and the data analysis report in S6 to conduct a comprehensive analysis of the sales volume data. The generated report includes title setting, background information, research questions, analysis results and data visualization. The background information provides the background and purpose of the analysis. The research questions clarify the specific problems and goals of the analysis. The analysis results are presented in the form of charts and text to highlight important findings. Data visualization uses charts to help readers understand the results. The charts include complex network diagrams, bar charts, pie charts and line charts. The complex network diagram shows the node connection relationship in S3 and is used to explain the root cause of data fluctuations.

[0014] The present invention also provides a data analysis system, comprising: A data collection module, which is used to collect the analyzed data and classify and summarize them; A chart module, which is used to store charts used for data analysis and supports filling chart data to form data charts; A chart control module, which is used to merge, draw lines, compare and modify charts; An analysis module, configured to analyze data based on the chart control module and generate an analysis report; A display module, which is used to display charts and analysis results generated by data analysis; a correlation module for collecting market basket data and performing comparative analysis on the analyzed data; The one-click report generation module is used to generate a report with one click for the analyzed problem.

[0015] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, a data analysis method is implemented.

[0016] Technical effects and advantages of the present invention: The present invention utilizes a data analysis method and adopts a time bar graph display mode, with time as the X-axis and value as the Y-axis, so that the trend of sales data changes can be clearly seen at a glance, which helps decision makers quickly understand the information behind the data. In addition, by drawing lines on the time bar graph, the trend of sales data changes can be clearly seen at a glance, which helps decision makers quickly understand the information behind the data. The present invention utilizes data analysis methods and, through an association method based on market basket analysis, can deeply explore the relationships between sales data, identify potential sales opportunities and customer behavior patterns, and thus provide data support for market strategies. By combining multi-dimensional data analysis results, a comprehensive analysis report is generated. The report not only contains the analysis results, but also provides background information and research questions, which helps decision makers fully understand the purpose and significance of the analysis. Through real-time data analysis and visual display, enterprises can quickly respond to market changes, optimize sales strategies, and improve the accuracy and efficiency of decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flowchart of the data analysis method of the present invention. DETAILED DESCRIPTION

[0018] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0019] The present invention provides Figure 1 A data analysis method shown includes the following specific steps: S1: Collect the data to be analyzed, classify and summarize the data, check the collected data to ensure the accuracy of the data, and classify and summarize the data; S2: Draw a time bar chart based on the classified and summarized data. Use the time bar chart to display the data information. The time bar chart uses time as the X-axis and value as the Y-axis. The time bar chart is used to facilitate intuitive understanding of the data information. S3: Manages and controls time histograms, allowing for fusion, line drawing, comparison, and modification of different types of time histograms. Before line drawing, a visibility algorithm is applied to convert time series data into a complex network diagram to identify and filter out noisy data points. The visibility algorithm constructs a network based on the visibility edges of adjacent data points. Edge weights are dynamically calculated based on the Y-value difference and X-time interval between data points. This dynamic weighting algorithm effectively identifies and filters noisy data points. Weights are set based on the Y-value difference and X-time interval between adjacent data points, making analysis more accurate and better reflecting the true trends and fluctuations of the data. S4: Display and control the time bar graph and store the time bar graph data; S5: Analyze the data based on the time histogram and generate a data analysis report. The analysis includes calculating the data fluctuation anomaly index based on the node degree centrality of the complex network graph; S6: Data association is performed based on the market basket, and correlation analysis is performed on the data to generate a data analysis report. The correlation analysis uses the market basket weighted association formula: weight = (promotion day impact factor × volatility) / (1 + time decay coefficient). Edge weights are dynamically calculated using a visibility algorithm. This dynamic weight calculation can effectively identify and filter out noisy data points. The weight is set based on the Y value difference and X time interval of adjacent data points, making the analysis more accurate and better reflecting the true trends and fluctuations of the data. S7: Comprehensive analysis, generate data analysis report, and optimize sales plan based on the analysis report.

[0020] Specifically, the data collected in S1 includes time, sales, sales cost and region. The data is classified and summarized based on the constraints of cycle time, sales, sales cost and region. The cycle time can be one week, one month or one quarter. A classification storage folder is established to store the classified data, and data labels are established based on the data in the classification storage folder. Therefore, the required data can be quickly retrieved based on the classification storage folder information.

[0021] Furthermore, the time bar chart drawn in S2 uses time as the X-axis and value as the Y-axis. The data labels in S1 are input into the time bar chart, and the value of each time point is displayed using a bar chart. The height of the column represents the size of the value, and a classified time bar chart is generated. The classified time bar chart represents the sales time bar chart and sales cost time bar chart of a single period and a single region, so that the time bar chart clearly shows the data information for easy viewing by personnel.

[0022] Furthermore, the fusion of different types of time histograms in S3 is carried out by fusion of classified time histograms. The sales time histogram and sales cost time histogram of a single region in a single period are fused to generate a single-period single-region profit time histogram, a multi-period single-region profit time histogram, a single-period multi-region profit time histogram, a multi-period multi-region profit time histogram, a multi-period multi-region profit time histogram, a multi-period single-region sales time histogram, a single-period multi-region sales time histogram, a multi-period multi-region sales time histogram, a multi-period single-region sales time histogram, a single-period multi-region sales time histogram, a multi-period multi-region sales time histogram, a multi-period single-region sales cost time histogram, a single-period multi-region sales cost time histogram, a multi-period multi-region sales cost time histogram. The drawing of different types of time histograms in S3 is to draw straight lines and curved lines on the time histogram. The drawing of straight lines includes the highest line, the average line and the lowest line that are dynamically adjusted based on the abnormality index of the complex network diagram. The average line calculation excludes data points with an abnormality index greater than 1. The single-period single-region profit time histogram, the multi-period single-region profit time histogram, the single-period multi-region profit time histogram The bar chart and the multi-period multi-region profit time bar chart represent the profit charts of a region in a single period, a multi-period single region, a single-period multi-region, and a multi-period multi-region respectively. When the profit in a single time unit is a negative value, the corresponding column on the time bar chart faces the bottom of the X-axis, so that people can intuitively check whether it is profitable. The multi-region profit time bar chart has two display methods. One is to set the profit columns of multiple regions in one unit on the X-axis side by side. The side-by-side time bar chart facilitates intuitive comparison of the profit differences of different regions. The other is to add the profit of multiple regions in one unit on the X-axis so that one column shows the sum of the profits of multiple regions. The multi-period single-region sales time bar chart is formed by combining the sales of a single region in chronological order with multiple single-period single-region sales time bar charts. The single-period multi-region sales cost time bar chart is formed by combining the sales costs of a single region in chronological order with multiple single-period single-region sales cost time bar charts. The multi-region sales time bar chart and the sales cost time bar chart are combined with the multi-region profit time bar chart. Figure 1 Each sample has two display modes.

[0023] Furthermore, in S3, drawing lines for different types of time bar charts is to draw straight lines and curved lines on the time bar chart. The drawing of straight lines includes the drawing of the highest line, average line, lowest line and preset line. The highest line represents the highest column data on the Y axis of the time bar chart, the average line represents the average column data in the time bar chart, the lowest line represents the lowest column data on the Y axis of the time bar chart, and the preset line represents a straight line in the time bar chart with Y equal to the preset value. The preset value corresponds to the sales target estimated or set by the enterprise. By changing the value of the preset straight line Y, the position of the preset line on the time bar chart can be changed, and the value of the highest line Y, the value of the average line Y, and the value of the lowest line Y are entered in the time bar chart formula. The value and the value of the preset line Y are set, and the value range of X is limited, that is, the value range of X is within the period of the time bar chart, then the corresponding high line, average line, lowest line and preset line can be generated. The drawing of the bending line is to connect the adjacent columns in the time bar chart with straight lines, so that the vertices of the columns in the time bar chart are connected to form a bending line. The trend of sales data in the time bar chart can be intuitively seen through the bending line. Comparison is used to compare the data of the time bar chart, including comparison of time bar charts in the same period, line comparison and correlation comparison. Modification is used to control the display of the time bar chart, including zooming in, zooming out and stretching operations, so as to facilitate the operation and viewing of the time bar chart, and modification is also used to modify the preset value.

[0024] Furthermore, the display control of the time bar chart in S4 is to carry out visual design of the time bar chart, including the design of clear axis and data visualization. The design of clear axis is to set clear labels for the time X-axis and data Y-axis to indicate the units and meanings, and set reasonable scale intervals for the time X-axis and data Y-axis. The design of data visualization is to set the column width so that the width corresponds to the time X-axis and adjacent columns have visual intervals to avoid the situation where the bar charts in adjacent units are crowded and indistinguishable. Then, the columns are filled with colors and different colors are used to distinguish data series and categories, so that the color information can be used to facilitate the observation of data information of the same series or category. Data labels are added to the bar chart and specific values ​​are displayed at the top of the column to improve the integrity and clarity of the time bar chart. Graphic labels are set on the time bar chart to retrieve the time bar chart according to the graphic labels, and a database is established to store the time bar chart data information to avoid data loss and improve the security of data use.

[0025] In particular, the data analysis based on the time bar chart in S5 is performed by comparing the lines drawn on the time bar chart, that is, by comparing the height relationship between the preset line and the highest line, the average line and the lowest line. When the preset line is higher than the highest line, the data in the time bar chart does not reach the preset value. When the preset line is lower than the lowest line, the data in the time bar chart reaches the preset value. When the preset line is higher than the average line and lower than the highest line, the average data in the time bar chart does not reach the preset value. When the preset line is lower than the average line and higher than the lowest line, the average data in the time bar chart reaches the preset value. The volatility of the profit is calculated by the following formula:

[0026] Where, is the earnings volatility, It is the Y-axis value of the average line in the profit time histogram. It is the Y-axis value of the preset line in the time histogram. If it is a positive value, the profit in one cycle reaches the expected preset value. The larger the value, the higher the profit will be than the expected preset value, and vice versa. When it is a negative value, the profit in one cycle does not reach the expected preset value, and when there is no profit in one cycle, is a negative value, which minus is still a negative value, and by calculating The value of can be used to get the difference between the specific profit amount and the expected preset value, and the volatility calculation formula can also be used to calculate the volatility of sales, so that The value of can be used to get the difference between the actual sales value and the expected sales value, and the sales volatility can be calculated based on The inventory is adjusted dynamically based on the value of .

[0027] Furthermore, the correlation analysis of data in S6 includes the following steps: S6.1: Data collection, collecting data corresponding to the analysis data based on the market basket, and processing and classifying the data; S6.2: Draw a time histogram of the collected market basket data and fuse it to generate a time histogram corresponding to the analyzed data; S6.3: Draw lines on the market basket time histogram and compare it with the time histogram of the analyzed data to determine the differences between the highest line, lowest line, average line, and bend line between the market basket time histogram and the analyzed data time histogram. Remove the promotion day data information from the time histogram and calculate the volatility value using the profit volatility calculation formula. Add the promotion day data information to the time histogram at intervals. Add the promotion day data information to the histogram at different time intervals and calculate the volatility value respectively. Analyze the impact of the promotion day on the profit volatility and reversely calibrate the promotion day impact factor through the volatility value. Adding the promotion day data information includes at least three iterative calibrations: the first adding basic promotion day data, the second adding data from the 7 days before the promotion day, and the third adding data from the 7 days after the promotion day. After each iteration, recalculate the volatility and compare the differences to calibrate the impact factor. When adding the promotion day data information, recollect daily data from the market basket to ensure the accuracy of the profit volatility calculation. S6.4: Generate a data analysis report based on the results of the market basket time histogram and the analyzed data time histogram.

[0028] Specifically, the comprehensive analysis in S7 combines the data analysis reports in S5 and S6 to conduct a comprehensive analysis of sales data. The generated report includes the title setting, background information, research questions, analysis results and data visualization. The background information provides the background and purpose of the analysis. The research questions clarify the specific problems and goals of the analysis. The analysis results are presented in the form of charts and text to highlight important findings. Data visualization uses charts to help readers understand the results. Charts include but are not limited to bar charts, pie charts and line charts. Figure 1 The complex network diagram shows the node connection relationship in S3 and is used to explain the root cause of data fluctuations. The pie chart can be used to observe the percentage data for data analysis. The line chart can be used to observe the data change trend. The generated report can be used to optimize the sales strategy to improve the accuracy and efficiency of decision-making.

[0029] The present invention also provides a data analysis system, which includes a data acquisition module, a chart module, a chart control module, an analysis module, a display module, an association module and a one-click report generation module. The data acquisition module is used to collect the analyzed data and classify and summarize it. The chart module is used to store the charts used for data analysis and supports the filling of chart data to form data charts. When the chart to be used is input in the chart module, the chart to be used is displayed first, and the associated charts are displayed. The chart control module is used to merge, draw lines, compare and modify the charts. The analysis module is used to analyze the data based on the chart control module and generate an analysis report. The display module is used to display the charts and analysis results generated by the data analysis. The association module is used to collect market basket data and compare and analyze the analyzed data. The one-click report generation module is used to generate a report with one click for the analyzed problem. The data analysis system, the data acquisition module, the chart module, the chart control module, the analysis module, the display module, the association module and the one-click report generation module are connected.

[0030] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, a data analysis method is implemented.

[0031] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A data analysis method, characterized in that: The specific steps include: S1: Collect the data to be analyzed and classify and summarize the data; S2: Draw a time bar chart based on the classified and summarized data, and use the time bar chart to display data information, wherein the time bar chart uses time as the X-axis and value as the Y-axis; S3: Manages and controls time histograms, including merging, drawing lines, comparing, and modifying different types of time histograms. Before drawing lines, a visibility algorithm is applied to convert the time series data into a complex network diagram to identify and filter out noisy data points. The visibility algorithm constructs a network based on the visibility edges of adjacent data points, with edge weights dynamically calculated based on the Y-value difference and X-time interval between data points. S4: Display and control the time bar graph and store the time bar graph data; S5: Analyze the data according to the time histogram and generate a data analysis report, wherein the analysis includes calculating the data fluctuation anomaly index based on the node degree centrality of the complex network graph; S6: Perform data association based on the market basket, perform association analysis on the data, and generate a data analysis report. The association analysis uses the market basket weighted association formula: weight = (promotion day impact factor × volatility) / (1 + time decay coefficient); S7: Comprehensive analysis and generation of data analysis report.

2. A data analysis method according to claim 1, characterized in that: The data collected in S1 includes time, sales, sales cost and region. The data is classified and summarized based on the cycle time, sales, sales cost and region as constraints. A classification storage folder is established to store the classified data, and data labels are established based on the data in the classification storage folder.

3. A data analysis method according to claim 2, characterized in that: The time bar chart drawn in S2 uses time as the X-axis and value as the Y-axis. The data labels in S1 are input into the time bar chart, and the value of each time point is displayed using a bar chart. The height of the column represents the size of the value, and a classified time bar chart is generated. The classified time bar chart represents the sales time bar chart and sales cost time bar chart of a single period and a single region.

4. A data analysis method according to claim 3, characterized in that: The fusion of different types of time histograms in S3 is performed by fusing classified time histograms, and the sales time histogram and sales cost time histogram of a single region in a single cycle are fused to generate a single-cycle single-region profit time histogram, a multi-cycle single-region profit time histogram, a single-cycle multi-region profit time histogram, a multi-cycle multi-region profit time histogram, a multi-cycle multi-region profit time histogram, a multi-cycle single-region sales time histogram, a single-cycle multi-region sales time histogram, a multi-cycle multi-region sales time histogram, a multi-cycle multi-region sales time histogram, a multi-cycle single-region sales cost time histogram, a single-cycle multi-region sales cost time histogram, a multi-cycle multi-region sales cost time histogram, and a multi-cycle multi-region sales cost time histogram. The drawing of different types of time histograms in S3 is to draw straight lines and curved lines on the time histogram, and the drawing of the straight lines includes the highest line, the average line and the lowest line dynamically adjusted based on the abnormality index of the complex network diagram, wherein the average line calculation excludes data points with an abnormality index greater than 1.

5. A data analysis method according to claim 4, characterized in that: The display control of the time bar chart in S4 is to perform visual design on the time bar chart, including the design of clear axis and data visualization. The design of clear axis is to set clear labels for the time X-axis and the data Y-axis to indicate the units and meanings, and to set reasonable scale intervals for the time X-axis and the data Y-axis. The design of data visualization is to set the column width so that the width corresponds to the time X-axis and adjacent columns have a visual interval, and then fill the column with color, use different colors to distinguish data series and categories, add data labels to the bar chart, display specific values ​​at the top of the column, and set graphic labels on the time bar chart to call the time bar chart according to the graphic labels, and establish a database to store the time bar chart data information.

6. A data analysis method according to claim 5, characterized in that: The association analysis of the data in S6 includes the following steps: S6.1: Data collection, collecting data corresponding to the analysis data based on the market basket, and processing and classifying the data; S6.2: Draw a time histogram of the collected market basket data and fuse it to generate a time histogram corresponding to the analyzed data; S6.3: Draw lines on the market basket time histogram and compare it with the time histogram of the analyzed data to determine the differences between the highest line, lowest line, average line, and curved line between the market basket time histogram and the analyzed data time histogram. Remove the promotion day data from the time histogram and calculate the volatility using the profit volatility calculation formula. Add promotion day data to the time histogram at intervals. Add promotion day data to the histogram at different time intervals and calculate the volatility values ​​for each. Analyze the impact of promotion day on profit volatility and reversely calibrate the promotion day impact factor using the volatility value. Adding promotion day data includes at least three iterative calibrations: the first step is to add basic promotion day data, the second step is to add data from the 7 days before the promotion day, and the third step is to add data from the 7 days after the promotion day. After each iteration, the volatility is recalculated and the differences are compared to calibrate the impact factor. Furthermore, when adding promotion day data, daily data must be recollected from the market basket to ensure the accuracy of profit volatility calculation. S6.4: Generate a data analysis report based on the results of the market basket time histogram and the analyzed data time histogram.

7. A data analysis method according to claim 6, characterized in that: The comprehensive analysis in S7 is a combination of the data analysis report in S5 and the data analysis report in S6 to conduct a comprehensive analysis of the sales data. The generated report includes title setting, background information, research questions, analysis results and data visualization. The background information provides the background and purpose of the analysis. The research questions clarify the specific problems and goals of the analysis. The analysis results are presented in the form of charts and text to highlight important findings. Data visualization uses charts to help readers understand the results. The charts include complex network diagrams, bar charts, pie charts and line charts. The complex network diagram shows the node connection relationship in S3 and is used to explain the root cause of data fluctuations.

8. A data analysis system, implementing a data analysis method according to any one of 1 to 7, characterized in that: The data analysis system includes: A data collection module, which is used to collect the analyzed data and classify and summarize them; A chart module, which is used to store charts used for data analysis and supports filling chart data to form data charts; A chart control module, which is used to merge, draw lines, compare and modify charts; An analysis module, configured to analyze data based on the chart control module and generate an analysis report; A display module, which is used to display charts and analysis results generated by data analysis; a correlation module for collecting market basket data and performing comparative analysis on the analyzed data; The one-click report generation module is used to generate a report with one click for the analyzed problem.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, implements a data analysis method according to any one of claims 1 to 7.

Images