Tea leaf sales data analysis method and system based on cloud computing

Abstract

The invention discloses a tea sales data analysis method and system based on cloud computing, and relates to the technical field of data analysis, and the method comprises the steps: obtaining tea sales data in real time; obtaining an abnormal value with statistical significance in the tea sales data; screening out a plurality of tea leaf selling strategies at the moment corresponding to the abnormal value; determining an influence index of each to-be-verified tea sales strategy on the tea sales data; judging whether the influence index of the to-be-verified tea sales strategy on the tea sales data is greater than a preset value or not; designating an expected index of each piece of tea leaf sales data; obtaining the optimal tea leaf sales strategy execution degree of the current stage; and the cloud data analysis center issues the optimal tea sales strategy execution degree to each terminal. The method has the advantages that the optimal tea leaf sales strategy can be formulated based on the tea leaf sales data, it is guaranteed that the tea leaf sales strategy meets the formulated sales demand, and efficient and accurate information prediction and decision assistance are provided for tea leaf sales.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to a cloud computing-based tea sales data analysis method and system. Background Art

[0002] In the prior art, there is a lack of intelligent analysis of the correlation between tea sales strategies and tea sales data. Therefore, the formulation of tea sales strategies often relies on the personal experience of decision makers. This approach results in the formulation of tea sales strategies not being able to guarantee the acquisition of optimal tea sales data, making it difficult to determine the optimal tea sales strategy execution plan. Summary of the invention

[0003] In order to solve the above technical problems, a cloud computing-based tea sales data analysis method and system are provided. The technical solution solves the problem that the above-mentioned prior art lacks intelligent analysis of the correlation between tea sales strategies and tea sales data. Therefore, the formulation of tea sales strategies often relies on the personal experience of decision makers. This approach results in that the formulation of tea sales strategies cannot guarantee the acquisition of optimized tea sales data, making it difficult to determine the optimal tea sales strategy execution plan.

[0004] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0005] A tea sales data analysis method based on cloud computing, comprising:

[0006] Obtain tea sales data in real time and upload it to the cloud data analysis center;

[0007] The cloud data analysis center screens the tea sales data to obtain statistically significant outliers in the tea sales data;

[0008] Filter out several tea sales strategies at the time corresponding to the outliers and record them as tea sales strategies to be verified;

[0009] Conduct a comprehensive analysis of each tea sales strategy to be verified, and determine the impact indicators of each tea sales strategy to be verified on tea sales data;

[0010] Determine whether the impact index of the tea sales strategy to be verified on the tea sales data is greater than a preset value. If so, the tea sales strategy to be verified is recorded as a high-impact weight sales strategy. If not, the tea sales strategy to be verified is recorded as a low-impact weight sales strategy.

[0011] Combined with the current sales demand for tea, specify the expected indicators for each tea sales data;

[0012] Based on the impact indicators of high-impact weighted sales strategies on tea sales data and the expected indicators of each tea sales data, a comprehensive analysis is conducted to obtain the optimal tea sales strategy execution level at the current stage;

[0013] The cloud data analysis center delegates the optimal tea sales strategy execution level to each terminal, and each terminal executes the tea sales strategy according to the set execution level.

[0014] Preferably, the step of screening the tea sales data to obtain statistically significant outliers in the tea sales data specifically includes:

[0015] Set the statistical period;

[0016] Sum the tea sales data within the statistical period to obtain the total tea sales data within the current period;

[0017] Based on the Grubbs criterion, peak screening conditions were established;

[0018] Based on the peak screening condition, the abnormal data points that exceed the normal fluctuation in the total tea sales data within several periods are screened out as statistically significant outliers in the tea sales data;

[0019] Wherein, the expression of the peak screening condition is:

[0020]

[0021] In the formula, x i is the total tea sales data in the i-th period, is the average value of the total tea sales data in all periods, s is the standard deviation of the total tea sales data in all periods, bpn is the Grubbs critical value, if the expression of the peak screening condition is met, then x i is an outlier;

[0022] The Grubbs critical value is obtained as follows:

[0023] A detection level value is set based on the abnormal sensitivity of the total tea sales data;

[0024] Determine the total number of all tea sales data;

[0025] Based on the total number and the detection level value of the total tea sales data, the corresponding value is retrieved from the Grubbs table as the Grubbs critical value.

[0026] Preferably, the comprehensive analysis of each tea sales strategy to be verified to determine the impact index of each tea sales strategy to be verified on tea sales data specifically includes:

[0027] From the historical sales data of tea, select a number of tea sales data when the current tea sales strategy to be verified is executed to different degrees, and record them as initial training data;

[0028] Analyze the initial training data in combination with the historical sales data of tea, and obtain the change rate of tea sales data when the current tea sales strategy to be verified is executed at different degrees as standardized data;

[0029] Based on the standardized data, the impact index between the tea sales strategy to be verified and the tea sales data is calculated through the correlation calculation formula.

[0030] Preferably, the correlation calculation formula is specifically:

[0031]

[0032] Where R is the impact index between the tea sales strategy to be verified and the tea sales data, n is the total number of standardized data, V j is the execution degree of the tea sales strategy to be verified in the jth standardized data, p j is the rate of change of tea sales data in the jth standardized data.

[0033] Preferably, analyzing the initial training data in combination with the historical sales data of tea to obtain the change rate of tea sales data when the current tea sales strategy to be verified is executed at different degrees specifically includes:

[0034] Obtain all other tea sales strategies executed at the time corresponding to the initial training data of the tea sales strategy to be verified at each execution level, and record them as comparison tea sales strategies;

[0035] From the historical sales data of tea, select a number of comparison moments at which only the comparison tea sales strategy is implemented, and determine the tea sales data at the comparison moments;

[0036] The tea sales data at several comparison moments are averaged to obtain comparison standard data;

[0037] Based on the initial training data and the comparison standard data, the change rate of tea sales data is calculated according to the standardized calculation formula; wherein the standardized calculation formula is specifically:

[0038]

[0039] Where p is the rate of change of tea sales data, x is the initial training data, and x0 is the comparison standard data.

[0040] Preferably, the expected indicators of each tea sales data are specified in combination with the current tea sales demand, specifically including:

[0041] Based on the current sales demand of tea, establish the tea sales data improvement demand matrix A, A=[a1 ... a k ...a H ], where H is the total number of tea sales data types, a k is the demand weight value of the k-th tea sales data.

[0042] Preferably, the comprehensive analysis based on the impact index of the high-impact weighted sales strategy on the tea sales data and the expected index of each tea sales data to obtain the optimal tea sales strategy execution degree at the current stage specifically includes:

[0043] Establish a linear regression prediction model between the execution degree of each high-impact weighted sales strategy and the sales data improvement rate of each type of tea;

[0044] According to the set value interval, establish several high-impact weight sales strategy execution degree matrices Bv, Bv = [b v1 ... b v1 ... b vM ], where M is the total number of high-impact weighted sales strategies, Bv is the execution degree matrix of the vth high-impact weighted sales strategy, and b vl is the execution degree of the lth high-influence weighted sales strategy in the execution degree matrix of the vth high-influence weighted sales strategy;

[0045] Calculate the tea sales data prediction matrix Cv corresponding to the execution degree moment of each high-impact weight sales strategy, Cv = [c v1 ... c vk ... c vH ], Cv is the tea sales data prediction matrix corresponding to the execution degree matrix of the vth high-impact weighted sales strategy, c vk Improve the predicted value of the k-th tea sales data in the v-th tea sales data prediction matrix;

[0046] Calculate the sales demand fit between each tea sales data prediction matrix and the tea sales data improvement demand matrix;

[0047] The execution degree matrix of the high-influence weighted sales strategy corresponding to the tea sales data prediction matrix with the largest sales demand fitting degree is screened out, and the execution degree matrix of the high-influence weighted sales strategy is used as the optimal tea sales strategy execution degree matrix.

[0048] Preferably, the calculation formula for the predicted value of tea sales data improvement is:

[0049]

[0050] In the formula, c vk Improve the prediction value for tea sales data, It is the linear regression prediction equation between the execution degree of the lth high-impact weighted sales strategy and the sales data of the kth type of tea.

[0051] Preferably, the calculation formula for calculating the sales demand fitting degree between each tea sales data prediction matrix and the tea sales data improvement demand matrix is:

[0052]

[0053] Where N h is the sales demand fit, α k is the demand weight of the k-th tea sales data.

[0054] Furthermore, a cloud computing-based tea sales data analysis system is proposed, which is used to implement the cloud computing-based tea sales data analysis method as described above, including:

[0055] A sales terminal, wherein the sales terminal is used to implement tea sales strategies;

[0056] A data acquisition module, which is electrically connected to the sales terminal and is used to obtain tea sales data from the sales terminal in real time and upload it to a cloud data analysis center;

[0057] A cloud data analysis center is provided, wherein the cloud data analysis center establishes a communication connection with the data acquisition module by wired or wireless means, and the cloud data analysis center is used for screening the tea sales data, obtaining statistically significant outliers in the tea sales data, screening out several tea sales strategies at the time corresponding to the outliers, recording them as tea sales strategies to be verified, conducting a comprehensive analysis on each tea sales strategy to be verified, determining the impact index of each tea sales strategy to be verified on the tea sales data, screening out sales strategies with high impact weights, and specifying the expected index of each tea sales data and the impact index of the high impact weight sales strategies on the tea sales data and the expected index of each tea sales data in combination with the current tea sales demand, conducting a comprehensive analysis, and obtaining the optimal tea sales strategy execution degree at the current stage.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] The present invention proposes a tea sales data analysis method based on cloud computing, which screens and identifies abnormal sales data in tea data, screens the tea sales strategies corresponding to the abnormal sales data, and screens several high-impact weighted sales strategies that have a high impact on the tea sales data, and intelligently analyzes the execution degree of each high-impact weighted sales strategy in combination with the current tea sales demand. In this way, the most optimized tea sales strategy can be effectively formulated based on the tea sales data, ensuring that the tea sales strategy meets the formulated sales demand, and providing efficient and accurate information forecasting and decision-making assistance for tea sales. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 This is a flow chart of the tea sales data analysis method based on cloud computing proposed by the present invention;

[0061] Figure 2 A flow chart of a method for obtaining statistically significant outliers in tea sales data in the present invention;

[0062] Figure 3 A flow chart of a method for determining the impact index of each tea sales strategy to be verified on tea sales data in the present invention;

[0063] Figure 4 A flow chart of a method for obtaining a change rate of tea sales data when the current tea sales strategy to be verified is executed at different degrees in the present invention;

[0064] Figure 5 The present invention is a flow chart of the method for obtaining the optimal tea sales strategy execution degree at the current stage. DETAILED DESCRIPTION

[0065] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0066] Reference Figure 1 As shown, a tea sales data analysis method based on cloud computing includes:

[0067] Acquire tea sales data in real time and upload it to the cloud data analysis center. Tea sales data at least includes tea sales, tea sales volume, tea promotion, and tea sales profit;

[0068] The cloud data analysis center screens the tea sales data to obtain statistically significant outliers in the tea sales data;

[0069] Filter out several tea sales strategies at the time corresponding to the outliers and record them as tea sales strategies to be verified;

[0070] Conduct a comprehensive analysis of each tea sales strategy to be verified, and determine the impact indicators of each tea sales strategy to be verified on tea sales data;

[0071] Determine whether the impact index of the tea sales strategy to be verified on the tea sales data is greater than a preset value. If so, the tea sales strategy to be verified is recorded as a high-impact weight sales strategy. If not, the tea sales strategy to be verified is recorded as a low-impact weight sales strategy.

[0072] Combined with the current sales demand for tea, specify the expected indicators for each tea sales data;

[0073] Based on the impact indicators of high-impact weighted sales strategies on tea sales data and the expected indicators of each tea sales data, a comprehensive analysis is conducted to obtain the optimal tea sales strategy execution level at the current stage;

[0074] The cloud data analysis center delegates the optimal tea sales strategy execution level to each terminal, and each terminal executes the tea sales strategy according to the set execution level.

[0075] This solution screens and identifies abnormal sales data in tea data, screens the tea sales strategies corresponding to the abnormal sales data, and selects several high-influence weighted sales strategies that have a high impact on the tea sales data, and intelligently analyzes the execution degree of each high-influence weighted sales strategy in combination with the current tea sales demand.

[0076] Reference Figure 2 As shown, based on the data screening of tea sales data, the statistically significant outliers in the tea sales data are obtained, including:

[0077] Set the statistical period, which can be 1 day, 1 week, 1 month, 1 quarter or 1 year. The specific period duration is determined according to data statistical requirements;

[0078] Sum the tea sales data within the statistical period to obtain the total tea sales data within the current period;

[0079] Based on the Grubbs criterion, peak screening conditions were established;

[0080] Based on the peak screening condition, the abnormal data points that exceed the normal fluctuation in the total tea sales data within several periods are screened out as statistically significant outliers in the tea sales data;

[0081] The expression of the peak screening condition is:

[0082]

[0083] In the formula, x iis the total tea sales data in the i-th period, is the average value of the total tea sales data in all periods, s is the standard deviation of the total tea sales data in all periods, bpm is the Grubbs critical value, if the expression of the peak screening condition is met, then x i is an outlier;

[0084] The Grubbs critical value is obtained as follows:

[0085] A detection level value is set based on the abnormal sensitivity of the total tea sales data;

[0086] Determine the total number of all tea sales data;

[0087] Based on the total number and the detection level value of the total tea sales data, the corresponding value is retrieved from the Grubbs table as the Grubbs critical value.

[0088] It is understandable that, under the status of tea sales, the fluctuation of the total tea sales data in each cycle should be within the range of the normal distribution. Therefore, when the total tea sales data shows an abnormal peak that does not conform to the normal distribution, it means that the total tea sales data has increased significantly. The sales strategy at this time has a greater reference significance for the improvement of the total tea sales data. In this way, finding out the sales strategy that causes the positive abnormal fluctuation of the total tea sales data can effectively ensure the accuracy of the subsequent determination of the tea sales strategy.

[0089] Reference Figure 3 As shown, a comprehensive analysis is conducted on each tea sales strategy to be verified, and the impact indicators of each tea sales strategy to be verified on tea sales data are determined, including:

[0090] From the historical sales data of tea, select a number of tea sales data when the current tea sales strategy to be verified is executed to different degrees, and record them as initial training data;

[0091] Analyze the initial training data in combination with the historical sales data of tea, and obtain the change rate of tea sales data when the current tea sales strategy to be verified is executed at different degrees as standardized data;

[0092] Based on the standardized data, the impact index between the tea sales strategy to be verified and the tea sales data is calculated through the correlation calculation formula.

[0093] The correlation calculation formula is as follows:

[0094]

[0095] Where R is the impact index between the tea sales strategy to be verified and the tea sales data, n is the total number of standardized data, Vj is the execution degree of the tea sales strategy to be verified in the jth standardized data, p j is the rate of change of tea sales data in the jth standardized data.

[0096] It is understandable that different sales strategies have different impacts on different tea sales data. For example, the discount strategy is positively correlated with the tea sales volume and the degree of tea promotion, the advertising strategy is positively correlated with the tea sales volume and the degree of tea promotion, and negatively correlated with the sales profit. Therefore, in this solution, by calculating the correlation between each sales strategy and the tea sales data as an impact indicator, the sales strategy with a large impact on the tea sales data can be quickly and accurately identified as a high-impact weight sales strategy.

[0097] Reference Figure 4 As shown, the initial training data is analyzed in combination with the historical sales data of tea, and the change rate of tea sales data when the current tea sales strategy to be verified is executed at different degrees is obtained, which specifically includes:

[0098] Obtain all other tea sales strategies executed at the time corresponding to the initial training data of the tea sales strategy to be verified at each execution level, and record them as comparison tea sales strategies;

[0099] From the historical sales data of tea, select a number of comparison moments at which only the comparison tea sales strategy is implemented, and determine the tea sales data at the comparison moments;

[0100] The tea sales data at several comparison moments are averaged to obtain comparison standard data;

[0101] Based on the initial training data and the comparison standard data, the change rate of tea sales data is calculated according to the standardized calculation formula;

[0102] The standardized calculation formula is as follows:

[0103]

[0104] Where p is the rate of change of tea sales data, x is the initial training data, and x0 is the comparison standard data.

[0105] It is understandable that since multiple strategies are usually executed simultaneously during the tea sales process, it is necessary to eliminate the impact of other strategies when analyzing a single strategy. Therefore, this scheme uses the change rate of tea sales data for the tea sales strategy to be verified as an indicator for correlation calculation. This method can effectively reduce the impact of other strategies on the tea sales strategy to be verified, and ensure the accuracy of the calculation of influencing indicators.

[0106] Combined with the current sales demand for tea, the expected indicators for each tea sales data are specified, including:

[0107] Based on the current sales demand of tea, establish the tea sales data improvement demand matrix A, A=[a1 … a k …a H ], where H is the total number of tea sales data types, a k is the demand weight value of the k-th tea sales data.

[0108] Reference Figure 5 As shown in the figure, based on the impact indicators of high-impact weighted sales strategies on tea sales data and the expected indicators of each tea sales data, a comprehensive analysis is conducted to obtain the optimal tea sales strategy execution degree at the current stage, which specifically includes:

[0109] Establish a linear regression prediction model between the execution degree of each high-impact weighted sales strategy and the sales data improvement rate of each type of tea;

[0110] According to the set value interval, establish several high-impact weight sales strategy execution degree matrices Bv, Bv = [b v1 … b vl … b vM ], where M is the total number of high-impact weighted sales strategies, Bv is the execution degree matrix of the vth high-impact weighted sales strategy, and b vl is the execution degree of the lth high-influence weighted sales strategy in the execution degree matrix of the vth high-influence weighted sales strategy;

[0111] Calculate the tea sales data prediction matrix Cv corresponding to the execution degree moment of each high-impact weight sales strategy, Cv = [c v1 … c vk … c vH ], Cv is the tea sales data prediction matrix corresponding to the execution degree matrix of the vth high-impact weighted sales strategy, c vk Improve the predicted value of the k-th tea sales data in the v-th tea sales data prediction matrix;

[0112] Calculate the sales demand fit between each tea sales data prediction matrix and the tea sales data improvement demand matrix;

[0113] The execution degree matrix of the high-influence weighted sales strategy corresponding to the tea sales data prediction matrix with the largest sales demand fitting degree is screened out, and the execution degree matrix of the high-influence weighted sales strategy is used as the optimal tea sales strategy execution degree matrix.

[0114] The calculation formula for the predicted value of tea sales data improvement is:

[0115]

[0116] In the formula, c vk Improve the prediction value for tea sales data, It is the linear regression prediction equation between the execution degree of the lth high-impact weighted sales strategy and the sales data of the kth type of tea.

[0117] The calculation formula for calculating the sales demand fit between each tea sales data prediction matrix and the tea sales data improvement demand matrix is:

[0118]

[0119] Where N h is the sales demand fit, α k is the demand weight of the k-th tea sales data.

[0120] It is understandable that the demand for tea sales data is different at different stages. In the promotion stage, the main need is to increase the tea sales volume and tea promotion degree. In the stable stage, the main need is to increase the tea sales profit and tea sales volume. Therefore, in this plan, based on the current tea sales demand, a tea sales data improvement demand matrix is ​​established. Then, based on the establishment of a linear regression prediction model between the execution degree of each high-impact weighted sales strategy and the sales data improvement rate of each tea, the sales data improvement degree of each high-impact weighted sales strategy under different execution degree combinations is predicted and calculated, a tea sales data prediction matrix is ​​established, and the overall improvement rate of the tea sales data prediction matrix relative to the tea sales data improvement demand matrix is ​​calculated as the sales demand fit, which can effectively ensure that the formulated optimized tea sales strategy can achieve the maximum improvement that fits the current tea sales demand.

[0121] Further, based on the same inventive concept as the above-mentioned cloud computing-based tea sales data analysis method, this solution also proposes a cloud computing-based tea sales data analysis system, including:

[0122] Sales terminal, which is used to implement tea sales strategy;

[0123] A data collection module, which is electrically connected to the sales terminal and is used to obtain tea sales data from the sales terminal in real time and upload it to a cloud data analysis center;

[0124] The cloud data analysis center establishes a communication connection with the data acquisition module by wired or wireless means. The cloud data analysis center is used for data screening of tea sales data, obtaining statistically significant outliers in the tea sales data, screening out several tea sales strategies at the time corresponding to the outliers, recording them as tea sales strategies to be verified, conducting a comprehensive analysis on each tea sales strategy to be verified, determining the impact indicators of each tea sales strategy to be verified on the tea sales data, screening out sales strategies with high impact weights, and specifying the expected indicators of each tea sales data and the impact indicators of the high impact weight sales strategies on the tea sales data and the expected indicators of each tea sales data in combination with the current tea sales demand, conducting a comprehensive analysis, and obtaining the optimal tea sales strategy execution level at the current stage.

[0125] The use process of the tea sales data analysis system based on cloud computing is as follows:

[0126] The data collection module obtains tea sales data from sales terminals in real time and uploads it to the cloud data analysis center;

[0127] The cloud data analysis center screens the tea sales data to obtain statistically significant outliers in the tea sales data;

[0128] Filter out several tea sales strategies at the time corresponding to the outliers and record them as tea sales strategies to be verified;

[0129] Conduct a comprehensive analysis of each tea sales strategy to be verified, and determine the impact indicators of each tea sales strategy to be verified on tea sales data;

[0130] Determine whether the impact index of the tea sales strategy to be verified on the tea sales data is greater than a preset value. If so, the tea sales strategy to be verified is recorded as a high-impact weight sales strategy. If not, the tea sales strategy to be verified is recorded as a low-impact weight sales strategy.

[0131] Combined with the current sales demand for tea, specify the expected indicators for each tea sales data;

[0132] Based on the impact indicators of high-impact weighted sales strategies on tea sales data and the expected indicators of each tea sales data, a comprehensive analysis is conducted to obtain the optimal tea sales strategy execution level at the current stage;

[0133] The cloud data analysis center delegates the optimal tea sales strategy execution level to each terminal, and each terminal executes the tea sales strategy according to the set execution level.

[0134] In summary, the advantages of the present invention are: it can realize the formulation of the most optimized tea sales strategy based on tea sales data, ensure that the tea sales strategy meets the formulated sales needs, and provide efficient and accurate information forecasting and decision-making assistance for tea sales.

[0135] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A tea sales data analysis method based on cloud computing, characterized in that: include: Obtain tea sales data in real time and upload it to the cloud data analysis center; The cloud data analysis center screens the tea sales data to obtain statistically significant outliers in the tea sales data; Filter out several tea sales strategies at the time corresponding to the outliers and record them as tea sales strategies to be verified; Conduct a comprehensive analysis of each tea sales strategy to be verified, and determine the impact indicators of each tea sales strategy to be verified on tea sales data; Determine whether the impact index of the tea sales strategy to be verified on the tea sales data is greater than a preset value. If so, the tea sales strategy to be verified is recorded as a high-impact weight sales strategy. If not, the tea sales strategy to be verified is recorded as a low-impact weight sales strategy. Combined with the current sales demand for tea, specify the expected indicators for each tea sales data; Based on the impact indicators of high-impact weighted sales strategies on tea sales data and the expected indicators of each tea sales data, a comprehensive analysis is conducted to obtain the optimal tea sales strategy execution level at the current stage; The cloud data analysis center delegates the optimal tea sales strategy execution level to each terminal, and each terminal executes the tea sales strategy according to the set execution level.

2. The tea sales data analysis method based on cloud computing according to claim 1 is characterized in that: The method of obtaining statistically significant outliers in the tea sales data based on data screening specifically includes: Set the statistical period; Sum the tea sales data within the statistical period to obtain the total tea sales data within the current period; Based on the Grubbs criterion, peak screening conditions were established; Based on the peak screening condition, the abnormal data points that exceed the normal fluctuation in the total tea sales data within several periods are screened out as statistically significant outliers in the tea sales data; Wherein, the expression of the peak screening condition is: In the formula, x i is the total tea sales data in the i-th period, is the average value of the total tea sales data in all periods, s is the standard deviation of the total tea sales data in all periods, bpn is the Grubbs critical value, if the expression of the peak screening condition is met, then x i is an outlier; The Grubbs critical value is obtained as follows: A detection level value is set based on the abnormal sensitivity of the total tea sales data; Determine the total number of all tea sales data; Based on the total number and the detection level value of the total tea sales data, the corresponding value is retrieved from the Grubbs table as the Grubbs critical value.

3. The tea sales data analysis method based on cloud computing according to claim 2 is characterized in that: The comprehensive analysis of each tea sales strategy to be verified to determine the impact indicators of each tea sales strategy to be verified on tea sales data specifically includes: From the historical sales data of tea, select a number of tea sales data when the current tea sales strategy to be verified is executed to different degrees, and record them as initial training data; Analyze the initial training data in combination with the historical sales data of tea, and obtain the change rate of tea sales data when the current tea sales strategy to be verified is executed at different degrees as standardized data; Based on the standardized data, the impact index between the tea sales strategy to be verified and the tea sales data is calculated through the correlation calculation formula.

4. The tea sales data analysis method based on cloud computing according to claim 3 is characterized in that: The correlation calculation formula is specifically: Where R is the impact index between the tea sales strategy to be verified and the tea sales data, n is the total number of standardized data, V j is the execution degree of the tea sales strategy to be verified in the jth standardized data, p j is the rate of change of tea sales data in the jth standardized data.

5. The tea sales data analysis method based on cloud computing according to claim 4 is characterized in that: The analyzing of the initial training data in combination with the historical sales data of tea to obtain the change rate of the tea sales data when the current tea sales strategy to be verified is executed at different degrees specifically includes: Obtain all other tea sales strategies executed at the time corresponding to the initial training data of the tea sales strategy to be verified at each execution level, and record them as comparison tea sales strategies; From the historical sales data of tea, select a number of comparison moments at which only the comparison tea sales strategy is implemented, and determine the tea sales data at the comparison moments; The tea sales data at several comparison moments are averaged to obtain comparison standard data; Based on the initial training data and the comparison standard data, the change rate of tea sales data is calculated according to the standardized calculation formula; The standardized calculation formula is as follows: Where p is the rate of change of tea sales data, x is the initial training data, and x0 is the comparison standard data.

6. The tea sales data analysis method based on cloud computing according to claim 5, characterized in that: In combination with the current sales demand for tea, the expected indicators for each tea sales data are specified to include: Based on the current sales demand of tea, establish the tea sales data improvement demand matrix A, A=[a1 ... a k ...a H ], where H is the total number of tea sales data types, a k is the demand weight value of the k-th tea sales data.

7. The tea sales data analysis method based on cloud computing according to claim 6 is characterized in that: The above-mentioned comprehensive analysis of the impact indicators of the high-impact weighted sales strategy on the tea sales data and the expected indicators of each tea sales data to obtain the optimal tea sales strategy execution degree at the current stage specifically includes: Establish a linear regression prediction model between the execution degree of each high-impact weighted sales strategy and the sales data improvement rate of each type of tea; According to the set value interval, establish several execution degree matrices Bv of high-impact weighted sales strategies, Bv = [bv1...bvl...bvM], where M is the total number of high-impact weighted sales strategy types, Bv is the execution degree matrix of the vth high-impact weighted sales strategy, and bvl is the execution degree of the lth high-impact weighted sales strategy in the execution degree matrix of the vth high-impact weighted sales strategy; Calculate the tea sales data prediction matrix Cv corresponding to the execution degree moment of each high-impact weight sales strategy, Cv = [c v1 ... c vk ... c vH ], Cv is the tea sales data prediction matrix corresponding to the execution degree matrix of the vth high-impact weighted sales strategy, c vk Improve the predicted value of the k-th tea sales data in the v-th tea sales data prediction matrix; Calculate the sales demand fit between each tea sales data prediction matrix and the tea sales data improvement demand matrix; The execution degree matrix of the high-influence weighted sales strategy corresponding to the tea sales data prediction matrix with the largest sales demand fitting degree is screened out, and the execution degree matrix of the high-influence weighted sales strategy is used as the optimal tea sales strategy execution degree matrix.

8. The tea sales data analysis method based on cloud computing according to claim 7 is characterized in that: The calculation formula for the predicted value of tea sales data improvement is: In the formula, c vk Improve the prediction value for tea sales data, It is the linear regression prediction equation between the execution degree of the lth high-impact weighted sales strategy and the sales data of the kth type of tea.

9. The tea sales data analysis method based on cloud computing according to claim 8, characterized in that: The calculation formula for calculating the sales demand fitting degree between each tea sales data prediction matrix and the tea sales data improvement demand matrix is: Where N h is the sales demand fit, α k is the demand weight of the k-th tea sales data.

10. A tea sales data analysis system based on cloud computing, characterized in that: The method for analyzing tea sales data based on cloud computing according to any one of claims 1 to 9 comprises: A sales terminal, wherein the sales terminal is used to implement tea sales strategies; A data acquisition module, which is electrically connected to the sales terminal and is used to obtain tea sales data from the sales terminal in real time and upload it to a cloud data analysis center; A cloud data analysis center is provided, wherein the cloud data analysis center establishes a communication connection with the data acquisition module by wired or wireless means, and the cloud data analysis center is used for screening the tea sales data, obtaining statistically significant outliers in the tea sales data, screening out several tea sales strategies at the time corresponding to the outliers, recording them as tea sales strategies to be verified, conducting a comprehensive analysis on each tea sales strategy to be verified, determining the impact index of each tea sales strategy to be verified on the tea sales data, screening out sales strategies with high impact weights, and specifying the expected index of each tea sales data and the impact index of the high impact weight sales strategies on the tea sales data and the expected index of each tea sales data in combination with the current tea sales demand, conducting a comprehensive analysis, and obtaining the optimal tea sales strategy execution degree at the current stage.

Images