Agricultural product supply and demand monitoring and early warning method and system

Abstract

The invention discloses an agricultural product supply and demand monitoring and early warning method and system, and belongs to the technical field of market monitoring, and the method specifically comprises the steps: collecting market data in a fixed period every fixed period; recording seasonal indexes and marketing costs of the agricultural products corresponding to the fixed period; the collected data is preprocessed; respectively training the relationship between the market data and the seasonal indexes and the relationship between the market data and the marketing cost based on machine learning, respectively establishing a seasonal market model and a marketing market model, and respectively verifying the two models to obtain corresponding correlation coefficients; when the correlation coefficient of the seasonal market model is larger than that of the marketing market model, the marketing cost is reduced, and the production quantity of agricultural products is increased, and when the correlation coefficient of the marketing market model is larger than that of the seasonal market model, the marketing cost is increased, and the production quantity of the agricultural products is adjusted to be consistent with the average sales quantity of the agricultural products; according to the invention, judgment of agricultural product sales volume dominant factors is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of market monitoring, and in particular to a method and system for monitoring and early warning of supply and demand of agricultural products. Background Art

[0002] Supply and demand monitoring is crucial to the stable operation of the agricultural product market. Effective supply and demand monitoring can not only keep abreast of market trends, but also provide a basis for decision-making for management departments and market players, and avoid market fluctuations caused by information lags.

[0003] In the field of agricultural production and sales, the supply and demand of agricultural products is an important factor affecting farmers' income and market stability. Traditional monitoring of agricultural product supply and demand mainly relies on manual observation and empirical judgment. This method often has lags and inaccuracies, and cannot reflect market changes and predict future trends in a timely manner. With the development of information technology, it has become a new trend to use big data and machine learning technology to conduct real-time monitoring and early warning of agricultural product markets. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for monitoring and early warning of agricultural product supply and demand, and solve the following technical problems: Traditional agricultural product supply and demand monitoring mainly relies on manual observation and empirical judgment. This method often has lags and inaccuracies, and is unable to reflect market changes and predict future trends in a timely manner.

[0005] The purpose of the present invention can be achieved through the following technical solutions: A method for monitoring and early warning of agricultural product supply and demand, comprising the following steps: Collect market data within a fixed period at regular intervals, the market data including agricultural product sales, search volume, and new customer ratio; and record the seasonal index and marketing cost of the agricultural products in the corresponding fixed period; Pre-process the collected data, remove the data from e-commerce marketing days, and remove abnormal fake order data; Based on machine learning, we train the relationship between market data and seasonal index, and the relationship between market data and marketing cost, respectively, establish seasonal market model and marketing market model, and verify the two models to obtain the corresponding correlation coefficients. When the correlation coefficient of the seasonal market model is greater than that of the marketing market model, the marketing cost is reduced and the output of agricultural products is increased. When the correlation coefficient of the marketing market model is greater than that of the seasonal market model, the marketing cost is increased and the output of agricultural products is adjusted to be consistent with the average sales volume of agricultural products.

[0006] As a further solution of the present invention: the specific meaning of the seasonal index is: Get the seasonal index segment [m,n] of the agricultural product, mark the middle date of the seasonal index segment as the calibration time, calculate the difference d between the current time and the calibration time, then (nm) / 2-d is the seasonal index. At this time, the seasonal index is an extremely large indicator.

[0007] As a further solution of the present invention: the specific process of eliminating e-commerce marketing days is: The market data and dates corresponding to the e-commerce marketing days are deleted from the fixed period, and the e-commerce marketing days include but are not limited to June 18, November 11, the fourth Friday of November, and December 12.

[0008] This patented method eliminates e-commerce marketing day and abnormal order-placing data during the data preprocessing stage, ensuring the authenticity and reliability of the data, thereby improving the accuracy of the analysis.

[0009] As a further solution of the present invention: the specific process of removing abnormal order-brushing data is: Get all customers who purchase current agricultural products in a fixed period, set several screening conditions, including but not limited to purchase frequency, platform login frequency and product page browsing time, extract customers who meet all the screening conditions at the same time and mark them as valid customers, and count the purchase data of target customers into market data.

[0010] As a further solution of the present invention: the process of model verification is: Input the seasonal index and marketing cost into the corresponding model, obtain the corresponding market data, establish a plane rectangular coordinate system, use the agricultural product sales volume y1, search volume y2 and new customer ratio y3 output by the model as the vertical coordinates, and use the seasonal index x1 and marketing cost x2 as the horizontal coordinates to establish the corresponding one-dimensional regression fitting function: y1=a1x1+b1, y2=a2x1+b2, y3=a3x1+b3; y1=m1x2+n1, y2=m2x2+n2, y3=m3x2+n3; Calculate the parameters a1, a2, a3, b1, b2, b3 and m1, m2, m3, n1, n2, n3 according to the values ​​of the horizontal and vertical coordinates of the corresponding functions.

[0011] As a further solution of the present invention: the process of model verification also includes calculating the correlation coefficients of the corresponding parameters of the seasonal market model and the marketing market model according to the following formula: ； ； ； ； ；

[0012] Obtain the correlation coefficients r1, r2, r3, r4, r5, and r6, and calculate the mean and variance of the correlation coefficients of the seasonal market model; obtain the correlation coefficients s1, s2, s3, s4, s5, and s6, and calculate the mean and variance of the correlation coefficients of the marketing market model, and take the weighted average of the mean and variance as the final correlation coefficient for comparison.

[0013] By comparing the correlation coefficients of the two models, it is possible to decide whether to reduce or increase marketing costs. This means that marketing strategies can be adjusted according to the actual market conditions, ineffective advertising expenditures can be avoided, and the return on advertising investment can be increased.

[0014] An agricultural product supply and demand monitoring and early warning system, comprising: A data collection module is used to collect market data within a fixed period at regular intervals, wherein the market data includes agricultural product sales, search volume, and new customer ratio; and record seasonal indexes and marketing costs of agricultural products corresponding to the fixed period; The data filtering module is used to pre-process the collected data, remove the data of e-commerce marketing days, and remove abnormal fake order data; The model verification module is used to train the relationship between market data and seasonal index and the relationship between market data and marketing cost based on machine learning, establish seasonal market model and marketing market model respectively, and verify the two models respectively to obtain the corresponding correlation coefficients; The model comparison module is used to reduce marketing costs and increase the output of agricultural products when the correlation coefficient of the seasonal market model is greater than that of the marketing market model. When the correlation coefficient of the marketing market model is greater than that of the seasonal market model, the marketing cost is increased and the output of agricultural products is adjusted to be consistent with the average sales volume of agricultural products.

[0015] Beneficial effects of the present invention: The present invention can provide a scientific basis for adjusting the production volume of agricultural products by regularly collecting market data, including sales volume, search volume and new customer ratio, and combining seasonal index and marketing cost, and using time market model and marketing market model established by machine learning; when the correlation coefficient of the time market model is high, it indicates that the current period is the seasonal period of agricultural products, and the production volume should be increased accordingly; otherwise, the production volume may need to be reduced to avoid oversupply. This patent can more accurately reflect the market supply and demand situation of agricultural products; this method reduces dependence on experience and improves the efficiency of data-driven decision-making; reasonable production scheduling and marketing input contribute to the rational use of resources and environmental protection, and promote the sustainable development of agriculture. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described below in conjunction with the accompanying drawings.

[0017] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0019] See also Figure 1 As shown, the present invention is a method and system for monitoring and early warning of agricultural product supply and demand, comprising the following steps: Collect market data within a fixed period at regular intervals, the market data including agricultural product sales, search volume, and new customer ratio; and record the seasonal index and marketing cost of the agricultural products in the corresponding fixed period; Pre-process the collected data, remove the data from e-commerce marketing days, and remove abnormal fake order data; Based on machine learning, we train the relationship between market data and seasonal index, and the relationship between market data and marketing cost, respectively, establish seasonal market model and marketing market model, and verify the two models to obtain the corresponding correlation coefficients. When the correlation coefficient of the seasonal market model is greater than that of the marketing market model, the marketing cost is reduced and the output of agricultural products is increased. When the correlation coefficient of the marketing market model is greater than that of the seasonal market model, the marketing cost is increased and the output of agricultural products is adjusted to be consistent with the average sales volume of agricultural products.

[0020] In another preferred embodiment of the present invention, the specific meaning of the seasonal index is: Get the seasonal index segment [m,n] of the agricultural product, mark the middle date of the seasonal index segment as the calibration time, calculate the difference d between the current time and the calibration time, then (nm) / 2-d is the seasonal index. At this time, the seasonal index is an extremely large indicator.

[0021] In another preferred embodiment of the present invention, the specific process of removing the e-commerce marketing day is: The market data and dates corresponding to the e-commerce marketing days are deleted from the fixed period, and the e-commerce marketing days include but are not limited to June 18, November 11, the fourth Friday of November, and December 12.

[0022] In another preferred embodiment of the present invention, the specific process of removing abnormal order-brushing data is as follows: Get all customers who purchase current agricultural products in a fixed period, set several screening conditions, including but not limited to purchase frequency, platform login frequency and product page browsing time, extract customers who meet all the screening conditions at the same time and mark them as valid customers, and count the purchase data of target customers into market data.

[0023] In another preferred embodiment of the present invention, the process of model verification is: Input the seasonal index and marketing cost into the corresponding model, obtain the corresponding market data, establish a plane rectangular coordinate system, use the agricultural product sales volume y1, search volume y2 and new customer ratio y3 output by the model as the vertical coordinates, and use the seasonal index x1 and marketing cost x2 as the horizontal coordinates to establish the corresponding one-dimensional regression fitting function: y1=a1x1+b1, y2=a2x1+b2, y3=a3x1+b3; y1=m1x2+n1, y2=m2x2+n2, y3=m3x2+n3; Calculate the parameters a1, a2, a3, b1, b2, b3 and m1, m2, m3, n1, n2, n3 according to the values ​​of the horizontal and vertical coordinates of the corresponding functions.

[0024] In another preferred embodiment of the present invention, the process of model verification further includes calculating the correlation coefficients of the corresponding parameters of the seasonal market model and the marketing market model according to the following formula: ； ； ； ； ；

[0025] Obtain the correlation coefficients r1, r2, r3, r4, r5, and r6, and calculate the mean and variance of the correlation coefficients of the seasonal market model; obtain the correlation coefficients s1, s2, s3, s4, s5, and s6, and calculate the mean and variance of the correlation coefficients of the marketing market model, and take the weighted average of the mean and variance as the final correlation coefficient for comparison.

[0026] An agricultural product supply and demand monitoring and early warning system, comprising: A data collection module is used to collect market data within a fixed period at regular intervals, wherein the market data includes agricultural product sales, search volume, and new customer ratio; and record seasonal indexes and marketing costs of agricultural products corresponding to the fixed period; The data filtering module is used to pre-process the collected data, remove the data of e-commerce marketing days, and remove abnormal fake order data; The model verification module is used to train the relationship between market data and seasonal index and the relationship between market data and marketing cost based on machine learning, establish seasonal market model and marketing market model respectively, and verify the two models respectively to obtain the corresponding correlation coefficients; The model comparison module is used to reduce marketing costs and increase the output of agricultural products when the correlation coefficient of the seasonal market model is greater than that of the marketing market model. When the correlation coefficient of the marketing market model is greater than that of the seasonal market model, the marketing cost is increased and the output of agricultural products is adjusted to be consistent with the average sales volume of agricultural products.

[0027] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for monitoring and early warning of agricultural product supply and demand, characterized in that: The following steps are involved: Collect market data within a fixed period at regular intervals, the market data including agricultural product sales, search volume, and new customer ratio; and record the seasonal index and marketing cost of the agricultural products in the corresponding fixed period; Pre-process the collected data, remove the data from e-commerce marketing days, and remove abnormal fake order data; Based on machine learning, we train the relationship between market data and seasonal index, and the relationship between market data and marketing cost, respectively, establish seasonal market model and marketing market model, and verify the two models to obtain the corresponding correlation coefficients. When the correlation coefficient of the seasonal market model is greater than that of the marketing market model, the marketing cost is reduced and the output of agricultural products is increased. When the correlation coefficient of the marketing market model is greater than that of the seasonal market model, the marketing cost is increased and the output of agricultural products is adjusted to be consistent with the average sales volume of agricultural products.

2. A method for monitoring and early warning of agricultural product supply and demand according to claim 1, characterized in that: The specific meaning of the seasonal index is: Get the seasonal index segment [m,n] of the agricultural product, mark the middle date of the seasonal index segment as the calibration time, calculate the difference d between the current time and the calibration time, then (nm) / 2-d is the seasonal index. At this time, the seasonal index is an extremely large indicator.

3. The agricultural product supply and demand monitoring and early warning method according to claim 1 is characterized in that: The specific process of removing e-commerce marketing days is as follows: The market data and dates corresponding to the e-commerce marketing days are deleted from the fixed period, and the e-commerce marketing days include but are not limited to June 18, November 11, the fourth Friday of November, and December 12.

4. The agricultural product supply and demand monitoring and early warning method according to claim 1 is characterized in that: The specific process of removing abnormal order-brushing data is as follows: Get all customers who purchase current agricultural products in a fixed period, set several screening conditions, including but not limited to purchase frequency, platform login frequency and product page browsing time, extract customers who meet all the screening conditions at the same time and mark them as valid customers, and count the purchase data of target customers into market data.

5. The agricultural product supply and demand monitoring and early warning method according to claim 1 is characterized in that: The process of model verification is as follows: Input the seasonal index and marketing cost into the corresponding model, obtain the corresponding market data, establish a plane rectangular coordinate system, use the agricultural product sales volume y1, search volume y2 and new customer ratio y3 output by the model as the vertical coordinates, and use the seasonal index x1 and marketing cost x2 as the horizontal coordinates to establish the corresponding one-dimensional regression fitting function: y1=a1x1+b1, y2=a2x1+b2, y3=a3x1+b3; y1=m1x2+n1, y2=m2x2+n2, y3=m3x2+n3; Calculate the parameters a1, a2, a3, b1, b2, b3 and m1, m2, m3, n1, n2, n3 according to the values ​​of the horizontal and vertical coordinates of the corresponding functions.

6. The agricultural product supply and demand monitoring and early warning method according to claim 5 is characterized in that: The process of model verification also includes calculating the correlation coefficients of the corresponding parameters of the seasonal market model and the marketing market model according to the following formula: ； ； ； ； ； Obtain the correlation coefficients r1, r2, r3, r4, r5, and r6, and calculate the mean and variance of the correlation coefficients of the seasonal market model; obtain the correlation coefficients s1, s2, s3, s4, s5, and s6, and calculate the mean and variance of the correlation coefficients of the marketing market model, and take the weighted average of the mean and variance as the final correlation coefficient for comparison.

7. An agricultural product supply and demand monitoring and early warning system, characterized in that: include: A data collection module is used to collect market data within a fixed period at regular intervals, wherein the market data includes agricultural product sales, search volume, and new customer ratio; and record seasonal indexes and marketing costs of agricultural products corresponding to the fixed period; The data filtering module is used to pre-process the collected data, remove the data of e-commerce marketing days, and remove abnormal fake order data; The model verification module is used to train the relationship between market data and seasonal index and the relationship between market data and marketing cost based on machine learning, establish seasonal market model and marketing market model respectively, and verify the two models respectively to obtain the corresponding correlation coefficients; The model comparison module is used to reduce marketing costs and increase the output of agricultural products when the correlation coefficient of the seasonal market model is greater than that of the marketing market model. When the correlation coefficient of the marketing market model is greater than that of the seasonal market model, the marketing cost is increased and the output of agricultural products is adjusted to be consistent with the average sales volume of agricultural products.

Images