Whole-network supply and marketing early warning management and control system based on crawler and large model technology

Abstract

The invention discloses a whole-network supply and marketing early warning management and control system based on a crawler and a large model technology, relates to the technical field of economic management and control, and solves the problems that in the prior art, abnormal prices of commodities are detected only through predicted values, and detection of the abnormal prices is inaccurate in some periods; and enterprise development is not facilitated. Collecting historical supply and marketing data through a crawler tool; dividing the historical supply and marketing data according to seasons to obtain periods of the historical supply and marketing data; predicting the demand quantity according to the period of the historical supply and marketing data to obtain a predicted demand quantity; obtaining supply quantity data of commodities; analyzing and determining the price of the commodity according to the predicted demand quantity and the supply quantity; acquiring market supply and marketing data, and judging whether the market is abnormal or not according to the market supply and marketing data; if yes, an early warning signal is generated, and an administrator is notified to adjust the price; if not, continuously detecting the market supply and marketing data; the price of the commodity can be reasonably set, and the price can be timely adjusted.

Description

Technical Field

[0001] The present invention belongs to the field of economic control, and relates to supply and marketing price control early warning technology, specifically a full-network supply and marketing early warning control system based on crawler and large model technology. Background Art

[0002] The supply and demand relationship is the interaction and mutual influence between the supply and demand of goods or services in the market, which determines the price, quantity and other market conditions of goods or services; the primary benefit of price control is to stabilize the market order. By formulating reasonable pricing strategies, brand owners and merchants can jointly maintain a fair and orderly market environment, which is conducive to fair competition among merchants; through price control strategies, companies can ensure that product prices are consistent across channels and platforms, thereby establishing a unified image of the company. Price control can also prevent the occurrence of low prices and random prices, and avoid damage to the corporate image; by formulating reasonable pricing strategies, companies can ensure the profit margins of their products and provide financial support for various links of the company such as R&D, production, and marketing; and stable prices also help companies establish stable customer relationships and promote long-term development.

[0003] The prior art (invention patent application with publication number CN111383080A) discloses a method for controlling abnormal commodity prices, which includes: creating a linear regression model, based on a training set, continuously adjusting the parameters of the linear regression model through a gradient descent algorithm, so that the cost function of the linear regression model tends to zero, and calculating the parameters of the linear regression model; substituting a test set into the linear regression model, evaluating the linear regression model, and adjusting the parameters of the linear regression model according to the evaluation results; receiving an input value of the commodity price, and calculating a predicted value of the commodity price; if the predicted value and the input value are within an allowable offset range, controlling to allow the input value to be input; if not within the allowable offset range, prompting that the input value is an abnormal price.

[0004] In the prior art, abnormal prices of commodities are detected by machine learning. However, some commodities are launched according to the season. In different seasons, the demand for some commodities is different, resulting in different prices of commodities in different periods. Therefore, only detecting abnormal prices of commodities by predicted values ​​may lead to inaccurate detection of abnormal prices in some periods, which is not conducive to the development of enterprises.

[0005] The present invention provides a full-network supply and marketing early warning management and control system based on crawler and large model technology to solve the above technical problems. Summary of the invention

[0006] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a full-network supply and marketing early warning management and control system based on crawler and large model technology, which is used to solve the technical problem that some commodities in the prior art are listed according to the season, and the demand for some commodities is different in different seasons, resulting in different prices of commodities in different periods. Therefore, only detecting abnormal prices of commodities through predicted values ​​will lead to inaccurate detection of abnormal prices in some periods; which is not conducive to the development of enterprises.

[0007] To achieve the above-mentioned purpose, the first aspect of the present invention provides a network-wide supply and marketing early warning control system based on crawler and large model technology, including: a supply and marketing control module, and a data acquisition module and an early warning adjustment module connected thereto;

[0008] Data collection module: used to collect historical supply and marketing data through crawler tools;

[0009] Supply and marketing control module: used to divide historical supply and marketing data by season to obtain the period of historical supply and marketing data; predict demand based on the period of historical supply and marketing data to obtain predicted demand; obtain commodity supply data; determine commodity prices based on predicted demand and supply analysis;

[0010] Early warning adjustment module: used to obtain market supply and marketing data, and determine whether there is any abnormality in the market based on the market supply and marketing data; if so, an early warning signal is generated and the administrator is notified to adjust the price; if not, the market supply and marketing data is continuously tested.

[0011] Preferably, forecasting the demand according to the period of historical supply and marketing data to obtain the forecasted demand includes:

[0012] Retrieve historical supply and marketing data; historical supply and marketing data includes historical sales, historical sales volume, historical pricing and historical market average price;

[0013] Retrieve the period i of historical supply and marketing data; calculate the seasonal factor Sk based on the period of historical supply and marketing data;

[0014] The forecast demand is calculated by the formula YX=F(m)×Sk; wherein the demand function F(m) is obtained based on the period analysis of historical supply and marketing data, where m represents the period.

[0015] It should be noted that the historical supply and marketing data are classified according to seasons, with each season being three months apart. December and January and February of the following year are marked as winter, March to May are marked as spring, June to August are marked as summer, and September to November are marked as autumn.

[0016] Preferably, the calculation of the seasonal factor Sk according to the period of historical supply and marketing data includes:

[0017] Retrieve the historical sales volume Di from the historical supply and marketing data of each period; where i represents the period number, and i is set to a positive integer greater than 0; through the formula Calculate the theoretical demand Dt for each period;

[0018] Divide the historical sales volume of each period by the theoretical sales volume to obtain the seasonal factor St of each period; The seasonal factors of each season are calculated; where r means there are r seasonal cycles; S 4j+k Corresponding to St.

[0019] It should be noted that k in Sk = 1, 2, 3, 4, corresponding to four different seasons respectively.

[0020] The present invention calculates theoretical demand for several periods based on historical sales volume for several periods; calculates seasonal factors for each period based on theoretical demand, and calculates seasonal factors corresponding to each season based on the number of seasonal cycles; and can take into account the impact of seasons on commodity demand, which is beneficial to improving the accuracy of demand forecasting.

[0021] Preferably, the demand function F(m) is obtained based on the period analysis of historical supply and marketing data, including:

[0022] Retrieve the theoretical demand Dt of historical sales data, and integrate the theoretical demand Dt of historical sales data into a discrete function H(t); assume that the regression model of the discrete function H(t) is y=a+bx;

[0023] The values ​​of coefficients a and b are calculated according to the least squares method. The steps are as follows:

[0024] Set the sum of squares of the errors

[0025] Find the partial derivatives of a and b with respect to S and set them equal to zero;

[0026]

[0027]

[0028] After simplifying the above formula and solving the simultaneous equations, we get:

[0029]

[0030] Solving the equation yields:

[0031]

[0032] Substitute the solved a and b into the regression model y respectively, and mark the regression model as the demand function F(m).

[0033] The present invention integrates the theoretical sales data of historical sales data into a discrete curve, and obtains the regression model function of the discrete curve; solves the coefficient value in the regression model function according to the least squares method, and marks the regression model function after the solution as a demand function; it can provide a functional basis for the prediction of demand, which is conducive to improving the accuracy of the prediction results.

[0034] Preferably, determining the price of the commodity based on the predicted demand and supply analysis includes:

[0035] Retrieve the forecasted demand; obtain the demand price database; match the forecasted demand with the demand price database to obtain the initial price corresponding to the product;

[0036] Retrieve the supply of the corresponding product; calculate the demand difference between the supply and demand; mark the initial price and the demand difference as CJ and XC respectively; calculate the price of the product through the formula JG=CJ-[(α×CZ)^3+β]; where α is a proportional coefficient greater than 0.

[0037] It should be noted that when calculating the demand difference between supply and demand, the demand difference is obtained by subtracting the demand from the supply; the initial price is always higher than the lowest selling price of the product and lower than the highest price of the product; α is set based on experience from actual conditions, and when the calculated price is lower than the lowest selling price, the product is sold at the lowest selling price; when the calculated price is greater than the highest selling price, the product is sold at the highest selling price.

[0038] The present invention determines the initial price of a commodity according to the predicted demand, calculates the demand difference between the supply and the demand, and adjusts the initial price based on the demand difference to obtain the price of the corresponding commodity; it can ensure the rationality of the commodity price setting, which is conducive to increasing the sales volume of the enterprise's commodities and increasing the profits of the enterprise's commodities.

[0039] Preferably, judging whether the market is abnormal based on the market supply and marketing data includes:

[0040] Retrieve the market supply and marketing data within a set time period; use interpolation to fit the market supply and marketing data into a price fluctuation curve Q(T); where the market supply and marketing data includes the collection time and the average market price;

[0041] Solve the derivative q(T) of the price fluctuation curve Q(T); determine whether the absolute value of the derivative q(T) is greater than the fluctuation threshold; if yes, it is determined that the market is abnormal; if not, it is determined that the market is not abnormal; wherein the fluctuation threshold is set according to the season.

[0042] The present invention fits the market supply and marketing data within a set time period into a price fluctuation curve; and solves the derivative of the price fluctuation curve. When the absolute value of the derivative is greater than the set fluctuation threshold, it is determined that the market is abnormal, and the average price of the market can be monitored in time. When the price fluctuates greatly, the commodity price is adjusted in time, which is conducive to ensuring that the company's commodities have a certain competitiveness in the market.

[0043] Preferably, the fluctuation threshold is set according to the season, including:

[0044] Obtain the time the product is in the market, and set the fluctuation range according to the time the product is in the market; the fluctuation range includes a violent fluctuation area and a gentle fluctuation area;

[0045] Obtain the fluctuation standard value; when the fluctuation range is a violent fluctuation area, adjust the fluctuation standard value upward to obtain the corresponding fluctuation threshold; when the fluctuation range is a gentle fluctuation area, adjust the fluctuation standard value downward to obtain the corresponding fluctuation threshold.

[0046] It should be noted that the time a commodity is on the market is determined according to the season; the range of upward and downward adjustment of the fluctuation standard value is set according to the type of commodity after expert evaluation, and when adjusting upward, each commodity has a maximum price fluctuation threshold. When adjusting upward, it is necessary to ensure that the fluctuation threshold is not greater than the maximum price fluctuation threshold.

[0047] The present invention sets the fluctuation range of commodities according to the time when commodities are on the market; adjusts the fluctuation standard value upward or downward according to the fluctuation range of commodities to obtain the fluctuation threshold value of the corresponding fluctuation range; can set the fluctuation threshold value according to different periods, and dynamically adjusts the fluctuation threshold value, which is beneficial to improving the rationality of the setting of the fluctuation threshold value.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] 1. The present invention collects historical supply and marketing data through a crawler tool, divides the historical supply and marketing data according to seasons, and obtains the periods of the historical supply and marketing data; calculates the theoretical demand for several periods according to the historical sales volume of several periods; calculates the seasonal factor of each period according to the theoretical demand, and calculates the seasonal factor corresponding to each season according to the number of seasonal cycles; can consider the impact of season on commodity demand, which is conducive to improving the accuracy of predicted demand; integrates the theoretical sales data of historical sales data into a discrete curve, and calculates the regression model function of the discrete curve; solves the coefficient value in the regression model function according to the least squares method, and marks the regression model function after the solution as a demand function; can provide a function basis for demand prediction, which is conducive to improving the accuracy of prediction results; determines the initial price of the commodity according to the predicted demand, calculates the demand difference between the supply and demand, adjusts the initial price based on the demand difference, and obtains the price of the corresponding commodity; can ensure the rationality of commodity price setting, which is conducive to increasing the sales volume of the enterprise's commodities and increasing the profit of the enterprise's commodities.

[0050] 2. The present invention fits the market supply and marketing data within a set time period into a price fluctuation curve; and solves the derivative of the price fluctuation curve. When the absolute value of the derivative is greater than the set fluctuation threshold, it is determined that the market is abnormal, and the average price of the market can be monitored in time. When the price fluctuates greatly, the commodity price is adjusted in time, which is conducive to ensuring that the company's commodities have a certain competitiveness in the market; according to the commodity's time on the market, the commodity's fluctuation range is set; according to the commodity's fluctuation range, the fluctuation standard value is adjusted upward or downward to obtain the fluctuation threshold of the corresponding fluctuation range; the fluctuation threshold can be set according to different periods, and the dynamic adjustment of the fluctuation threshold is conducive to improving the rationality of the fluctuation threshold setting. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0052] Figure 1 It is a schematic diagram of the overall steps of the present invention;

[0053] Figure 2 A schematic diagram of the steps of determining commodity prices according to the present invention;

[0054] Figure 3 A schematic diagram of the market abnormality early warning steps of the present invention;

[0055] Figure 4 Schematic diagram of the derivative of the wave curve of the present invention. DETAILED DESCRIPTION

[0056] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than 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.

[0057] See also Figure 1 , the first aspect of the present invention provides a network-wide supply and marketing early warning control system based on crawler and large model technology, including: a supply and marketing control module, and a data acquisition module and an early warning adjustment module connected thereto;

[0058] Data collection module: used to collect historical supply and marketing data through crawler tools;

[0059] Supply and marketing control module: used to divide historical supply and marketing data by season to obtain the period of historical supply and marketing data; predict demand based on the period of historical supply and marketing data to obtain predicted demand; obtain commodity supply data; determine commodity prices based on predicted demand and supply analysis;

[0060] Early warning adjustment module: used to obtain market supply and marketing data, and determine whether there is any abnormality in the market based on the market supply and marketing data; if so, an early warning signal is generated and the administrator is notified to adjust the price; if not, the market supply and marketing data is continuously tested.

[0061] It should be noted that crawler tools include WebMagic, XPath Helper, CooSeeker, etc. Enterprises can choose appropriate crawler tools according to their needs to obtain historical supply and marketing data.

[0062] See also Figure 2 , collect historical supply and marketing data through crawler tools; divide the historical supply and marketing data according to seasons to obtain the periods of historical supply and marketing data; retrieve the historical sales volume Di in the historical supply and marketing data of each period; where i represents the period number, and i is set to a positive integer greater than 0; through the formula The theoretical demand Dt for each period is calculated; where t is a positive integer greater than 2, and t corresponds to i;

[0063] Retrieve the theoretical demand Dt of historical sales data, and integrate the theoretical demand Dt of historical sales data into a discrete function H(t); assume that the regression model of the discrete function H(t) is y=a+bx;

[0064] The values ​​of coefficients a and b are calculated according to the least squares method. The steps are as follows:

[0065] Set the sum of squares of the errors

[0066] Find the partial derivatives of a and b with respect to S and set them equal to zero;

[0067]

[0068] After simplifying the above formula and solving the simultaneous equations, we get:

[0069]

[0070] Solving the equation yields:

[0071]

[0072] Substitute the solved a and b into the regression model y respectively, and mark the regression model as the demand function F(m) where n represents the maximum value of the period corresponding to the theoretical demand; the value of m is t-2;

[0073] Divide the historical sales volume of each period by the theoretical sales volume to obtain the seasonal factor St of each period; The seasonal factors of each season are calculated; where r means there are r seasonal cycles; S 4j+k Corresponding to St.

[0074] For example, suppose that Company A is a down jacket merchant. Now we use the crawler tool XPath Helper to collect the historical sales data of Company A. The historical sales data of the company in the past three years is as follows:

[0075] years season Period t Historical sales volume Theoretical sales volume Seasonal influence 2021 spring 1 8000 10529 0.76 2021 summer 2 10000 11983 0.83 2021 autumn 3 22000 13834 1.59 2021 winter 4 32000 14667 2.18 2022 spring 5 9000 23667 0.38 2022 summer 6 16000 18334 0.87 2022 autumn 7 24000 21167 1.13 2022 winter 8 36000 16334 2.20 2023 spring 9 10000 25667 0.39 2023 summer 10 14000 26166 0.54 2023 autumn 11 30000 26523 1.13 2023 winter 12 41000 27977 1.47

[0076] Integrate the theoretical demand of historical sales data into a discrete function H(t); assume that the regression model of the discrete function H(t) is y=a+bx; calculate the coefficients a and b in the regression model according to the above table;

[0077] By calculating the formula, we get a≈13437; b≈1454; substituting the coefficients into the regression model, we get the regression model y=13437+1454x; completing the above table according to the regression model, and marking the regression model as the demand function F(m); calculating the seasonal factors of each period, and completing the above table;

[0078] Calculated by the formulas, we obtain S1=0.51; S2=0.75; S3=1.28; S4=1.95.

[0079] The forecast demand is calculated by the formula YX=F(m)×Sk; wherein the demand function F(m) is obtained based on the period analysis of historical supply and marketing data, where m represents the period.

[0080] Retrieve the forecasted demand; obtain the demand price database; match the forecasted demand with the demand price database to obtain the initial price corresponding to the product;

[0081] Retrieve the supply of the corresponding product; calculate the demand difference between the supply and demand; mark the initial price and the demand difference as CJ and XC respectively; calculate the price of the product using the formula JG=CJ-(α×CZ)^3; where α is a proportional coefficient greater than 0.

[0082] For example: suppose that the demand of enterprise A in spring and winter in 2024 is predicted; according to the formula, the predicted demand in spring is YX=(13437+1454×11)×0.51=15010;

[0083] Winter forecast demand YX = (13437 + 1454 × 14) × 1.95 = 65896;

[0084] According to the predicted demand, the initial price of the spring down jacket is set at 560 per piece; the supply of spring down jackets is obtained as 20,000; the difference between the two is calculated to be 4990; then α is set to 0.001; the price of the down jacket calculated by the formula is 435.75, which is greater than the minimum selling price of 400; when the down jacket is sold in spring, the price is set to 435.75.

[0085] The initial price of winter down jackets is set at 500 per piece; the supply of winter down jackets is obtained as 60,000; the difference between the two is calculated to be -5896; then α is set to 0.001; the price of the down jacket is calculated by the formula to be 704.96; which is less than the maximum selling price of 800; when selling down jackets in winter, the price is set to 704.96.

[0086] See also Figure 3-Figure 4 , obtain the market supply and marketing data within the set time period; use interpolation to fit the market supply and marketing data into a price fluctuation curve Q(T); where the market supply and marketing data includes the collection time and the market average price; solve the derivative q(T) of the price fluctuation curve Q(T); determine whether the absolute value of the derivative q(T) is greater than the fluctuation threshold; if yes, generate an early warning signal and notify the management personnel to adjust the price; if not, continue to detect the market supply and marketing data; where the fluctuation threshold is set according to the season.

[0087] It is worth noting that the volatility thresholds are set according to seasons and include:

[0088] Obtain the time the product is in the market, and set the fluctuation range according to the time the product is in the market; the fluctuation range includes a violent fluctuation area and a gentle fluctuation area;

[0089] Obtain the fluctuation standard value; when the fluctuation range is a violent fluctuation area, adjust the fluctuation standard value upward to obtain the corresponding fluctuation threshold; when the fluctuation range is a gentle fluctuation area, adjust the fluctuation standard value downward to obtain the corresponding fluctuation threshold.

[0090] For example, suppose that the price fluctuation threshold of down jackets sold by Enterprise A is set; if the down jackets are available in the market in winter, November 20 to December 5 and February 25 to March 10 are divided into the violent fluctuation zone; December 6 to February 24 are divided into the gentle fluctuation zone;

[0091] The fluctuation standard value for obtaining price fluctuations is 100. In the area of ​​drastic fluctuations, the fluctuation standard value is adjusted upward to 170; in the area of ​​gentle fluctuations, the fluctuation standard value is adjusted downward to 50; the fluctuation threshold is the fluctuation threshold within two days.

[0092] Obtain market supply and marketing data in real time, and fit the market supply and marketing data into a price fluctuation curve Q(T); solve the derivative q(T) of the price fluctuation curve Q(T); judge the absolute value of the derivative, and the value of q(T) collected on November 27 was 160; if the absolute value is less than the fluctuation threshold, it is a normal fluctuation in market prices; the value of q(T) collected on January 13 was 70; if the absolute value is greater than the fluctuation threshold, a warning signal is generated and management personnel are notified to adjust prices.

[0093] It should be noted that when collecting market supply and marketing data, the collection frequency is based on the specifically set fluctuation range. In the above example, based on the setting of the fluctuation range, the collection frequency is to collect market supply and marketing data once a day.

[0094] It should be noted that when adjusting prices, the prices of corresponding commodities can be raised or lowered by a corresponding percentage according to changes in the average market price; or the adjustments can be made by the company's internal personnel based on their experience.

[0095] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0096] The working principle of the present invention is as follows: historical supply and marketing data are collected through crawler tools; the historical supply and marketing data are divided according to seasons to obtain the period of the historical supply and marketing data; the demand is predicted according to the period of the historical supply and marketing data to obtain the predicted demand; the supply data of the goods are obtained; the price of the goods is determined according to the predicted demand and supply analysis; the market supply and marketing data is obtained, and whether the market is abnormal is determined according to the market supply and marketing data; if yes, an early warning signal is generated and the administrator is notified to adjust the price; if not, the market supply and marketing data is continuously tested.

[0097] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. The whole network supply and marketing early warning control system based on crawler and large model technology, including: The supply and marketing control module, and the data collection module and early warning adjustment module connected thereto are characterized in that: Data collection module: used to collect historical supply and marketing data through crawler tools; Supply and marketing control module: used to divide historical supply and marketing data by season to obtain the period of historical supply and marketing data; predict demand based on the period of historical supply and marketing data to obtain predicted demand; obtain commodity supply data; determine commodity prices based on predicted demand and supply analysis; Early warning adjustment module: used to obtain market supply and marketing data, and determine whether there is any abnormality in the market based on the market supply and marketing data; if so, an early warning signal is generated and the administrator is notified to adjust the price; if not, the market supply and marketing data is continuously tested.

2. The whole network supply and marketing early warning control system based on crawler and large model technology according to claim 1 is characterized in that: The forecasting of demand according to the period of historical supply and marketing data to obtain the forecasted demand includes: Retrieve historical supply and marketing data; historical supply and marketing data includes historical sales, historical sales volume, historical pricing and historical market average price; Retrieve the period i of historical supply and marketing data; calculate the seasonal factor Sk based on the period of historical supply and marketing data; The forecast demand is calculated by the formula YX=F(m)×Sk; wherein the demand function F(m) is obtained based on the period analysis of historical supply and marketing data, where m represents the period.

3. The whole network supply and marketing early warning control system based on crawler and large model technology according to claim 2 is characterized in that: The seasonal factor Sk is calculated based on the period of historical supply and marketing data, including: Retrieve the historical sales volume Di from the historical supply and marketing data of each period; where i represents the period number, and i is set to a positive integer greater than 0; through the formula Calculate the theoretical demand Dt for each period; Divide the historical sales volume of each period by the theoretical sales volume to obtain the seasonal factor St of each period; The seasonal factors of each season are calculated; where r means there are r seasonal cycles; S 4j+k Corresponding to St.

4. The whole network supply and marketing early warning control system based on crawler and large model technology according to claim 2 is characterized in that: The demand function F(m) is obtained by period analysis based on historical supply and marketing data, including: Retrieve the theoretical demand Dt of historical sales data, and integrate the theoretical demand Dt of historical sales data into a discrete function H(t); assume that the regression model of the discrete function H(t) is y=a+bx; The values ​​of coefficients a and b are calculated according to the least squares method. The steps are as follows: Set the sum of squares of the error Find the partial derivatives of a and b with respect to S and set them equal to zero; After simplifying the above formula and solving the simultaneous equations, we get: Solving the equation yields: Substitute the solved a and b into the regression model y respectively, and mark the regression model as the demand function F(m).

5. The whole network supply and marketing early warning control system based on crawler and large model technology according to claim 1 is characterized in that: The method of determining the price of a commodity based on the forecasted demand and supply analysis includes: Retrieve the forecasted demand; obtain the demand price database; match the forecasted demand with the demand price database to obtain the initial price corresponding to the product; Retrieve the supply of the corresponding product; calculate the demand difference between the supply and demand; mark the initial price and the demand difference as CJ and XC respectively; calculate the price of the product using the formula JG=CJ-(α×CZ)^3; where α is a proportional coefficient greater than 0.

6. The whole network supply and marketing early warning control system based on crawler and large model technology according to claim 1 is characterized in that: The determining whether the market is abnormal based on the market supply and marketing data includes: Retrieve the market supply and marketing data within a set time period; use interpolation to fit the market supply and marketing data into a price fluctuation curve Q(T); where the market supply and marketing data includes the collection time and the average market price; Solve the derivative q(T) of the price fluctuation curve Q(T); determine whether the absolute value of the derivative q(T) is greater than the fluctuation threshold; if yes, it is determined that the market is abnormal; if not, it is determined that the market is not abnormal; wherein the fluctuation threshold is set according to the season.

7. The whole network supply and marketing early warning control system based on crawler and large model technology according to claim 6 is characterized in that: The fluctuation threshold is set according to the season and includes: Obtain the time the product is in the market, and set the fluctuation range according to the time the product is in the market; the fluctuation range includes a violent fluctuation area and a gentle fluctuation area; Obtain the fluctuation standard value; when the fluctuation range is a violent fluctuation area, adjust the fluctuation standard value upward to obtain the corresponding fluctuation threshold; when the fluctuation range is a gentle fluctuation area, adjust the fluctuation standard value downward to obtain the corresponding fluctuation threshold.

Images