Fresh milk price prediction system based on big data
By using a combination of adaptive clustering algorithms and game theory models in the raw milk market, the shortcomings of fixed algorithms and linear models in processing complex market data are solved, and more efficient and accurate price predictions are achieved.
Patent Information
- Application Number
- CN202411104126.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-13
- Publication Date
- 2024-11-29
AI Technical Summary
In the data processing and analysis of raw milk market, the clustering algorithm structure is fixed, which is difficult to adapt to data changes, ignores the interaction between market participants, and the linear model is difficult to capture price nonlinear fluctuations, resulting in low prediction accuracy and deviation.
Adopt a fresh milk price prediction system based on big data, including adaptive clustering algorithm, game theory modeling and mixed strategy Nash equilibrium analysis, dynamically adjust market status, build game theory models, use the optimal strategy combination to predict prices, and use the Historical data verification and correction model.
It improves data processing efficiency and accuracy, enhances simulation accuracy for complex market behaviors, accurately captures nonlinear price fluctuations, and improves prediction accuracy and model adaptability.
Smart Images

Figure CN119048138A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of agricultural economy, and in particular to a fresh milk price prediction system based on big data. Background Art
[0002] Raw milk market data is characterized by large scale, multi-dimensionality, high timeliness and complex correlation, which poses severe technical challenges to data processing and analysis. At present, raw milk market data processing and analysis mainly rely on traditional data processing methods and machine learning algorithms. These methods include time series analysis, regression analysis and various clustering algorithms. These technologies process and analyze historical market data and build data models to perform data classification, pattern recognition and trend analysis. However, these methods still have technical limitations when processing large-scale and complex raw milk market data.
[0003] The existing technology faces the following technical challenges in the processing and analysis of raw milk market data: Most of the existing clustering algorithms adopt a fixed algorithm structure and lack a dynamic adjustment mechanism, which results in the inability to effectively adapt to changes in data distribution when processing variable market data, affecting the efficiency and accuracy of data processing; especially when it is necessary to divide different market stages and identify complex market states, fixed-structure algorithms are difficult to accurately capture the dynamic changes of the market; traditional market analysis models often ignore the complex interactive relationships between market participants; the raw milk market involves multiple participants such as producers, wholesalers, retailers and consumers, and there are complex game relationships between them. It is difficult for existing analysis methods to fully consider these factors, resulting in limited prediction accuracy of the model; existing price prediction methods are mainly based on linear models, which are difficult to accurately capture the nonlinear fluctuation characteristics of raw milk market prices; and linear prediction models often show large prediction deviations when processing such complex price data. Therefore, it is urgent to develop a new technical solution to solve the above problems. Summary of the invention
[0004] The invention provides a raw milk price prediction system based on big data, so as to solve the problems that most of the existing clustering algorithms adopt a fixed algorithm structure and lack a dynamic adjustment mechanism, which leads to the inability to effectively adapt to the change of data distribution when processing changeable market data, affecting the efficiency and accuracy of data processing; when it is necessary to divide different market stages and identify complex market states, the fixed structure algorithm is difficult to accurately capture the dynamic changes of the market; the traditional market analysis model often ignores the complex interactive relationship between market participants; the raw milk market involves multiple participants such as producers, wholesalers, retailers and consumers, and there are complex game relationships between them. It is difficult for the existing analysis method to fully consider these factors, resulting in limited prediction accuracy of the model; the existing price prediction method is mainly based on a linear model, which is difficult to accurately capture the nonlinear fluctuation characteristics of the raw milk market price; and the linear prediction model often shows a large prediction deviation when processing such complex price data.
[0005] The present invention provides a fresh milk price prediction system based on big data, which specifically includes the following technical solutions:
[0006] A fresh milk price prediction system based on big data comprises: a data acquisition module for collecting original market data; a data processing module for preprocessing the original market data to obtain market data; a data analysis module for implementing an adaptive clustering algorithm on the market data to dynamically adjust the market status; a game theory modeling module for constructing a game theory model, introducing a mixed strategy Nash equilibrium analysis, and calculating an optimal strategy combination; a price prediction module for predicting the price of fresh milk using the optimal strategy combination, revising the game theory model, and predicting the price of fresh milk again using the revised game theory model to obtain a prediction result; and a display module for visually displaying the prediction result; wherein the data acquisition module is connected to the data processing module by means of data transmission, the data processing module is connected to the data analysis module by means of data transmission, the data analysis module is connected to the game theory modeling module by means of data transmission, the game theory modeling module is connected to the price prediction module by means of data transmission, and the price prediction module is connected to the display module by means of data transmission.
[0007] Preferably, the adaptive clustering algorithm in the data analysis module divides the market data into two or more stages, each stage representing a different market state and price fluctuation pattern; wherein the market state is represented by the number of clusters and cluster centers, and the price fluctuation pattern is represented by the intra-cluster variance and the inter-cluster variance.
[0008] Preferably, the implementation process of the adaptive clustering algorithm includes: the first step of setting the initial number of clusters and the initial cluster center; the second step of dynamically adjusting the cluster center; the third step of calculating the intra-cluster variance and the inter-cluster variance; the fourth step of dynamically adjusting the number of clusters by comparing the ratio of the intra-cluster variance to the inter-cluster variance.
[0009] Preferably, the process of dynamically adjusting the cluster centers includes calculating and updating the cluster centers of each cluster according to the data points belonging to each cluster in the current iteration; and optimizing the adaptive clustering result through the dynamically adjusted number of clusters and cluster centers.
[0010] Preferably, the game theory model constructed by the game theory modeling module includes defining a producer's profit function.
[0011] Preferably, the mixed strategy Nash equilibrium analysis introduced by the game theory modeling module includes calculating the producer's optimal mixed strategy, which is based on the producer's profit function and the strategy combination of other participants; based on the optimal mixed strategy, determining the optimal mixed strategy selection of the participants to obtain the optimal strategy combination for each market stage.
[0012] Preferably, the process of the price prediction module predicting the price of raw milk using the optimal strategy combination includes considering wholesale price, retail price, market demand and market benchmark demand.
[0013] Preferably, the process of the price prediction module correcting the game theory model includes verifying the predicted raw milk price using historical market data, and optimizing the game theory model by adjusting fitting parameters.
[0014] A method for predicting raw milk prices based on big data comprises the following steps:
[0015] S1: pre-processing the collected original market data to obtain market data; implementing an adaptive clustering algorithm on the market data to dynamically adjust the market status;
[0016] S2: Construct a game theory model and introduce mixed strategy Nash equilibrium analysis to calculate the optimal strategy combination;
[0017] S3: using the optimal strategy combination to predict the price of raw milk;
[0018] S4: modifying the game theory model, and using the modified game theory model to predict the price of raw milk to obtain a prediction result;
[0019] S5: Visualize the prediction results.
[0020] Preferably, the S2 specifically includes:
[0021] Through mixed strategy Nash equilibrium analysis, the optimal mixed strategy of participants under different strategy combinations is calculated; based on the optimal mixed strategy, the optimal mixed strategy selection of participants is determined to obtain the optimal strategy combination.
[0022] The beneficial effects of the technical solution of the present invention are:
[0023] 1. The adaptive clustering algorithm improves the processing capability of market data. By dynamically adjusting the number of clusters and cluster centers, the algorithm can more accurately identify and characterize different market conditions, providing a more reliable technical basis for subsequent data analysis and improving the data processing efficiency and accuracy of the raw milk price prediction system.
[0024] 2. Through the game theory model, the raw milk price prediction system can more comprehensively process and analyze the complex interaction data between market participants; the introduction of mixed strategy Nash equilibrium analysis further enhances the data processing capabilities of the game theory model, improves the reliability of data analysis, and improves the simulation accuracy of the raw milk price prediction system for complex market behaviors.
[0025] 3. The introduction of nonlinear terms enhances the ability of the raw milk price forecasting system to process nonlinear price data, enabling the raw milk price forecasting system to more accurately simulate and predict complex price fluctuations, improve the accuracy of the forecast, and improve the reliability of data analysis.
[0026] 4. By introducing historical market data to verify and correct the game theory model, the accuracy and reliability of the prediction model have been significantly improved. This technical means enables the raw milk price prediction system to continuously optimize the prediction model based on actual market data, improves the reliability of data analysis, and thus improves the adaptability and prediction accuracy of the raw milk price prediction system. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 This is a structural diagram of a raw milk price prediction system based on big data according to the present invention;
[0028] Figure 2 This is a flow chart of a method for predicting raw milk prices based on big data described in the present invention. DETAILED DESCRIPTION
[0029] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0030] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
[0031] The specific scheme of the fresh milk price prediction system based on big data provided by the present invention is described in detail below with reference to the accompanying drawings.
[0032] See attached Figure 1 , which shows a structure diagram of a fresh milk price prediction system based on big data provided by an embodiment of the present invention, the system includes the following parts:
[0033] Data acquisition module, data processing module, data analysis module, game theory modeling module, price prediction module and display module;
[0034] The data collection module collects raw market data from dairy product manufacturers, wholesale markets, retailers, and e-commerce platforms for subsequent processing. The data collection module is connected to the data processing module by means of data transmission;
[0035] The data processing module performs pre-processing such as data cleaning and data normalization on the original market data from the data acquisition module and outputs the market data. The data processing module is connected to the data analysis module by means of data transmission;
[0036] The data analysis module implements an adaptive clustering algorithm on the market data to divide the market data into multiple stages, each stage representing a different market state and price fluctuation pattern; wherein the market state is represented by the number of clusters and cluster centers, and the price fluctuation pattern is represented by the intra-cluster variance and the inter-cluster variance; specifically, the number of clusters and cluster centers are dynamically adjusted according to the distribution of data points in the market data, and the market stages are divided using the dynamically adjusted number of clusters and the cluster center of each cluster;
[0037] The implementation process of the adaptive clustering algorithm includes: the first step is to set the initial number of clusters and the initial cluster center; the second step is to dynamically adjust the cluster center; the third step is to calculate the intra-cluster variance and the inter-cluster variance; the fourth step is to dynamically adjust the number of clusters by comparing the ratio of the intra-cluster variance to the inter-cluster variance; the process of dynamically adjusting the cluster center includes calculating and updating the cluster center of each cluster according to the data points belonging to each cluster in the current iteration; the adaptive clustering result is optimized by the dynamically adjusted number of clusters and cluster centers; the data analysis module is connected to the game theory modeling module by data transmission;
[0038] The game theory modeling module performs game modeling on producers, wholesalers, retailers and consumers in each market stage, defines the producer's profit function, and the profit function takes into account wholesale price, production volume and production cost; uses mixed strategy Nash equilibrium analysis to calculate the optimal mixed strategy in each market stage, and the optimal mixed strategy is based on the producer's profit function and the strategy combination of other participants; based on the optimal mixed strategy, the optimal mixed strategy selection of each participant in the market stage is determined through the mixed strategy probability distribution of each participant, and the game theory modeling module is connected to the price prediction module by data transmission;
[0039] The price prediction module performs price prediction based on the optimal strategy combination to obtain the predicted raw milk price, and the formula is as follows:
[0040]
[0041] Among them, P future is the predicted raw milk price; N is the number of market stages, and the value is the number of clusters in the adaptive clustering algorithm; is the wholesale price of the participants in the θth market stage in the optimal strategy portfolio; is the retail price chosen by the participants in the optimal strategy combination in the θth market stage; Q θ is the market demand in the θth market stage, obtained by time series analysis of historical market demand data from historical market data; ∈ is a constant, which is set according to the specific scenario to avoid logarithmic zero value; is the market benchmark demand in the θth market stage; v is the response speed of the game theory model; and v are fitting parameters, and the specific values are obtained by regression analysis based on historical market data, reflecting the basic level and response speed of market demand. Time series analysis and calculation of fitting parameters are existing technologies and will not be elaborated here. The predicted raw milk price is verified based on historical market data, and the game theory model is corrected by adjusting the fitting parameters. The raw milk price is predicted using the corrected game theory model to obtain the prediction result. The price prediction module is connected to the display module by data transmission.
[0042] The display module visualizes the forecast results and intuitively displays the forecast results through charts and other means to help users understand market trends in a timely manner.
[0043] See attached Figure 2 , which shows a flow chart of a method for predicting raw milk prices based on big data provided by an embodiment of the present invention, the method comprising the following steps:
[0044] S1: Preprocess the collected raw market data to obtain market data; implement an adaptive clustering algorithm on the market data, dynamically adjust the market status, and divide the market stages;
[0045] The original market data of the raw milk market is collected from dairy product manufacturers, wholesale markets, retailers, and e-commerce platforms, including prices (producer prices, wholesale prices, and retail prices), transaction volumes (daily or weekly sales), and supply and demand information (supply and demand balance data in the market, which describes the relationship between the supply and demand of raw milk in the market). Next, the collected original market data is preprocessed to obtain market data; preprocessing includes data cleaning and normalization, and data cleaning includes processing missing values and outliers. The above preprocessing processes all use existing technologies and will not be described in detail here.
[0046] After data preprocessing is completed, the adaptive clustering algorithm is implemented;
[0047] The adaptive clustering algorithm divides market data into multiple stages, each of which represents a different market state and price fluctuation pattern. The market state is represented by the number of clusters and cluster centers. The number of clusters is a numerical value that indicates the number of currently identified market states, reflecting the complexity of the market; the cluster center is a numerical value that indicates the central value of each cluster (market state), reflecting the characteristic mean of the market state, and the characteristic can be price or trading volume. The data points in the market data are specific market data values, such as prices (producer prices, wholesale prices, and retail prices) and trading volumes at different time points. The number of data points indicates the number of data belonging to a certain market state. The adaptive clustering algorithm dynamically adjusts the number of clusters and cluster centers, and continuously optimizes the adaptive clustering results according to the distribution of data points in the market data to better reflect the actual market situation.
[0048] The price fluctuation pattern is represented by the intra-cluster variance and the inter-cluster variance. The intra-cluster variance is a numerical value used to measure the degree of dispersion between data points in the same cluster. The larger the intra-cluster variance value, the greater the price and trading volume fluctuations under the market state; the smaller the intra-cluster variance value, the more stable the price and trading volume under the market state. The inter-cluster variance is a numerical value used to measure the degree of dispersion between different clusters; the larger the inter-cluster variance value, the greater the difference between different market states; the smaller the inter-cluster variance value, the more similar the market states are.
[0049] The adaptive clustering algorithm process is as follows:
[0050] First, statistical analysis and distribution analysis are performed on historical market data; the initial cluster number K0 and initial cluster center μ are set based on expert experience. k,0 The specific setting method is set according to the specific implementation scenario.
[0051] Dynamically adjust the cluster center, the formula is as follows:
[0052]
[0053] Among them, μ k,t+1 is the cluster center of the kth cluster after the t+1th iteration; k is the index of the cluster; n k,t is the number of data points belonging to the kth cluster in the tth iteration; x i,t represents the i-th data point in the t-th iteration, which is the actual data point value obtained in the market data; exp represents the exponential function; ∈ is a constant, which is set according to the specific scenario and is used to avoid logarithmic zero value.
[0054] The formula for calculating the intra-cluster variance is as follows:
[0055]
[0056] Among them, σ within,t The within-cluster variance at the t-th iteration, where the subscript within means within the cluster; K t is the number of clusters at the tth iteration; μ k,t is the cluster center of the kth cluster after the tth iteration;
[0057] The formula for calculating the inter-cluster variance is as follows:
[0058]
[0059] Among them, σ between,t is the between-cluster variance at the t-th iteration, the subscript between indicates between clusters; μ all,t is the mean of all data points in the tth iteration, that is, the comprehensive average of the entire market. The subscript all represents all data points; N t is the number of all data points in the tth iteration.
[0060] The adaptive clustering algorithm dynamically adjusts the number of clusters by comparing the ratio of intra-cluster variance to inter-cluster variance to better reflect the actual market situation and price fluctuation patterns. The formula is as follows:
[0061]
[0062] Among them, K t+1 is the number of clusters at the t+1th iteration; α and β are preset thresholds, which are set according to the specific implementation scenario.
[0063] The number of clusters and the cluster center of each cluster after the adaptive clustering algorithm dynamically adjusts, that is, the adaptive clustering results, reflect the different stages of market data, for example: the high demand and low supply stage, characterized by high prices and large trading volumes, the values of price and trading volume data points at the cluster center are higher than the mean and variance of the overall market data; the supply and demand balance stage, characterized by relatively stable prices and trading volumes, the values of price and trading volume data points at the cluster center are close to the mean of the overall market data, and the variance within the cluster is small, indicating small fluctuations; the low demand and high supply stage, this stage is characterized by low prices and small trading volumes, and the values of price and trading volume data points at the cluster center are significantly lower than the mean and variance of the overall market data. The specific market stage division criteria can be set according to the specific implementation scenario.
[0064] S2: Based on the divided market stages, a game theory model is constructed, and mixed strategy Nash equilibrium analysis is introduced to calculate the optimal strategy combination for each market stage; the price of raw milk is predicted using the optimal strategy combination; the game theory model is revised, and the price of raw milk is predicted using the revised game theory model, and the prediction results are visualized.
[0065] Based on the market stages divided according to the adaptive clustering results, the participants in each market stage are gamed, including producers, wholesalers, retailers and consumers. A game theory model is designed to consider the strategies and benefits of the participants.
[0066] The game theory model simulates the strategic choices and benefits between participants to find the market equilibrium state and predict the price of raw milk. The input data of the game theory model include the strategy combination, cost, price and demand of the participants. Among them, the strategy combination is the production volume and ex-factory price of the producer, the wholesale price and purchase volume of the wholesaler, the retail price and inventory of the retailer, and the purchase volume and purchase channel selection of the consumer. The output data is the optimal strategy combination, including the production volume, wholesale price and retail price of each participant at different market stages.
[0067] The introduction of mixed strategy Nash equilibrium analysis enables the game theory model to more accurately reflect market dynamics. Mixed strategy Nash equilibrium analysis is a game theory method used to find the optimal mixed strategy of participants under different strategy combinations. In the optimal mixed strategy, each participant's strategy selection is optimal, and no participant can obtain higher returns by unilaterally changing his or her strategy.
[0068] The process of building a game theory model is as follows:
[0069] The producer's profit function is defined as:
[0070]
[0071] Among them, Uj is the revenue of producer j; pw j is the wholesale price of the wholesaler; q j is the output of the producer; C(q j ) represents the production quantity q j cost.
[0072] Then, the optimal mixed strategy for each market stage is calculated using mixed strategy Nash equilibrium analysis:
[0073]
[0074]
[0075] in, is the optimal mixed strategy of producer j; ζ j is the mixed strategy of producer j; ζ j (s) is the probability that producer j chooses mixed strategy s; S j is the set of all possible mixed strategies, which is calculated using the strategy enumeration algorithm and will not be described here; is the strategy combination of producer j in mixed strategy s and the strategies of participants other than producer j The profit function under q j (s) is the production quantity of the producer under mixed strategy s; C(q j (s)) is the production quantity q of the producer under mixed strategy s j The cost of (s); s is a specific strategy in the set of mixed strategies; is the strategy combination of participants except producer j.
[0076] After obtaining the optimal mixed strategy in each market stage, the optimal mixed strategy selection of each participant in the market stage is determined through the probability distribution of the mixed strategy of each participant, and the optimal strategy combination in each market stage is obtained. The raw milk price is predicted using the optimal strategy combination. The formula is as follows:
[0077]
[0078] Among them, P future is the predicted raw milk price; N is the number of market stages, and the value is the number of clusters in the adaptive clustering algorithm; is the wholesale price of the participants in the θth market stage in the optimal strategy portfolio; is the retail price chosen by the participants in the optimal strategy combination in the θth market stage; Q θ is the market demand in the θth market stage, obtained by time series analysis of historical market demand data from historical market data; is the market benchmark demand in the θth market stage; v is the response speed of the game theory model; and v are fitting parameters. The specific values are obtained by regression analysis based on historical market data, reflecting the basic level and response speed of market demand. Time series analysis and calculation of fitting parameters are existing technologies and will not be elaborated here.
[0079] The predicted raw milk price is verified by historical market data, which can be obtained through dairy manufacturers, wholesale markets, retailers, e-commerce platforms and other channels. The method for verifying the predicted raw milk price adopts the backtesting method, which applies the game theory model to historical data and compares the difference between the predicted results and the actual data. The backtesting method is an existing technology and will not be described in detail here. By adjusting the fitting parameters, the game theory model is better adapted to the market data, the accuracy and reliability of the game theory model are improved, and finally a revised game theory model is obtained.
[0080] Finally, the revised game theory model is used to predict the price of raw milk, and the prediction results are visualized and displayed intuitively through charts and other means to help users understand market trends in a timely manner.
[0081] In summary, a raw milk price prediction system based on big data was completed.
[0082] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0083] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.
[0084] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.
Claims
1. A fresh milk price prediction system based on big data, characterized in that: include: Data collection module, used to collect raw market data; A data processing module, used for preprocessing the original market data to obtain market data; A data analysis module, used to implement an adaptive clustering algorithm on the market data and dynamically adjust the market status; a game theory modeling module, used to construct a game theory model, introduce a mixed strategy Nash equilibrium analysis, and calculate the optimal strategy combination; a price prediction module, used to predict the price of raw milk using the optimal strategy combination, modify the game theory model, and use the modified game theory model to predict the price of raw milk again to obtain a prediction result; And a display module for visually displaying the prediction results; wherein the data acquisition module is connected to the data processing module by means of data transmission, the data processing module is connected to the data analysis module by means of data transmission, the data analysis module is connected to the game theory modeling module by means of data transmission, the game theory modeling module is connected to the price prediction module by means of data transmission, and the price prediction module is connected to the display module by means of data transmission.
2. The fresh milk price prediction system based on big data according to claim 1, characterized in that: The adaptive clustering algorithm in the data analysis module divides the market data into two or more stages, each stage representing a different market state and price fluctuation pattern; wherein the market state is represented by the number of clusters and cluster centers, and the price fluctuation pattern is represented by the intra-cluster variance and the inter-cluster variance.
3. The fresh milk price prediction system based on big data according to claim 2, characterized in that: The implementation process of the adaptive clustering algorithm includes: the first step is to set the initial number of clusters and the initial cluster center; the second step is to dynamically adjust the cluster center; the third step is to calculate the intra-cluster variance and the inter-cluster variance; the fourth step is to dynamically adjust the number of clusters by comparing the ratio of the intra-cluster variance to the inter-cluster variance.
4. The fresh milk price prediction system based on big data according to claim 3 is characterized in that: The process of dynamically adjusting the cluster center includes calculating and updating the cluster center of each cluster according to the data points belonging to each cluster in the current iteration; and optimizing the adaptive clustering result through the dynamically adjusted cluster number and cluster center.
5. The raw milk price prediction system based on big data according to claim 1, characterized in that: The game theory model constructed by the game theory modeling module includes defining the producer's profit function.
6. The fresh milk price prediction system based on big data according to claim 5, characterized in that: The mixed strategy Nash equilibrium analysis introduced in the game theory modeling module includes calculating the optimal mixed strategy of the producer, which is based on the producer's profit function and the strategy combination of other participants; based on the optimal mixed strategy, determining the optimal mixed strategy selection of the participants, and obtaining the optimal strategy combination in each market stage.
7. The raw milk price prediction system based on big data according to claim 1, characterized in that: The process of the price prediction module predicting the price of raw milk using the optimal strategy combination includes considering wholesale price, retail price, market demand and market benchmark demand.
8. The raw milk price prediction system based on big data according to claim 1, characterized in that: The process of the price prediction module correcting the game theory model includes verifying the predicted raw milk price using historical market data and optimizing the game theory model by adjusting fitting parameters.
9. A method for predicting raw milk prices based on big data, characterized in that: The following steps are involved: S1: pre-processing the collected original market data to obtain market data; implementing an adaptive clustering algorithm on the market data to dynamically adjust the market status; S2: Construct a game theory model and introduce mixed strategy Nash equilibrium analysis to calculate the optimal strategy combination; S3: using the optimal strategy combination to predict the price of raw milk; S4: modifying the game theory model, and using the modified game theory model to predict the price of raw milk to obtain a prediction result; S5: Visualize the prediction results.
10. The method for predicting raw milk price based on big data according to claim 1, characterized in that: The S2 specifically includes: Through mixed strategy Nash equilibrium analysis, the optimal mixed strategy of participants under different strategy combinations is calculated; based on the optimal mixed strategy, the optimal mixed strategy selection of participants is determined to obtain the optimal strategy combination.
