Real-time data acquisition and analysis method and system for agricultural product digital transaction

Abstract

The invention discloses a real-time data acquisition and analysis method and a real-time data acquisition and analysis system for agricultural product digital transaction, and relates to the technical field of product transaction management. And a historical data set including key indexes such as price fluctuation, market demands, supply conditions and environmental factors is formed. The data are analyzed through the system, a data model is trained, the model can evaluate the advantages and disadvantages of transactions and potential risks according to the input data, and decision support is provided for investors and transactions in all stages. And then, the new data set is input into the model, the model rapidly outputs a transaction risk result of the new data set based on a historical learning result, risk levels are divided according to a preset threshold, and managers are helped to timely respond to market changes. And finally, through analogue simulation, a user can experience risk trends in different scenes, investment and operation strategies are optimized, and transaction security and efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of product transaction management, and in particular to a real-time data collection and analysis method and system for digital transactions of agricultural products. Background Art

[0002] In recent years, digital transactions have become increasingly important in the agricultural products market. Real-time data collection and analysis systems play a key role in improving transaction efficiency and ensuring transaction transparency. However, with the frequent occurrence of natural disasters in agriculture, frequent fluctuations in agricultural product prices, and significantly increased risks in the agricultural trading market, coupled with the wide variety of agricultural products and the real-time updating of agricultural product trading market conditions, traditional real-time data collection and analysis systems for digital transactions of agricultural products are difficult to obtain risk assessments based on transaction analysis data of agricultural products, in order to assess the risk level of the transaction advantages and disadvantages of individual agricultural product varieties, and to ensure the return of investment funds by avoiding transaction risks.

[0003] To sum up, how to conduct risk assessment on digital transactions of agricultural products to avoid investment risks is an urgent problem to be solved and optimized in the real-time data collection and analysis system of digital transactions of agricultural products. Summary of the invention

[0004] The present invention provides a real-time data collection and analysis method and system for digital transactions of agricultural products, which solves the technical problem of how to conduct risk assessment on digital transactions of agricultural products in related technologies to avoid investment risks.

[0005] In order to solve the above technical problems, the present invention provides a real-time data collection and analysis method and system for digital transactions of agricultural products. The specific technical solutions are as follows:

[0006] In a first aspect, a real-time data collection and analysis method for digital agricultural product transactions comprises the following steps:

[0007] Collect digital transaction information data of agricultural products to obtain a data set; train the historical data set to obtain a data analysis model to output the risk standards of agricultural product transactions;

[0008] Input the newly acquired data set into the data analysis model; based on the data analysis model, receive the newly acquired data set to output the transaction pros and cons risk results of the newly acquired data set, and set a preset threshold to obtain the transaction risk level of the data set;

[0009] Based on the transaction risk level, a simulated agricultural product trading scenario is designed to provide an immersive multi-period agricultural product trading risk experience. Based on the multi-period agricultural product trading risk experience, the risk assessment trend of corresponding agricultural product transactions in different periods is obtained. Based on the assessment results, the investment risks of agricultural product trading are avoided.

[0010] As a further optimization scheme of the present invention, the historical data set is trained to obtain a data analysis model to output a standard index of agricultural product transaction risk, including:

[0011] Collect agricultural product transaction information data through Internet of Things technology to obtain a data set; the data set includes agricultural disaster situation data, various agricultural product transaction quality data, market transaction data and supply chain data;

[0012] The data set is normalized by: ; Convert the digital transaction information data of various agricultural products into a unified range of [0, 1] to obtain the pre-processed historical data set;

[0013] Where D i is the data value of the i-th hour; min(D train ) is the minimum value of the training data; max(D train ) is the maximum value of the training data; D i,norm is the data value at the i-th moment after normalization;

[0014] The preprocessed historical data set is used to construct a training set to train a data analysis model.

[0015] As a further optimization scheme of the present invention, the preprocessed historical data set is used to construct a training set to train a data analysis model, including:

[0016] When the data set input into the data analysis model contains variables that are irrelevant to agricultural product transaction information data, by: ; Extract necessary features from agricultural product trading information data and convert them into high-quality training data;

[0017] Where r represents the Pearson correlation coefficient between vector x and vector y; , Represents the average value of vectors x and y, respectively. i ,y i They represent the values ​​of vectors x and y at the i-th moment respectively, n represents the total number of extracted features, and the value range of r is [-1, 1].

[0018] As a further optimization solution of the present invention, the data analysis model includes:

[0019] The historical data set is used as the original input data of the data analysis model to train the model generalization; based on the preprocessed data set, the data set is encoded into a feature vector G, and the feature vector G is input into the data analysis model. The leakage prediction model includes an update gate, a forget gate, a hidden layer and an output layer. The intermediate representation data of the hidden layer is transmitted to the output layer, and the output layer output represents the historical agricultural product digital transaction quality risk results, as follows:

[0020]

[0021] In the formula, z t is the update gate vector; r t is the forget gate vector; h t is the hidden layer state vector of the current time step t; c t is the candidate hidden layer state vector of the current time step t; W z , W r , W c are the parameter matrices of each gate vector; U z , U r , U c are the parameter vectors of each gate vector; b z 、b r 、b c are the offset vectors of each gate vector respectively; σ represents the sigmoid activation function; ▪ represents the dot product between vector elements; tanh represents the hyperbolic tangent activation function.

[0022] As a further optimization solution of the present invention, the newly acquired data set is input into a data analysis model; based on the data analysis model, the newly acquired data set is received to output a risk index of the newly acquired data set, including:

[0023] The newly acquired data set is encoded into a feature vector H, and the feature vector H is input into a data analysis model to output a result of predicting the quality risk of the newly acquired agricultural product digital transaction information data. The output layer:

[0024]

[0025] Where y represents the output of the hidden layer and represents the nth intermediate representation data, n represents the total number of data items in the input data set, and W y is the output weight parameter, by is the output bias parameter, and σ represents the sigmoid function.

[0026] As a further optimization scheme of the present invention, a preset threshold is set to obtain the transaction risk level, including:

[0027] Based on the historical data set, a transaction data risk factor is obtained; the transaction data risk factor includes agricultural product market fluctuation data; supply chain impact data; policy change data, credit information data of both parties to the transaction, and agricultural disaster risk data;

[0028] According to the transaction data risk factor, by: ; To obtain the risk threshold, where RI is the risk threshold, V is the market volatility, S is the supply chain impact parameter, P is the policy risk parameter, C is the credit risk parameter, I is the agricultural disaster risk parameter, and w1, w2, w3, w4, w5 are weight coefficients;

[0029] Based on the risk threshold, the newly acquired risk result of the data set output by the data analysis model is graded, and the grade classification includes two risk levels: excellent and poor.

[0030] As a further optimization scheme of the present invention, a simulated transaction scenario is designed based on the risk index to evaluate the transaction risk level, including:

[0031] Based on the risk results of agricultural product transactions output by the data analysis model, design immersive transaction experience scenarios for multiple agricultural products in multiple periods;

[0032] The various agricultural products include fruits and vegetables, livestock, poultry and grains;

[0033] The immersive trading experience scenarios at each stage include when agricultural products are growing, maturing, processing, transporting, trading and using;

[0034] Based on immersive agricultural product trading experience scenarios in multiple periods, users can conduct trading experiences in different periods and obtain corresponding risk level assessment trends.

[0035] As a further optimization scheme of the present invention, based on immersive agricultural product trading experience scenarios in multiple periods, users can experience trading in different periods and obtain corresponding risk assessment trends, including:

[0036] Acquire scenario features at multiple stages and generate a complete scenario experience for agricultural product transactions, so that users can fully participate in the transaction decision-making at the full cycle of agricultural products and obtain detailed feedback information on transactions at each agricultural product stage;

[0037] The scene features of the multiple stages include the transaction experience scene features of the agricultural product growth stage, the transaction experience scene features of the agricultural product maturity stage; the transaction experience scene features of the agricultural product processing stage; the transaction experience scene features of the agricultural product transportation stage; the transaction experience scene features of the agricultural product trading stage; and the transaction experience scene features of the agricultural product use stage.

[0038] Based on the transaction data risk factors, the experience scenario continuously monitors and dynamically updates the risk dynamics of each period, and generates corresponding risk level assessment results, so that users can understand the direction of risk dynamic changes in agricultural product transactions at multiple stages;

[0039] The risk trend is obtained based on the dynamic changes in risks during agricultural product transactions, and the final risk assessment report is generated based on the transaction experience and risk assessment results of all stages to support users in comprehensively analyzing market dynamics at each stage.

[0040] As a further optimization scheme of the present invention, based on the evaluation results, in order to avoid the investment risks of agricultural product trading, the following are included:

[0041] Based on the good and bad risk results of agricultural product transactions output by the data analysis model, when agricultural product transactions are at a low risk level, users can obtain the transaction status of the same agricultural product in different regions or different seasons through immersive agricultural product transaction experience scenarios;

[0042] Based on the trading situation of the same agricultural products, the best trading period of the same agricultural products in each region is obtained to obtain reference data of agricultural product trading information;

[0043] Based on the agricultural product transaction information reference data, it provides a reference for users to accurately invest funds when conducting agricultural product transactions.

[0044] In a second aspect, the system provides an electronic device including a memory, a processor, and a real-time data collection and analysis method program for digital transactions of agricultural products stored in the memory and executable on the processor. When the real-time data collection and analysis method program for digital transactions of agricultural products is executed by the processor, the steps of the real-time data collection and analysis method for digital transactions of agricultural products are implemented. The system includes:

[0045] Data collection module 201: It is used to collect digital transaction information data of agricultural products to obtain a data set; train the historical data set to obtain a data analysis model to output the risk standard of agricultural product transactions;

[0046] Data analysis module 202: used to input the newly acquired data set into the data analysis model; based on the data analysis model, receive the newly acquired data set to output the transaction risk result of the newly acquired data set, and set a preset threshold to obtain the transaction risk level of the data set;

[0047] Risk assessment module 203: It is used to provide an immersive multi-period agricultural product transaction risk experience by designing and simulating agricultural product transaction scenarios based on the transaction risk level, and to obtain risk assessment trends of agricultural product transactions corresponding to different periods based on the multi-period agricultural product transaction risk experience, and to avoid agricultural product transaction investment risks based on the assessment results.

[0048] The present invention has at least the following beneficial effects: the present invention firstly collects digital information of multiple transaction cycles involving agricultural products to form a historical data set. This data set includes but is not limited to price fluctuations, market demand, supply conditions, environmental factors and other key indicators that affect transactions. By systematically analyzing and processing these historical data, a data analysis model is trained. The model can output the pros and cons of the corresponding agricultural product transactions and the potential risk level results according to the input data set, thereby providing decision support for investors and agricultural product transactions at all stages.

[0049] Subsequently, the newly collected agricultural product transaction data is input into the trained data analysis model. The model processes the new data and quickly outputs the pros and cons of the new data set transactions and the potential risk level results based on historical learning results and algorithm logic. The system automatically identifies and classifies the risk level of transaction data by setting preset risk thresholds, allowing managers and investors to understand market changes in a timely manner and respond quickly.

[0050] Finally, based on the output results of the data analysis model, a simulation environment is designed to provide an immersive multi-period agricultural product trading risk experience scenario. Users can experience every stage of agricultural products from planting to sales to use in a simulated market environment, and evaluate risk trends in different periods and scenarios. Based on these risk assessment results, investors and agricultural product operators can better avoid potential market risks and optimize their investment and operation strategies. This provides participants with a new and scientific risk control method, which greatly improves the security and efficiency of transactions. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a flow chart of a method for real-time data collection and analysis of digital agricultural product transactions provided by an embodiment of the present invention;

[0052] Figure 2 It is a schematic diagram of a real-time data collection and analysis system for digital transactions of agricultural products provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The present application is further described in detail below in conjunction with the accompanying drawings. It is necessary to point out here that the following specific implementation methods are only used to further illustrate the present application and cannot be understood as limiting the scope of protection of the present application. Technical personnel in this field can make some non-essential improvements and adjustments to the present application based on the above application content.

[0054] This embodiment provides a real-time data collection and analysis method and system for digital agricultural product transactions, and the specific implementation methods are as follows:

[0055] like Figure 1 As shown, a real-time data collection and analysis method for digital transactions of agricultural products includes the following steps:

[0056] Step 1: Collect digital transaction information data of agricultural products to obtain a data set; train the historical data set to obtain a data analysis model to output the risk standard of agricultural product transactions;

[0057] Step 2: input the newly acquired data set into the data analysis model; based on the data analysis model, receive the newly acquired data set to output the transaction risk result of the newly acquired data set, and set a preset threshold to obtain the transaction risk level of the data set;

[0058] Step 3, based on the transaction risk level, simulate the agricultural product transaction scenario by designing to provide an immersive multi-period agricultural product transaction risk experience, based on the multi-period agricultural product transaction risk experience, to obtain the risk assessment trend of agricultural product transactions corresponding to different periods, based on the assessment results, to avoid agricultural product transaction investment risks.

[0059] It needs to be specifically explained in this embodiment that the method for obtaining agricultural product information data in this embodiment is to use crawler technology (such as Python's Scrapy, BeautifulSoup, etc.) to crawl data from online agricultural product trading platforms. When the trading platform provides an open API, the API interface is used to request trading data through programming; for platforms that do not have an open API or cannot be obtained through crawlers, personnel can be organized to collect data manually and enter the data into a database or table.

[0060] For crawler solutions, you need to write crawler scripts to define the crawled web page content and rules, and handle anti-crawling mechanisms; for API solutions, write API request scripts to obtain and store data; in the case of manual collection, use tools such as Excel to enter transaction data in a structured manner.

[0061] The collected raw digital agricultural product transaction data is stored in a database (such as MySQL, MongoDB, etc.). When the data volume is small, it can also be stored in CSV or Excel files to facilitate later analysis and enable real-time data interaction.

[0062] In this embodiment, digital transaction information data of agricultural products are collected through various channels (such as e-commerce platforms, trading markets, supply chain systems, etc.) through step 1. These data may include agricultural disaster conditions, prices, transaction volumes, information about buyers and sellers, transaction times, product quality, logistics information, etc. The collected data will be cleaned, deduplicated, and sorted to form a complete set of historical data; these historical data are used to train data analysis models. Training is performed based on patterns in historical data to learn how to assess the risks of different transactions;

[0063] By training the model with a large amount of historical data, it can capture the potential risk patterns in the agricultural product trading process and formulate risk standards for the quality of transactions. This standard can help the system quickly assess risks when facing future transactions; the trained model has the ability to predict transaction risks and improve the accuracy and reliability of decision-making.

[0064] In step 2, the newly collected agricultural product transaction data is input into the trained data analysis model. The model analyzes the new data based on the historical transaction risk standards and generates corresponding transaction risk results. In order to further refine the transaction risk, the system will set a risk threshold. The risk threshold divides the transaction into different risk levels (excellent and inferior). Based on the comparison between the output result and the threshold, the system will give the risk level of the transaction;

[0065] By inputting new data into the trained model, the system can analyze and output the risk status of the transaction in real time, allowing users to have a clearer understanding of the current transaction; the division of risk levels allows users to more intuitively see the risk level of different transactions, so that they can respond to transaction risks more flexibly.

[0066] According to the transaction risk level generated in step 2, a simulated agricultural product transaction scenario is designed through step 3. The simulation scenario will be constructed based on current transaction risks and past data patterns, and may simulate different market environments, price fluctuations, demand changes and other scenarios; through simulated transactions, the system can provide an immersive multi-period transaction experience, allowing users to feel how the risks of agricultural product transactions change at different time points. For example, the system can simulate the impact of fluctuations in agricultural product prices during the harvest season, transportation delays or climate change on transactions; through these immersive experiences, users can observe the corresponding transaction risk assessment trends in different periods, which helps to understand the transaction risks that may be encountered at different time points;

[0067] By simulating real trading scenarios, users can perceive and experience potential trading risks in advance in a low-risk environment; multi-period risk experience helps users understand the trend of agricultural product trading risks over time, so that they can make more informed investment decisions and avoid potential investment risks; the system can provide more accurate risk avoidance strategies based on the evaluation results to ensure that users reduce investment risks and improve investment returns in actual transactions.

[0068] Through the collaborative work of the above steps, it is possible to provide suggestions for digital transactions of agricultural products through multi-source data input and model processing in a complex and changing environment. The entire system improves the understanding of agricultural products in many aspects and the prediction accuracy in the process of processing and analyzing data, thereby reducing the investment risk in digital transactions of agricultural products.

[0069] In a preferred embodiment of the present invention, step 1 may further include the following steps:

[0070] Step 11, collecting agricultural product transaction information data through Internet of Things technology to obtain a data set; the data set includes agricultural disaster situation data, various agricultural product transaction quality data, market transaction data and supply chain data;

[0071] Step 12, normalizing the data set by: ; Convert the digital transaction information data of various agricultural products into a unified range of [0, 1] to obtain the pre-processed historical data set;

[0072] Where D i is the data value of the i-th hour; min(D train ) is the minimum value of the training data; max(D train ) is the maximum value of the training data; D i,norm is the data value at the i-th moment after normalization;

[0073] Step 13, constructing a training set from the preprocessed historical data set to train a data analysis model.

[0074] Specifically, in the embodiments of the present invention, IoT devices (such as smart sensors) are used in various links of agricultural product growth, collection, production, processing, transportation and use to collect data such as temperature, humidity, and transportation conditions in real time. These sensors can be connected to the cloud, and data can be automatically uploaded and stored. In addition, drones or satellite images can be used to monitor farmland, and combined with environmental data collected by sensors, information such as crop growth status and health level can be provided.

[0075] In step 11, the transaction information of agricultural products that need to be collected is clearly defined, such as grain, fruits and vegetables, livestock and other categories. It can be broken down to the variety level, and the time range (such as day, week, month) and geographical range (such as national, provincial, municipal, district, etc.) of the data are determined so that data collection is more targeted.

[0076] Determine the data source, such as agricultural product trading platforms, e-commerce platforms (such as Taobao, Pinduoduo, etc.), agricultural cooperative information systems, or wholesale market data, and determine the type of data to be collected based on the nature of the platform, for example:

[0077] Product name and variety; agricultural climate and disaster situation; transaction price and volume; supply and demand information; logistics information (transportation cost, route, etc.); transaction time and location; basic information of buyer and seller; clear data storage format (such as JSON, CSV, XML, etc.) to facilitate subsequent data processing.

[0078] Collect digital transaction information data of agricultural products through Internet of Things technology. In step 12, check whether the collected data is complete, remove duplicates, null values ​​or abnormal data, and ensure that all data items are in the same format by normalizing the data, such as date format, price unit, product category name, etc. For missing or abnormal data, take reasonable processing methods (such as interpolation, deletion or completion, etc.), merge data from different platforms or sources, and ensure that a unified data set is formed. The cleaned data can be stored in a structured database (such as an SQL database) or an unstructured database (such as a NoSQL database), and the data is backed up regularly to prevent data loss. The data is managed in layers, and access rights are set to ensure data security and privacy.

[0079] Finally, step 13 is used to construct a training set with the preprocessed historical data set, and the data analysis model is trained to predict the transaction risk results of the newly input agricultural product transaction information data in the future, so as to provide users with a reference for digital agricultural product transactions.

[0080] In a preferred embodiment of the present invention, the above step 13 further includes:

[0081] Step 131, when the data set input into the data analysis model contains variables irrelevant to agricultural product transaction information data, by: ; Extract necessary features from agricultural product trading information data and convert them into high-quality training data;

[0082] Where r represents the Pearson correlation coefficient between vector x and vector y; , Represents the average value of vectors x and y, respectively. i ,y iThey represent the values ​​of vectors x and y at the i-th moment respectively, n represents the total number of extracted features, and the value range of r is [-1, 1].

[0083] In the embodiment of the present invention, a data set including various agricultural product transaction information variables and some irrelevant variables (inaccurate data information, the influence of public opinion, malicious competition in the market economy, etc.) is obtained in step 11 through step 131. The irrelevant variables will interfere with the learning and training of the model, thereby providing a basis for subsequent analysis and ensuring that valuable features can be extracted therefrom.

[0084] Pearson correlation coefficient analysis is used to identify variables that are irrelevant to agricultural product trading information. This coefficient measures the linear correlation between two variables. By identifying irrelevant variables, the complexity of the data can be reduced, the noise in model training can be reduced, and the accuracy and efficiency of the model can be improved. Once irrelevant variables are identified, the next step is to extract features that are directly related to agricultural product trading. This can be achieved through methods such as calculating correlations and feature selection algorithms (such as Lasso, decision trees, etc.); after extracting necessary features, a more concise and highly relevant data set can be obtained. This will help the model better learn related patterns and improve prediction performance.

[0085] The extracted features are converted into high-quality training data sets. This usually includes steps such as data cleaning, standardization, and encoding to ensure that the data is suitable for machine learning model training. High-quality training data can significantly improve the generalization ability and prediction accuracy of the model, so that the model can be better applied to actual agricultural product trading predictions.

[0086] The value range of the Pearson correlation coefficient is [-1,1], where 1 indicates a perfect positive correlation, -1 indicates a perfect negative correlation, and 0 indicates no correlation. By calculating these coefficients, the linear relationship between the features can be quantified. This process enables the model to more clearly understand the relationship between the features, thereby better capturing the key factors affecting agricultural product transactions.

[0087] In a preferred embodiment of the present invention, the data analysis model includes:

[0088] The historical data set is used as the original input data of the data analysis model to train the model generalization; based on the preprocessed data set, the data set is encoded into a feature vector G, and the feature vector G is input into the data analysis model. The leakage prediction model includes an update gate, a forget gate, a hidden layer and an output layer. The intermediate representation data of the hidden layer is transmitted to the output layer, and the output layer output represents the historical agricultural product digital transaction quality risk results, as follows:

[0089]

[0090] In the formula, z t is the update gate vector; r t is the forget gate vector; h t is the hidden layer state vector of the current time step t; c t is the candidate hidden layer state vector of the current time step t; W z , W r , W c are the parameter matrices of each gate vector; U z , U r , U c are the parameter vectors of each gate vector; b z 、b r 、b c are the offset vectors of each gate vector respectively; σ represents the sigmoid activation function; ▪ represents the dot product between vector elements; tanh represents the hyperbolic tangent activation function.

[0091] In this embodiment, historical data sets are used as the original input of the model, which usually include transaction information of agricultural products, time series data, etc. By analyzing these data, the model can learn historical transaction patterns, thereby ensuring that the model can learn in a diverse historical context, thereby improving generalization capabilities and enabling it to better adapt to future data.

[0092] Preprocess the raw data to generate a feature vector G. This step may involve standardization, missing value processing, feature selection, etc. The feature vector will be used as the input of the model. By encoding the feature vector, the model can use the input data in a more effective way, extract key features, and enhance the expressiveness of the model.

[0093] Working principle: The model consists of an update gate, a forget gate, a hidden layer, and an output layer. The update gate and the forget gate are used to control the inflow and outflow of information, helping the model retain important information and discard useless information. This structure can handle the long-term dependencies of time series data, making the model perform better when processing complex time series information.

[0094] The hidden layer state at the current time step t is calculated by the update gate and the forget gate. The candidate hidden layer state is generated by the hyperbolic tangent activation function (tanh) and is used to further adjust the hidden layer state. This state update mechanism enables the model to capture the dynamic changes of historical information and improve the accuracy and flexibility of prediction.

[0095] The intermediate representation data of the hidden layer is transmitted to the output layer, which generates prediction results about the historical agricultural product trading risks based on the state of the hidden layer. This process combines the learning ability of the model with activation functions (such as sigmoid) to produce the final result. By converting historical information into specific risk assessments, the output layer can provide practical decision support and provide a basis for adjusting the strategy of agricultural product trading.

[0096] Nonlinear features are introduced through σ and tanh activation functions, and the use of parameters and activation functions is optimized, so that the model can better fit the data and enhance the expressiveness and predictive ability of the model.

[0097] In a preferred embodiment of the present invention, the above step 2 further includes:

[0098] Step 21, encode the newly acquired data set into a feature vector H, input the feature vector H into the data analysis model to output the predicted quality risk results of the newly acquired agricultural product digital transaction information data, the output layer:

[0099]

[0100] Where y represents the output of the hidden layer and represents the nth intermediate representation data, n represents the total number of data items in the input data set, and W y is the output weight parameter, by is the output bias parameter, and σ represents the sigmoid function.

[0101] In this embodiment, the newly acquired data set is encoded into a feature vector H. This step usually includes data cleaning, normalization, feature selection, etc. to ensure the quality and relevance of the input data. High-quality feature vectors can improve the model's ability to understand new data, enabling it to perform risk assessment more accurately.

[0102] The feature vector H is input into the data analysis model. The model uses the previously learned parameters and structures to process the input features. The model can quickly identify the hidden information in the feature vector and make predictions based on the patterns in the historical data, thus improving the prediction efficiency.

[0103] In the output layer, the output of the hidden layer represents that the output layer converts the features of the hidden layer into specific risk assessment values, and the sigmoid function provides a probability output between 0 and 1, making the results easy to interpret and facilitate decision-making.

[0104] The parameters learned by the model during training determine how the hidden layer information is mapped to the output results. The weight parameters control the impact of the features on the output, while the bias parameters adjust the baseline of the output. By optimizing these parameters, the model can better fit new data and enhance the accuracy and stability of the output.

[0105] The final output y represents the prediction of the advantages and disadvantages of the newly acquired agricultural product trading information. Based on the risk prediction, users can adopt corresponding strategies for management and decision-making. This risk prediction ability helps agricultural product traders identify potential market opportunities and risks, thereby optimizing trading strategies, reducing losses and increasing profits.

[0106] In a preferred embodiment of the present invention, the above step 2 may further include:

[0107] Step 22, based on the historical data set, to obtain transaction data risk factors; the transaction data risk factors include agricultural product market fluctuation data; supply chain impact data; policy change data, transaction parties' credit information data and agricultural disaster risk data;

[0108] Step 23, according to the transaction data risk factor, by: ; To obtain the risk threshold, where RI is the risk threshold, V is the market volatility, S is the supply chain impact parameter, P is the policy risk parameter, C is the credit risk parameter, I is the agricultural disaster risk parameter, and w1, w2, w3, w4, w5 are weight coefficients;

[0109] Step 24, based on the risk threshold, the newly acquired risk result of the data set output by the data analysis model is graded, and the grade classification includes two risk levels: excellent and poor.

[0110] In this embodiment, by analyzing historical data, key risk factors affecting agricultural product transactions are extracted, including:

[0111] Market volatility data (V): describes the volatility of agricultural product market prices or demand.

[0112] Supply Chain Impact Data(S): Assess risks in the supply chain that could cause transaction disruptions or delays.

[0113] Policy change data (P): reflects the potential impact of government policy changes on agricultural product transactions.

[0114] Credit information data of both parties to the transaction (C): Evaluate the historical credit records of both parties to the transaction and predict future default or performance risks.

[0115] Agricultural disaster risk data (I): Based on the impact of historical natural disasters or climate events on agricultural production.

[0116] The above include but are not limited to:

[0117] Market volatility: Agricultural product prices are affected by changes in supply and demand, climate conditions, and seasonal fluctuations. Volatility is an important indicator for measuring price risk.

[0118] Supply chain risks: Events that affect supply, such as transportation delays, disasters at the origin, and epidemics, increase risks.

[0119] Policy risks: Changes in government policies, trade restrictions, and subsidy policies have a significant impact on agricultural product transactions.

[0120] Credit risk: credit issues between buyers and sellers, payment defaults or delivery failures, etc.

[0121] International market risks: international market price fluctuations, exchange rate changes, tariffs and other changes in international trade conditions.

[0122] Through in-depth analysis of historical data, we can extract multiple risk factors that are crucial to agricultural product trading, laying a solid data foundation for subsequent risk forecasting.

[0123] In step 23, based on the risk factors extracted in step 22, a comprehensive risk threshold is obtained. The weights of different risk factors represent their relative importance in the total risk. Through the comprehensive calculation of each risk factor, a quantifiable risk threshold RI is generated. This threshold helps to quickly identify potential risks in transactions and help decision makers take appropriate countermeasures;

[0124] Step 24: Based on the calculated RI, the risk results of the new data set are graded. A common way of grading is to categorize the risk results into two grades: "excellent" and "poor":

[0125] Excellent grade: indicates that the risk of agricultural product transactions is low and market conditions are good.

[0126] Poor grade: indicates that agricultural product trading involves greater risks and may require hedging measures or cautious trading or investment.

[0127] The classification of risk levels can provide users with clear guidance, helping them quickly identify low-risk trading opportunities and avoid high-risk transactions, thereby optimizing management decisions and reducing possible losses.

[0128] In a preferred embodiment of the present invention, the above step 3 may include:

[0129] Step 31, based on the agricultural product transaction quality and risk results output by the data analysis model, designing immersive transaction experience scenarios for multiple agricultural products in multiple periods;

[0130] The various agricultural products include fruits and vegetables, livestock, poultry and grains;

[0131] The immersive trading experience scenarios at each stage include when agricultural products are growing, maturing, processing, transporting, trading and using;

[0132] Step 32, based on the immersive agricultural product trading experience scenarios in multiple periods, the user can experience trading in different periods and obtain the corresponding risk level assessment trend;

[0133] Step 33, based on the agricultural product transaction risk results output by the data analysis model, when the agricultural product transaction is at a low risk level, the user experiences an immersive agricultural product transaction scenario to obtain the transaction status of the same agricultural product in different regions or different seasons;

[0134] Step 34, based on the trading situation of the same agricultural products, the best trading period of the same agricultural products in each region is obtained to obtain reference data of agricultural product trading information;

[0135] Step 35, based on the agricultural product transaction information reference data, to provide a reference for users to accurately invest funds when conducting agricultural product transactions.

[0136] In this embodiment, in step 31, based on the risk results of agricultural product transactions output by the data analysis model, immersive trading experience scenarios for multiple agricultural products in multiple periods are designed. The types of agricultural products covered include fruits and vegetables, livestock, poultry, and grain, ensuring that users can be exposed to a rich variety of market scenarios. At the same time, for each agricultural product, experience scenarios are designed for multiple periods including growth, maturity, processing, transportation, trading, and use. This multi-dimensional scenario setting helps users fully understand the market risks and their changes at different stages, laying the foundation for subsequent trading decisions.

[0137] Step 32: In the designed immersive trading experience scenario, users can experience simulated trading at different times. These experiences allow users to gain an intuitive understanding of market fluctuations and obtain corresponding risk level assessment trends in real time. This process not only increases the user's sense of participation, but also enables them to have a deeper understanding of the dynamic changes in market risks through actual combat exercises, helping them to better formulate trading strategies.

[0138] Specifically, the trading risks of agricultural products during the growth period are closely related to natural conditions, agronomic measures, and market supply and demand expectations, such as:

[0139] Unpredictable weather, such as drought, flood, frost, etc., directly affects the growth of crops, leading to yield fluctuations, which in turn affect market expectations and price fluctuations;

[0140] Crops are susceptible to various diseases and pests, especially during the growing season, which, if not properly controlled, may lead to large-scale yield reduction or quality decline.

[0141] The increase in production costs such as seeds, fertilizers, and labor may increase farmers' production costs and affect the final profit margin.

[0142] Changes in government agricultural policies (such as planting subsidies and price protection policies) may lead to changes in farmers' expectations of market supply and demand, thereby affecting transactions.

[0143] Trading risks during the maturity period. Trading risks during the maturity period of agricultural products are mainly related to the harvest time and quality standards. Due to external factors such as weather, the harvest period of agricultural products may be delayed or advanced. This uncertainty may affect the timing of market supply and lead to an imbalance between supply and demand.

[0144] During the maturity period, the quality of the product will directly affect its market value. If there is abnormal weather, the product may not meet the expected quality standards, which will in turn affect its trading price.

[0145] The supply at maturity is usually concentrated and large, and this short-term supply surge may cause sharp fluctuations in market prices, especially in cases of oversupply.

[0146] Transaction risks during the processing period are an important stage in the value-added of agricultural products, but they are also accompanied by some unique risks. If the processing is not handled properly, it may lead to a decline in product quality, such as hygiene and safety issues or loss of nutrients, affecting market acceptance and prices.

[0147] Failure of processing equipment or limitations in technology may result in limited production or substandard quality, affecting delivery time and product market performance.

[0148] Processing companies rely on the timely supply of raw materials. If there is a disruption in the supply chain or an increase in the cost of raw materials, it may affect the overall cost of processing and the price of the final product.

[0149] Transaction risks during the transportation period. In the process of transporting agricultural products from the place of production to the market, transportation risks may significantly affect the transaction results. Too long transportation time may cause fresh agricultural products to spoil or deteriorate in quality, especially for perishable fruits and vegetables. The efficiency of logistics is directly related to the market value of the products.

[0150] During transportation, agricultural products may suffer physical damage such as bumps, squeezing or temperature changes (such as cold chain transportation failure), which can lead to product loss or depreciation.

[0151] Differences in traffic conditions and road infrastructure may also lead to uncontrollable delays in transportation or increase transportation costs.

[0152] Trading risks during the trading period. When actually buying and selling in the market, the risks brought by the market environment and external economic factors are particularly prominent. Market price fluctuation risk: Fluctuations in market demand and supply may lead to significant changes in prices. If market demand suddenly drops or supply increases, the price of agricultural products may fall sharply, affecting trading profits.

[0153] Trade barriers, import taxes or other market access requirements that may be implemented by different countries or regions may affect the circulation and prices of agricultural products.

[0154] Agricultural products may not be able to find buyers quickly or be sold through appropriate channels, which will lead to unsalable goods or inventory backlogs, thereby increasing transaction costs and losses.

[0155] Transaction risks during the use period: When agricultural products are finally used or consumed by consumers, there are still some risks, especially those related to food safety and market acceptance. Consumer preferences may change over time, such as health trends, food preferences, etc. If a certain type of agricultural product suddenly becomes unpopular, it may cause a sharp drop in demand and affect market transactions.

[0156] If food safety issues (such as contamination or quality problems) arise during the use stage of agricultural products, it will lead to product recall or damage to market trust, which will in turn affect its market value and subsequent sales.

[0157] When a product is not used up within the shelf life in the hands of consumers, it may affect subsequent purchasing decisions and indirectly affect transaction demand.

[0158] When agricultural product transactions are at a low risk level through step 33, users can obtain the transaction status of the same agricultural product in different regions or seasons through immersive agricultural product transaction experience scenarios. The acquisition of such information enables users to conduct actual combat drills under lower risk conditions, thereby enhancing their sensitivity and adaptability to the market.

[0159] Then, based on the transaction information of the same agricultural product obtained by the user, the best transaction period of the same agricultural product in each region is analyzed in step 34. By comparing the transaction data of different regions and seasons, the user can identify the most favorable transaction window, which not only improves the transaction efficiency, but also provides the user with a more accurate decision-making basis.

[0160] Finally, based on the agricultural product transaction information reference data, users can invest funds more accurately when trading agricultural products in step 35. The successful implementation of this link depends on the data accumulation and analysis of the previous steps, so that users can make more informed investment decisions in a data-driven manner when facing a complex market environment, thereby significantly reducing transaction risks and improving investment returns.

[0161] In a preferred embodiment of the present invention, the above step 32 may further include:

[0162] Step 321, obtaining scene features of multiple stages, generating a complete scene experience of agricultural product transactions, so that users can fully participate in the transaction decision-making of agricultural products in the whole cycle stage, and obtain detailed feedback information on transactions in each agricultural product stage;

[0163] Step 322, the scene features of the multiple stages include the transaction experience scene features of the agricultural product growth stage, the transaction experience scene features of the agricultural product maturity stage; the transaction experience scene features of the agricultural product processing stage; the transaction experience scene features of the agricultural product transportation stage; the transaction experience scene features of the agricultural product transaction stage; and the transaction experience scene features of the agricultural product use stage.

[0164] Step 323, based on the transaction data risk factors, the experience scenario continuously monitors and dynamically updates the risk dynamics of each period, and generates corresponding risk level assessment results, so that users can understand the direction of risk dynamic changes in the agricultural product trading process at multiple stages;

[0165] Step 324, based on the direction of risk dynamic changes in the agricultural product trading process, the risk trend is obtained, and the final risk assessment report is generated in combination with the trading experience and risk assessment results of all stages to support users in comprehensively analyzing the market dynamics of each stage.

[0166] In this embodiment, step 321 aims to obtain the scene characteristics of agricultural products at different stages in their life cycle through simulation or real data. By establishing a virtual scene experience or collecting real transaction data at each stage, users can experience the transaction process at different stages in the first person or in an interactive manner. The characteristics of each stage may include market prices, demand fluctuations, weather impacts, logistics conditions, etc.

[0167] Through immersive scene experience, users can intuitively perceive the specific characteristics and potential risks of each transaction stage, helping them better understand the transaction chain and its complexity of agricultural products from growth to final consumption. This will encourage users to fully participate in and master the transaction decisions of the entire cycle of agricultural products.

[0168] Step 322 refines the definition of multiple scenario features, including the growth, maturity, processing, transportation, trading and use stages of agricultural products. The experience scenario for each stage will be modeled based on actual transaction data and market changes, for example:

[0169] Growth stage: pay attention to climate change, production costs, pests and diseases, etc.

[0170] Maturity stage: involves harvest time, yield fluctuations, etc.

[0171] Processing stage: processing equipment, quality control, supply chain management, etc.

[0172] Transportation stage: logistics time, temperature control, transportation loss, etc.

[0173] Trading stage: market supply and demand, price fluctuations, policy impact, etc.

[0174] Usage stage: consumer preference, food safety, shelf life, etc.

[0175] Through these specific scenario experiences, users can perceive the risks and challenges at each stage in detail, enhance their understanding of complex transaction scenarios within the life cycle of agricultural products, and thus make more informed decisions.

[0176] Step 323 will continuously monitor and collect transaction data in different scenarios, and conduct risk assessment based on preset risk factors (such as market fluctuations, policy changes, weather impacts, etc.). As the trading environment changes, the risk dynamics in the scenario will also be updated in real time to generate risk level assessment results.

[0177] Users can keep abreast of the dynamic changes in risks at different stages of transactions. Dynamically updated risk assessments allow users to see potential market risk trends, allowing them to adjust trading strategies more flexibly to avoid or mitigate the negative impact of risks.

[0178] Step 324 integrates the risk dynamics of each stage, combines the previously experienced scenarios and real-time risk assessment data, and generates a final risk assessment report. This report will include the risk level, trend analysis, and market dynamics forecasts of each stage. Users can use this report to conduct a comprehensive market risk analysis and make corresponding trading decisions;

[0179] The final risk assessment report will help users make decisions after comprehensively considering the risks at all stages of the transaction. The core of this step is to summarize all relevant information to help users foresee future market trends, thereby optimizing their agricultural product trading strategies to maximize profits or minimize potential losses.

[0180] To further explain, the investment scale is determined according to the risk level:

[0181] The inferior risks are further divided into:

[0182] Low risk: When the assessment results show that the risk level is low, it means that the agricultural product transaction is relatively safe. You can consider increasing the investment scale to obtain greater potential returns. In this case, although the risk is low, you still need to pay close attention to market dynamics and external factors.

[0183] Medium risk: Medium risk means that the market has a certain degree of volatility. You can choose to diversify your investments, reduce your investment in a single product, or use hedging strategies to reduce potential losses. In such an environment, a sound investment strategy is more important.

[0184] High risk: High-risk transactions are usually accompanied by greater uncertainty. If the assessment shows that the risk level is high, you can choose to reduce the scale of investment or even temporarily not participate in the transaction. At the same time, you can also consider looking for other more stable investment opportunities.

[0185] According to the above risk assessment results, set clear stop loss and take profit points. If the risk is high, you can set a lower stop loss point to reduce potential losses. On the contrary, if the risk is low, you can allow more fluctuation space to obtain higher returns.

[0186] Stop loss point: When market volatility reaches the assessed risk threshold, you should exit in time to avoid greater losses.

[0187] Take-profit point: When market returns reach expectations, lock in profits in time to prevent subsequent market fluctuations from leading to profit taking.

[0188] By evaluating risk trends, you can dynamically adjust your investment portfolio to diversify risk. For example:

[0189] During times of high risk, you can move your funds into low-risk agricultural commodities or other stable assets, such as gold or Treasury bonds.

[0190] When market risk is low or on an upward trend, you can increase your investment in the agricultural product market to obtain higher returns.

[0191] The risk scenarios you experience in immersive simulations can help you better cope with unexpected situations in actual transactions. For example, by simulating different market environment changes (such as price fluctuations, supply chain disruptions, climate change, etc.), you can prepare response strategies in advance to avoid risks that you may encounter in real transactions.

[0192] By analyzing the risk assessment trends in different time periods, you can predict future market changes and help you make decisions to adjust your investment strategy in advance. For example, if the risk assessment shows that the risk of a certain agricultural product will gradually increase in the next few months, you can choose to sell the relevant assets in advance or reduce investment to avoid losses in market turmoil.

[0193] If the risk assessment shows that a certain agricultural product transaction has a higher risk, but you still want to participate in the market, you can adopt a hedging strategy to mitigate losses. For example, while conducting high-risk transactions, you can invest in assets that are negatively correlated with the product to offset possible losses.

[0194] Based on risk assessment, you can categorize investments in different agricultural products into short-term and long-term strategies:

[0195] Short-term strategy: If a particular agricultural product transaction has a higher risk in the short term, you can adopt a more conservative investment and wait for the market to stabilize before increasing your investment.

[0196] Long-term strategy: If risk assessment shows that a certain agricultural product has good return potential in the long term, you can consider holding it for the long term, even if there are market fluctuations in the short term.

[0197] In summary, by using the risk assessment results, you can make more flexible and rational investment decisions in different trading environments, maximize returns and effectively avoid risks. Specific steps include adjusting the investment scale according to the risk level, setting stop loss and take profit points, dynamically adjusting the investment portfolio, and formulating long-term and short-term response strategies through simulation experience and risk trends.

[0198] like Figure 2 As shown, a real-time data collection and analysis system for digital transactions of agricultural products includes:

[0199] Data collection module 201: It is used to collect digital transaction information data of agricultural products to obtain a data set; train the historical data set to obtain a data analysis model to output the risk standard of agricultural product transactions;

[0200] Data analysis module 202: used to input the newly acquired data set into the data analysis model; based on the data analysis model, receive the newly acquired data set to output the transaction risk result of the newly acquired data set, and set a preset threshold to obtain the transaction risk level of the data set;

[0201] Risk assessment module 203: It is used to provide an immersive multi-period agricultural product transaction risk experience by designing and simulating agricultural product transaction scenarios based on the transaction risk level, and to obtain risk assessment trends of agricultural product transactions corresponding to different periods based on the multi-period agricultural product transaction risk experience, and to avoid agricultural product transaction investment risks based on the assessment results.

[0202] When the functions of the above modules are implemented in the form of software functional units and used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks or optical disks.

[0203] Therefore, the purpose of the present invention can also be achieved by running a program or a group of programs on any computing device. The computing device can be a well-known general intelligent device. Therefore, the purpose of the present invention can also be achieved by simply providing a program product containing a program code that implements the method or system. That is to say, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any known storage medium or any storage medium developed in the future. It should also be pointed out that in the device and method of the present invention, it is obvious that each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. In addition, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but it is not necessary to perform them in chronological order. Some steps can be performed in parallel or independently of each other.

Claims

1. A real-time data collection and analysis method for digital transactions of agricultural products, characterized in that: The following steps are involved: Collect digital transaction information data of agricultural products to obtain a data set; train the historical data set to obtain a data analysis model to output the risk standards of agricultural product transactions; Inputting the newly acquired data set into a data analysis model; Based on the data analysis model, receiving the newly acquired data set, outputting the transaction pros and cons risk result of the newly acquired data set, and setting a preset threshold to obtain the transaction risk level of the data set; Based on the transaction risk level, a simulated agricultural product trading scenario is designed to provide an immersive multi-period agricultural product trading risk experience. Based on the agricultural product trading risk experience, the risk assessment trend of corresponding agricultural product transactions in different periods is obtained. Based on the assessment results, the investment risks of agricultural product trading are avoided.

2. The real-time data collection and analysis method for digital transaction of agricultural products according to claim 1 is characterized in that: The historical data set is trained to obtain a data analysis model to output a standard index of agricultural product transaction risk, including: Collect agricultural product transaction information data through Internet of Things technology to obtain a data set; the data set includes agricultural disaster situation data, various agricultural product transaction quality data, market transaction data and supply chain data; The data set is normalized by: ; Convert the digital transaction information data of various agricultural products into a unified range of [0, 1] to obtain the pre-processed historical data set; Where D i is the data value of the i-th hour; min(D train ) is the minimum value of the training data; max(D train ) is the maximum value of the training data; D i,norm is the data value at the i-th moment after normalization; The preprocessed historical data set is used to construct a training set to train a data analysis model.

3. The real-time data collection and analysis method for digital transaction of agricultural products according to claim 2 is characterized in that: The preprocessed historical data set is used to construct a training set to train a data analysis model, including: When the data set input into the data analysis model contains variables that are irrelevant to agricultural product transaction information data, by: ; Extract necessary features from agricultural product trading information data and convert them into high-quality training data; Where r represents the Pearson correlation coefficient between vector x and vector y; , Represents the average value of vectors x and y, respectively. i ,y i They represent the values ​​of vectors x and y at the i-th moment respectively, n represents the total number of extracted features, and the value range of r is [-1, 1].

4. The real-time data collection and analysis method for digital transaction of agricultural products according to claim 3 is characterized in that: The data analysis model comprises: The historical data set is used as the original input data of the data analysis model to train the model generalization; based on the preprocessed data set, the data set is encoded into a feature vector G, and the feature vector G is input into the data analysis model. The leakage prediction model includes an update gate, a forget gate, a hidden layer and an output layer. The intermediate representation data of the hidden layer is transmitted to the output layer, and the output layer output represents the historical agricultural product digital transaction quality risk results, as follows: ； In the formula, z t is the update gate vector; r t is the forget gate vector; h t is the hidden layer state vector of the current time step t; c t is the candidate hidden layer state vector of the current time step t; W z , W r , W c are the parameter matrices of each gate vector; U z , U r , U c are the parameter vectors of each gate vector; b z , b r , b c are the offset vectors of each gate vector respectively; σ represents the sigmoid activation function; ▪ represents the dot product between vector elements; tanh represents the hyperbolic tangent activation function.

5. The real-time data collection and analysis method for digital transaction of agricultural products according to claim 4 is characterized in that: Inputting the newly acquired data set into a data analysis model; Based on the data analysis model, receiving the newly acquired data set to output a risk index of the newly acquired data set includes: The newly acquired data set is encoded into a feature vector H, and the feature vector H is input into a data analysis model to output a result of predicting the quality risk of the newly acquired agricultural product digital transaction information data. The output layer: ； Where y represents the output of the hidden layer and represents the nth intermediate representation data, n represents the total number of data items in the input data set, and W y is the output weight parameter, by is the output bias parameter, and σ represents the sigmoid function.

6. The real-time data collection and analysis method for digital transaction of agricultural products according to claim 5 is characterized in that: And set preset thresholds to obtain transaction risk levels, including: Based on the historical data set, a transaction data risk factor is obtained; the transaction data risk factor includes agricultural product market fluctuation data; supply chain impact data; policy change data, credit information data of both parties to the transaction, and agricultural disaster risk data; According to the transaction data risk factor, by: ; To obtain the risk threshold, where RI is the risk threshold, V is the market volatility, S is the supply chain impact parameter, P is the policy risk parameter, C is the credit risk parameter, I is the agricultural disaster risk parameter, and w1, w2, w3, w4, w5 are weight coefficients; Based on the risk threshold, the newly acquired risk result of the data set output by the data analysis model is graded, and the grade classification includes two risk levels: excellent and poor.

7. A real-time data collection and analysis method for digital transactions of agricultural products according to claim 6, characterized in that: Design simulated trading scenarios based on risk index to assess the trading risk level, including: Based on the risk results of agricultural product transactions output by the data analysis model, design immersive transaction experience scenarios for multiple agricultural products in multiple periods; The various agricultural products include fruits and vegetables, livestock, poultry and grains; The immersive trading experience scenarios at each stage include when agricultural products are growing, maturing, processing, transporting, trading and using; Based on immersive agricultural product trading experience scenarios in multiple periods, users can conduct trading experiences in different periods and obtain corresponding risk level assessment trends.

8. The real-time data collection and analysis method for digital transaction of agricultural products according to claim 7 is characterized in that: Based on immersive agricultural product trading experience scenarios in multiple periods, users can experience trading in different periods and obtain corresponding risk assessment trends, including: Acquire scenario features at multiple stages and generate a complete scenario experience for agricultural product transactions, so that users can fully participate in the transaction decision-making at the entire cycle of agricultural products and obtain detailed feedback information on transactions at each stage of agricultural products; The scene features of the multiple stages include the transaction experience scene features of the agricultural product growth stage, the transaction experience scene features of the agricultural product maturity stage; the transaction experience scene features of the agricultural product processing stage; the transaction experience scene features of the agricultural product transportation stage; the transaction experience scene features of the agricultural product trading stage; and the transaction experience scene features of the agricultural product use stage. Based on the transaction data risk factors, the experience scenario continuously monitors and dynamically updates the risk dynamics of each period, and generates corresponding risk level assessment results, so that users can understand the direction of risk dynamic changes in the agricultural product trading process at multiple stages; The risk trend is obtained based on the dynamic changes in risks during agricultural product transactions, and the final risk assessment report is generated based on the transaction experience and risk assessment results of all stages to support users in comprehensively analyzing market dynamics at each stage.

9. The real-time data collection and analysis method for digital transaction of agricultural products according to claim 8 is characterized in that: Based on the assessment results, we can avoid agricultural product trading investment risks, including: Based on the good and bad risk results of agricultural product transactions output by the data analysis model, when agricultural product transactions are at a low risk level, users can obtain the transaction status of the same agricultural product in different regions or different seasons through immersive agricultural product transaction experience scenarios; Based on the trading situation of the same agricultural products, the best trading period of the same agricultural products in each region is obtained to obtain reference data of agricultural product trading information; Based on the agricultural product transaction information reference data, it provides a reference for users to accurately invest funds when conducting agricultural product transactions.

10. A real-time data collection and analysis system for digital transactions of agricultural products, characterized in that: The system is provided with an electronic device including a memory, a processor, and a real-time data collection and analysis method program for digital transactions of agricultural products stored in the memory and executable on the processor. When the real-time data collection and analysis method program for digital transactions of agricultural products is executed by the processor, the steps of the real-time data collection and analysis method for digital transactions of agricultural products as claimed in any one of claims 1 to 9 are implemented. The system includes: Data collection module: It is used to collect digital transaction information data of agricultural products to obtain a data set; train the historical data set to obtain a data analysis model to output the risk standards of agricultural product transactions; Data analysis module: used to input the newly acquired data set into the data analysis model; based on the data analysis model, receive the newly acquired data set to output the transaction pros and cons risk results of the newly acquired data set, and set a preset threshold to obtain the transaction risk level of the data set; Risk assessment module: It is used to provide an immersive multi-period agricultural product trading risk experience by designing and simulating agricultural product trading scenarios based on the transaction risk level. Based on the multi-period agricultural product trading risk experience, it is used to obtain the risk assessment trend of agricultural product transactions corresponding to different periods, and based on the assessment results, it is used to avoid agricultural product trading investment risks.

Images