Intelligent agricultural product price prediction assistant applet system

Abstract

The application discloses a kind of wisdom agricultural product price assistant applet system, belong to computer software, agricultural product big data and artificial intelligence technical field, to solve the problems of low efficiency of agricultural product price data entry, high operation threshold, insufficient prediction accuracy, function closed loop loss in prior art.The system contains seven core modules, each module is associated with each other through applet background and corresponds to exclusive interface (Fig.1-4), can realize agricultural product first-hand data entry verification, data standardization processing, historical data management, price trend visualization, adopts DeepSeek large model to realize 3-15 day price accurate prediction, output planting sales suggestion and collect user feedback to optimize system.The application is light weight and installation-free, convenient to operate, builds whole-process closed loop, adapts to the needs of grassroots farmers, improves the accuracy of price prediction and the scientific nature of transaction decision.

Description

Technical Field

[0001] This invention belongs to the fields of computer software, agricultural big data and artificial intelligence technology, and specifically relates to a smart agricultural product price assistant mini-program system. Specifically, it relates to a lightweight agricultural product price service system with functions such as transaction price input, data management, price visualization, AI prediction and feedback optimization, and corresponding to a specific mini-program interface. Background Technology

[0002] Agricultural product prices are the core decision-making basis for the entire agricultural production, circulation, and trading chain. Farmers, agricultural product dealers, and grassroots trading personnel generally face industry pain points such as price information asymmetry, difficulty in predicting price trends, and low efficiency in collecting first-hand transaction data during the agricultural product trading process. Currently, existing technologies in the field of agricultural product price services have the following core defects: 1. Low data collection efficiency and high operational threshold: Existing agricultural product price data collection systems are mostly complex forms on PCs with cumbersome operation processes. Grassroots users (such as farmers and individual dealers) find it difficult to quickly complete the first-hand data entry at the trading site, and cannot achieve real-time reporting in the fields and wholesale markets; for example, the existing "Agricultural Price Pass" APP has a cumbersome operation process, requiring registration and login to complete multiple forms, which is not suitable for the lightweight use needs of grassroots farmers, and lacks a simple visual input interface. 2. Insufficient Price Forecast Accuracy: Existing price forecasting tools rely solely on historical price data for basic trend analysis, failing to incorporate multi-dimensional spatiotemporal enhancement factors such as weather, supply chain, and special events. This results in significant discrepancies between forecasts and actual market conditions, failing to provide effective support for user trading decisions. Furthermore, the lack of a dedicated AI forecasting interface makes operation inconvenient. 3. Lack of Functional Loop: Existing products only allow for single data entry or price queries, failing to construct a complete closed-loop process of "data entry - data management - trend visualization - AI forecasting - feedback optimization." User feedback cannot directly impact algorithm model iteration and optimization, and the absence of a dedicated feedback interface makes feedback collection and tracking difficult. 4. Poor Adaptability: Existing systems are primarily designed for professional institutions, featuring redundant functions and complex operations. They cannot meet the lightweight usage needs of grassroots users, making promotion difficult. The lack of a dedicated integrated interface for data management and price visualization makes it difficult for users to quickly access historical data and grasp price trends. The aforementioned shortcomings lead to difficulties in collecting first-hand data for agricultural product transactions, low accuracy in price forecasting, and a lack of effective support for user decision-making. Therefore, there is an urgent need to develop a lightweight agricultural product price assistant system that is easy to operate, has a closed-loop function, provides accurate forecasts, and has a corresponding visual interface to solve the core pain points of existing technologies. Summary of the Invention

[0003] The technical problem this invention aims to solve is to overcome the shortcomings of existing technologies and provide a smart agricultural product price assistant mini-program system. This system addresses the issues of low data entry efficiency, high operational barriers, insufficient price prediction accuracy, and lack of functional closed-loop mechanisms in existing technologies. It achieves lightweight collection, intelligent management, accurate prediction, and closed-loop optimization of agricultural product transaction data, with each core function corresponding to a dedicated mini-program interface, enhancing user convenience. To solve the above technical problems, this invention provides the following technical solution: This invention provides a smart agricultural product price assistant mini-program system, comprising seven core modules: a transaction price entry module, a data preprocessing module, an entered data management module, a price trend visualization module, a DeepSeek AI intelligent prediction module, a suggestion generation module, and a project feedback and suggestion module. These modules are logically linked and interconnected through the mini-program's backend, collaboratively achieving a closed-loop functional process. Each core module corresponds to a dedicated mini-program interface, as follows: 1. Transaction Price Entry Module: [The text abruptly ends here, so the translation stops here as well.] Figure 1 The smart agricultural product price assistant mini-program shown here has a transaction price entry interface. This module is used to quickly enter first-hand agricultural product transaction data. The interface includes a basic information entry area and an AI-assisted enhancement factor entry area. The basic information entry area includes five core fields: product name input box, variety selection box, transaction location input box, transaction price input box, and transaction date selection box. It supports the rapid reuse of historical products, provides an entry point for manually entering new products, and also supports the advanced function of batch entry for multiple consecutive days. After the user has filled in the information, they can click the save button on the interface to save a single price data entry with one click. The AI-assisted enhancement factor entry area is optional and includes three dimensions: weather conditions, supply chain conditions, and special events / other factors. It is used to provide multi-dimensional enhancement data for the AI ​​prediction model and improve prediction accuracy. The module also includes an input interface unit and a data verification unit. The data verification unit is used to verify the format of the data entered by farmers, reminding farmers to supplement missing data and correct abnormal data. It has price range verification and duplicate entry reminder functions to ensure the accuracy of the input data. 2. Data Preprocessing Module: This module receives farmer input data and influencing factors from the transaction price entry module, organizes and verifies it, integrates and processes the acquired agricultural product data, removes abnormal data, fills in missing data, and converts input data of different formats into a standardized format to obtain a standardized and personalized database. This provides a rich and sufficient data source for data prediction, achieves standardized data management, and ensures the quality of input data for the prediction model. The processed standardized data is synchronized to both the entered data management module and the DeepSeek AI intelligent prediction module, providing support for subsequent data management and price prediction. 3. Entered Data Management Module: Corresponding to... Figure 2The data management area in the "Smart Agricultural Product Price Assistant" mini-program's data entry management and price trend visualization interface is shown. This module is used to view and manage historical data entries under the current user environment. The interface includes two quick operation buttons: refresh / load data and hide data. It displays core information about the entered data in a table format, including product name, variety, transaction location, transaction date, and transaction price, enabling unified management and quick access to historical data. Simultaneously, this module synchronously transfers stored historical price data to the price trend visualization module to generate price fluctuation trend charts. 4. Price Trend Visualization Module: Corresponding to... Figure 2 The visualization area in the "Data Management and Price Trend Visualization" interface of the Smart Agricultural Product Price Assistant mini-program is shown. This module works in conjunction with the Data Management module, automatically generating a line chart of price fluctuation trends based on historical price data transmitted from the Data Management module, intuitively displaying price change patterns. If the user has not entered any historical data, the interface will prompt "No data available, please enter prices to generate a chart," achieving data visualization and helping users quickly grasp price change trends. Simultaneously, this module also displays price prediction results generated by the DeepSeek AI Intelligent Prediction module and decision suggestions output by the suggestion generation module, achieving an integrated display of data, predictions, and suggestions. 5. DeepSeek AI Intelligent Prediction Module: Corresponding to... Figure 3The DeepSeek AI intelligent prediction interface of the Smart Agricultural Product Price Assistant mini-program shown is based on a deep learning model enhanced with historical data and spatiotemporal factors. Utilizing the DeepSeek large model, it achieves intelligent prediction of agricultural product prices. The interface features four core input fields: a prediction product input field, a variety selection field, a reference trading point input field, and a target prediction region input field. Variety selection is optional; leaving it blank indicates a general variety. Users can customize the prediction time span using a slider on the interface; a prediction period of 3-7 days is recommended for optimal accuracy, with a maximum supported prediction period of 15 days. After completing the input, users can click the "Start AI Model to Generate Prediction Report" button to instantly generate a prediction report, providing users with agricultural product price trend predictions and decision support. The module uses a regression prediction model, which calculates the supply and demand relationship of agricultural products and market fluctuation patterns to generate wholesale price prediction data for corresponding agricultural product categories. The prediction data is accurate to a single day or week, and the prediction results are simultaneously transmitted to the price trend visualization module and the suggestion generation module. 6. Suggestion Generation Module: This module, based on wholesale price forecast data for agricultural products generated by the DeepSeek AI intelligent prediction module, combines agricultural product market circulation patterns, planting and sales knowledge, and relevant hot policies to provide farmers with targeted planting and sales suggestions. These suggestions include recommendations on sales timing, procurement channels, and adjustments to planting categories. The suggestions are concise, easy to understand, and tailored to farmers' practical needs. The suggestions are also simultaneously transmitted to the price trend visualization module. Figure 2 The visualization interface shown displays price trend charts and forecast results together for easy user viewing. 7. Project Feedback and Suggestion Module: Corresponding to... Figure 4 The smart agricultural product price assistant mini-program shown here has a project feedback and suggestion interface. This module is used to collect user feedback and optimization suggestions. The interface includes a feedback content input box, an image upload button, and a submit feedback button, as well as a comment posting unit and a feedback collection unit. The comment posting unit supports both text and image feedback. Users can fill in the feedback content in the input box, upload relevant images for supplementary explanation via the image upload button, and click the submit feedback button to complete the submission. The feedback collection unit categorizes farmer feedback information (function optimization suggestions, prediction accuracy feedback, usage questions) and automatically summarizes it into a feedback report for system optimization. The interface also includes a recent feedback record display area to achieve full-process tracking of user feedback. Users can view their historical feedback content and processing status here, constructing a product closed loop of "use-feedback-optimization". Compared with existing technologies, the beneficial effects of this invention are as follows: 1. Significantly reduced operational threshold and significantly improved data entry efficiency: This invention is developed based on WeChat mini-programs, is lightweight and requires no installation, and each core function corresponds to a dedicated visual interface (…). Figure 1-4The core data entry process is simplified to within 3 steps, while supporting the reuse of historical products and batch entry functions, adapting to the rapid reporting needs of trading sites. Grassroots farmers and individual dealers can quickly get started, solving the problems of complex operation and low entry efficiency of existing systems; 2. Significantly improved price prediction accuracy: In addition to basic transaction data, this invention adds multi-dimensional AI-assisted enhancement factors such as weather, supply chain, and special events. Combined with the DeepSeek large model and spatiotemporal factor-enhanced deep learning algorithm, the prediction results significantly improve the matching degree with actual market trends, and through a dedicated AI prediction interface ( Figure 3 3. Achieve convenient operation and provide users with effective transaction decision support; 4. Achieve a closed-loop function throughout the entire process: This invention constructs a "data entry ( Figure 1 Data preprocessing - Data management Figure 2 Trend Visualization Figure 2 AI prediction Figure 3 - Suggestion generation - Feedback optimization Figure 4 The complete functional loop of ")" allows each link to be linked through a dedicated interface. User feedback can directly affect the iterative optimization of the algorithm model, enabling the product to continuously optimize and upgrade itself; 4. Strong scenario adaptability: This invention has been specially optimized for all scenarios of agricultural product circulation and trading. The interface design is simple and intuitive, which can not only meet the lightweight use needs of individual farmers and distributors, but also meet the use needs of professional institutions through batch input and multi-dimensional factor configuration. It has strong adaptability and low promotion threshold. Attached Figure Description

[0004] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of the transaction price input interface of the Smart Agricultural Product Price Assistant mini-program of this invention; Figure 1 The interface includes a product name input box (1), a variety selection box (2), a transaction location input box (3), a transaction price input box (4), a transaction date selection box (5), an AI-assisted enhancement factor input area (6), and a save button (7). Users can quickly enter agricultural product transaction data and influencing factors through this interface. The data verification unit verifies the input data in real time to ensure data validity. Figure 2 This is a schematic diagram of the data management and price trend visualization interface of the Smart Agricultural Product Price Assistant mini-program of this invention; Figure 21 is the table of entered data (corresponding to the entered data management module), 2 is the refresh / load data button, 3 is the hide data button, and 4 is the price trend line chart (corresponding to the price trend visualization module). Users can view historical entered data and operation data on this interface, as well as view price fluctuation trends, prediction results, and decision suggestions. When there is no historical data, a corresponding prompt will be displayed. Figure 3 This is a schematic diagram of the DeepSeek AI intelligent prediction interface of the smart agricultural product price assistant mini-program of this invention; Figure 3 The interface includes: 1 for the product prediction input box, 2 for the product selection box, 3 for the reference trading point input box, 4 for the target prediction region input box, 5 for the prediction time span slider, and 6 for the AI ​​model start button. Users can fill in prediction-related information, customize the prediction period, and start the AI ​​model to generate a price prediction report with one click. Figure 4 This is a schematic diagram of the project feedback and suggestion interface of the Smart Agricultural Product Price Assistant mini-program of this invention; Figure 4 The interface includes a feedback input box (1), an image upload button (2), a submit feedback button (3), and a recent feedback record display area (4). Users can submit text and image feedback, view historical feedback records and processing status, and track the entire feedback process. Detailed Implementation

[0005] The following is in conjunction with the appendix Figure 1-4 The preferred embodiments of the present invention will be described below. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention. In the accompanying drawings, all the same reference numerals refer to the same parts. Figure 1-4 As shown, this invention provides a smart agricultural product price assistant mini-program system, including seven core modules: a transaction price entry module, a data preprocessing module, a data entry management module, a price trend visualization module, a DeepSeek AI intelligent prediction module, a suggestion generation module, and a project feedback and suggestion module. These modules are logically linked and interconnected through the mini-program's backend to achieve data transfer and collaborative work. Each module corresponds to a dedicated mini-program interface. Specific implementation details are as follows: 1. Transaction Price Entry Module (corresponding to...) Figure 1 Users can directly access the mini-program system homepage by scanning the QR code or searching for the mini-program name "Maihaojia - Smart Agricultural Product Price Assistant" via WeChat, without needing to register, log in, or download and install; users can then click to enter the transaction price entry interface. Figure 1In the basic information entry area, users fill in the product name, variety, transaction location, transaction price, and transaction date. The transaction date is filled in by default with the current date, but users can quickly select it using the "Today" and "Yesterday" shortcut buttons, or manually customize the date. For products that have been entered previously, the basic information can be filled in with one click using the "Quickly Reuse Historical Products" function, eliminating the need for re-entry. In the AI-assisted enhancement factor entry area, users can select the corresponding weather conditions and supply chain conditions, and fill in special events / other factors to provide multi-dimensional enhancement data for the AI ​​prediction model. After filling in the information, users can click the "Save Single Price Data" button on the interface to complete the data submission. The system automatically transmits the data to the data preprocessing module, processes it, and stores it in the local and backend databases, synchronously updating the entered data management module. The data verification unit will verify the input data in real time. If there are format errors, missing data, or duplicate entries, the system will promptly remind the user to correct and supplement the data to ensure that the input data meets the requirements. 2. Data Preprocessing Module: After receiving data from the transaction price entry module, this module organizes, verifies, and integrates the data, removing abnormal data that does not conform to the actual production and trading patterns of agricultural products, filling in missing data, and converting input data of different formats into a standardized format to generate a standardized and personalized database. The standardized data is then transmitted to the entered data management module and the DeepSeek AI intelligent prediction module, providing a high-quality data source for data management and price prediction, ensuring the normal operation of subsequent functions. 3. Entered Data Management Module (corresponding to...) Figure 2 Data Management Area: After receiving standardized data from the data preprocessing module, this area organizes and archives the data chronologically, displaying core information such as product name, variety, transaction location, transaction date, and transaction price in tabular form. Users can click the "Refresh / Load Data" button to update the latest entered data and the "Hide Data" button to hide historical records, enabling convenient management of historical data. Simultaneously, this module transmits historical price data to the price trend visualization module to generate price fluctuation trend charts, achieving data and visualization linkage. 4. Price Trend Visualization Module (corresponding to...) Figure 2 Visualization Area: After receiving historical price data from the data management module, it automatically generates a line chart of price fluctuations to visually display the patterns of agricultural product price changes. If the user has not entered any historical data, the interface will display "No data available, please enter prices to generate the chart." Users can quickly grasp agricultural product price trends through this module, and simultaneously view price prediction results generated by the DeepSeek AI intelligent prediction module and decision-making suggestions output by the suggestion generation module, achieving an integrated display of data, predictions, and suggestions to provide reference for trading decisions. 5. DeepSeek AI Intelligent Prediction Module (corresponding to...) Figure 3After receiving standardized data from the data preprocessing module, the user clicks to enter the DeepSeek AI intelligent prediction interface. Figure 3 The user enters the predicted product, variety, reference trading point, and target prediction region. Variety is optional; leaving it blank indicates a general variety. Users can customize the prediction time span using a slider on the interface, selecting a prediction period of 3-15 days. The system defaults to a 7-day prediction period for optimal accuracy. After clicking the "Start AI Model to Generate Prediction Report" button, the system calls the DeepSeek large model combined with spatiotemporal factor enhancement deep learning algorithms to calculate the supply and demand relationship and market fluctuation patterns of agricultural products based on standardized data, generating wholesale price prediction data for the corresponding agricultural product category (accurate to a single day or week). Simultaneously, the prediction results are transmitted to the suggestion generation module and the price trend visualization module for user viewing. 6. Suggestion Generation Module: After receiving the price prediction results from the DeepSeek AI intelligent prediction module, it combines agricultural product market circulation patterns, local market characteristics, planting and sales common sense, and relevant hot policies to generate targeted planting and sales suggestions, including suggestions on sales timing, procurement channels, and adjustments to planting categories. These suggestions are then transmitted to the price trend visualization module. Figure 2 The visualization interface shown presents the prediction results and price trend charts to the user, allowing them to combine trends and suggestions to make trading decisions. 7. Project Feedback and Suggestions Module (corresponding to...) Figure 4 Users click to enter the project feedback and suggestion interface ( Figure 4 Users can fill in their feedback, feature suggestions, and optimization requests in the feedback input box. They can also upload relevant images to support their feedback. Clicking the "Submit Feedback" button completes the submission. The comment posting unit transmits user feedback to the feedback collection unit, which categorizes and organizes the feedback information, generating a feedback report that provides core evidence for system feature optimization and predictive model parameter adjustment. Users can view their historical feedback content and processing status in the "Recent Feedback Records" area, enabling full-process tracking of feedback and facilitating continuous system optimization. The working principle and usage process of this invention: This invention is developed based on the WeChat Mini Program ecosystem, with each core function corresponding to a dedicated interface (…). Figure 1-4 This provides users with lightweight, end-to-end agricultural product pricing services; users can access these services through... Figure 1 The transaction price entry interface shown allows for the rapid entry of primary transaction data using the transaction price entry module. After processing by the data preprocessing module, the data is transmitted to both the entered data management module and the DeepSeek AI intelligent prediction module. The entered data management module then... Figure 2 The interface shown enables the management of historical data, and the price trend visualization module is located within... Figure 2 The interface generates price trend charts; users can... Figure 3 The interface shown allows users to initiate price predictions using the DeepSeek AI intelligent prediction module. After the system generates the prediction results, the suggestion generation module outputs targeted decision-making suggestions. The prediction results and suggestions are then shared through... Figure 2 The interface shown is presented to the user; the user can... Figure 4 The interface shown allows users to submit feedback through the project feedback and suggestion modules, which helps to iterate and optimize the system and prediction model. The entire process is convenient and adaptable to the needs of grassroots users. Furthermore, the interfaces work together smoothly to achieve a closed loop of functions throughout the entire process.

Claims

1. A smart agricultural product price assistant mini-program system, characterized in that, It includes a transaction price entry module, a data preprocessing module, a DeepSeek AI intelligent prediction module, a suggestion generation module, a project feedback and suggestion module, a data entry management module, and a price trend visualization module. These modules are logically linked and interconnected through the mini-program's backend, collaboratively achieving the prediction, data management, visualization, and feedback optimization of agricultural product market prices. The transaction price entry module is used to obtain wholesale price data of agricultural products and related influencing factors, including real-time weather conditions, supply chain conditions, and special events provided by farmers. This module corresponds to the transaction price entry interface of the mini-program, and sets up a visual input window, which includes input fields for product name, variety, transaction location, transaction price, and transaction date, as well as an AI-assisted enhancement factor input area, and supports data saving function. The data preprocessing module is used to receive data and influencing factors input by farmers, organize and verify them, integrate and process the acquired agricultural product data to obtain a standardized and personalized database, providing a rich and sufficient data source for data prediction. The processed data is synchronized to the data management module and the DeepSeek AI intelligent prediction module. The DeepSeek AI intelligent prediction module is used to predict agricultural product-related indicators through a preset model. It adopts a deep learning algorithm that combines the DeepSeek large model with spatiotemporal factor enhancement to make multi-dimensional predictions of agricultural product market prices and generate predicted wholesale price data. The module corresponds to the DeepSeek AI intelligent prediction interface in the mini-program, which allows users to set input fields for predicted products, varieties, reference trading points, target prediction regions, and a prediction time span slider. It also supports one-click startup of the AI ​​model to generate prediction reports. The suggestion generation module, based on the generated wholesale price forecast data for agricultural products, combined with the market circulation patterns of agricultural products, common knowledge about planting and sales, and relevant hot policies, provides farmers with targeted planting and sales suggestions. These suggestions are simultaneously displayed on the price trend visualization interface. The project feedback and suggestion module, corresponding to the mini-program's project feedback and suggestion interface, includes a feedback content input box, an image upload button, a submit feedback button, and a recent feedback record display area. Farmers can post comments and feedback in text and image formats. The received feedback information is categorized and summarized to form a feedback report, providing data support for the optimization of the system and prediction model. The data entry management module corresponds to the data entry management interface of the mini-program, which is used to view and manage the historical data entry records in the current environment of the user. It has two quick operation buttons: refresh / load data and hide data. The core information of the data entry is displayed in a table format, realizing unified management and quick access to historical data. The historical price data stored in it is synchronized to the price trend visualization module. The price trend visualization module corresponds to the price trend visualization interface of the mini-program. It works in conjunction with the data entry management module. Based on the historical price data entered by the user, it automatically generates a price fluctuation trend chart in the form of a line graph, which intuitively displays the price change pattern. When no data is entered, it displays the corresponding prompt information. At the same time, it displays the prediction results generated by the DeepSeek AI intelligent prediction module and the decision suggestions output by the suggestion generation module.

2. The smart agricultural product price assistant mini-program system according to claim 1, characterized in that, The transaction price entry module includes an input interface unit and a data verification unit. The input interface unit corresponds to the transaction price entry interface shown in Figure 1, providing a visual input window for farmers to input agricultural product categories, wholesale price data, and market influencing factors. It supports the rapid reuse of historical products and batch entry for multiple consecutive days, and includes a button for saving a single price data entry with one click. The data verification unit is used to verify the format of the data entered by farmers, reminding them to supplement missing data and correct abnormal data. It also includes price range verification and duplicate entry reminder functions.

3. The smart agricultural product price assistant mini-program system according to claim 1, characterized in that, The DeepSeekAI intelligent prediction module uses a regression prediction model as its preset model. Based on the agricultural product-related data and influencing factors input by farmers, the model calculates the supply and demand relationship of agricultural products and market fluctuation patterns to generate wholesale price prediction data for the corresponding category of agricultural products. The prediction data is accurate to a single day or a single week. This module corresponds to the DeepSeekAI intelligent prediction interface shown in Figure 3. Users can customize the prediction time span from 3 to 15 days using a slider. The default recommendation is a 7-day prediction period to obtain the best accuracy. Clicking the "Start AI Model" button will generate a prediction report.

4. The intelligent agricultural product price assistant mini-program system according to claim 1, characterized in that, The sales suggestions generated by the suggestion generation module, combined with agricultural product wholesale price forecast data and local market circulation characteristics, specifically include suggestions on sales timing, recommendations on purchasing channels, and suggestions on adjusting planting categories. The suggestions are concise, easy to understand, and tailored to the actual operational needs of farmers. They are simultaneously displayed on the price trend visualization interface shown in Figure 2, presented in conjunction with price trend charts and forecast results.

5. The smart agricultural product price assistant mini-program system according to claim 1, characterized in that, The project feedback and suggestion module includes a comment posting unit and a feedback collection unit. The comment posting unit corresponds to the project feedback and suggestion interface shown in Figure 4, supporting both text and image feedback formats, and includes a feedback content input box, an image upload button, and a submit feedback button. The feedback collection unit categorizes farmer feedback information (function optimization suggestions, prediction accuracy feedback, and usage questions) and automatically summarizes it into a feedback report for system optimization. It also includes a recent feedback record display area on the interface, allowing users to view historical feedback records and processing status.

6. A prediction method based on the smart agricultural product price assistant mini-program system according to any one of claims 1-5, characterized in that, Includes the following steps: Step 1: Using the transaction price entry interface shown in Figure 1, obtain the historical price data of the target agricultural product at the designated transaction point and the corresponding auxiliary information data through the transaction price entry module. The auxiliary information data includes at least one of weather data and supply chain status data. After filling in the information, click the save button to submit the data. Step 2: The historical price data and auxiliary information data are cleaned, missing values ​​are filled, and outliers are identified through the data preprocessing module to generate standardized time series data, which is then synchronously transmitted to the data management module and the DeepSeek AI intelligent prediction module. Step 3: Using the DeepSeek AI intelligent prediction interface shown in Figure 3, and with the help of the DeepSeek AI intelligent prediction module, based on the standardized time series data, the DeepSeek large model combined with the spatiotemporal factor enhancement deep learning algorithm generates the price prediction results for the specified time period in the future. After the prediction is completed, it is synchronously transmitted to the suggestion generation module and the price trend visualization module. Step 4: Based on the price forecast results, the suggestion generation module generates decision-making suggestions for the target farmers. The decision-making suggestions include: suggested delivery time, suggested batch delivery strategy, current price risk level and corresponding response prompts. Step 5: Through the price trend visualization interface shown in Figure 2, and with the help of the price trend visualization module, the price prediction results and decision suggestions are presented to the user in the form of text reports, line charts, or trend summaries. If there is no historical data, the prompt "No data available, please enter the price to generate the chart" will be displayed. Step 6: Through the data entry management interface shown in Figure 2, the historical data entry and prediction-related data can be managed in a unified manner using the data entry management module. Data can be operated by refreshing / loading data and hiding data buttons. Through the project feedback and suggestion interface shown in Figure 4, farmer feedback can be collected using the project feedback and suggestion module to provide support for the optimization of the system and prediction model.

7. The prediction method of the smart agricultural product price assistant mini-program system according to claim 6, characterized in that, The DeepSeek large model in step 3 is only used to: convert structured prediction results into easy-to-understand natural language statements; generate risk warnings and uncertainty explanations based on auxiliary information data; and the DeepSeek large model does not modify the numerical prediction results. The prediction results are simultaneously displayed on the AI ​​prediction interface shown in Figure 3 and the visualization interface shown in Figure 2.

Images