Intention prediction method and system based on AI machine learning

Abstract

The present invention discloses an intention prediction method and system based on AI machine learning, and belongs to the technical field of artificial intelligence intention prediction. The method specifically comprises: collecting historical behavior data of customers purchasing agricultural products, building a customer database, preprocessing the collected historical behavior data of customers purchasing agricultural products, extracting the preference characteristics of customers purchasing agricultural products from the preprocessed historical behavior data of customers purchasing agricultural products, including: the frequency, price, time, agricultural product category preference and browsing behavior characteristics of customers purchasing agricultural products, building a customer purchase intention machine learning model based on the extracted customer purchase preference characteristics of agricultural products, predicting the customer's intention to purchase agricultural products, optimizing and improving the model based on actual application conditions and feedback information, and improving the accuracy and effectiveness of purchase intention prediction.

Description

Technical Field

[0001] The present invention belongs to the technical field of artificial intelligence intention prediction, and specifically to an intention prediction method and system based on AI machine learning. Background Art

[0002] Online shopping is increasingly becoming the mainstream of shopping. When customers browse and purchase products, a large amount of data is generated. How to mine this data to help sales is a hot topic in the current market.

[0003] Currently, there is no problem of deterioration during transportation for mainstream shopping commodities. However, for agricultural products, they may deteriorate during transportation due to temperature, moisture and other factors, which will have a negative impact on customers' repurchase of the agricultural products.

[0004] For example, a Chinese patent with authorization announcement number CN110232589B discloses a potential customer analysis system based on big data, including a client that communicates with the system. The system also includes a product classification module, a customer behavior acquisition module, a product intention grading module, a keyword extraction module, a product function extraction module, and a storage module. The product intention grading module divides the customer's intention into two intention levels, which can more accurately determine the customer's purchase intention and provide data support for later sales. The keyword extraction module generates necessary function keywords, irrelevant function keywords, and preferred function keywords to indicate the functions that customers pay most attention to and the functions that they do not pay attention to for a category of goods, which will further enhance the function of product evaluation. Further mining of different intention products is achieved, which can be used for later advertising push and consumption prediction.

[0005] For example, the Chinese patent with publication number CN115131074A discloses an intention prediction method and related devices, which can be applied to the fields of artificial intelligence and finance. The invention can obtain the information of the target customer; characterize the information to obtain the corresponding feature data; input the feature data into a pre-trained intention prediction model, thereby obtaining the intention of the target customer to purchase a lump-sum deposit and withdrawal product output by the intention prediction model. It can be seen from this that the invention can predict the customer's intention to purchase a lump-sum deposit and withdrawal product based on the intention prediction model, and then promote the lump-sum deposit and withdrawal product to customers with the intention to purchase, which can effectively improve the success rate and improve the utilization of resources, thereby saving resources.

[0006] The defects of the above patent are: 1) it cannot be directly applied to the prediction of purchase intention of agricultural products; 2) it does not take into account the problem of deterioration of agricultural products during transportation. Summary of the invention

[0007] In view of the shortcomings of the prior art, the present invention proposes an intention prediction method and system based on AI machine learning, collects historical behavior data of customers purchasing agricultural products, constructs a customer database, preprocesses the collected historical behavior data of customers purchasing agricultural products, and extracts the customer's preference characteristics for purchasing agricultural products from the preprocessed historical behavior data of customers purchasing agricultural products, including: the frequency, price, time, agricultural product category preference and browsing behavior characteristics of customers purchasing agricultural products. According to the extracted customer preference characteristics for purchasing agricultural products, a customer purchase intention machine learning model is constructed to predict the customer's intention to purchase agricultural products. According to actual application conditions and feedback information, the model is optimized and improved to improve the accuracy and effectiveness of purchase intention prediction.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] The intention prediction method based on AI machine learning includes the following specific steps:

[0010] Step S1: Collect historical behavior data of customers purchasing agricultural products and build a customer database;

[0011] Step S2: pre-processing the collected historical behavior data of customers purchasing agricultural products;

[0012] Step S3: extracting the preference characteristics of customers in purchasing agricultural products from the pre-processed historical behavior data of customers purchasing agricultural products, including: the frequency, price, time, category preference and browsing behavior characteristics of customers purchasing agricultural products;

[0013] Step S4: construct a customer purchase intention machine learning model based on the extracted customer preference characteristics for purchasing agricultural products, and predict the customer's intention to purchase agricultural products;

[0014] Step S5: Optimize the model according to actual application conditions and feedback information to determine the degree of intention.

[0015] Specifically, the historical behavior data of the customer purchasing agricultural products in step S1 includes: historical record data of purchasing agricultural products, historical browsing data of agricultural products and user personal information data.

[0016] Specifically, the preprocessing in step S2 includes: cleaning, deduplication, missing value processing, removing redundant and repeated data, correcting missing data, and ensuring the quality and integrity of the data.

[0017] Specifically, the specific steps of step S3 are:

[0018] Step S301: Set the pre-processed historical behavior data set of customers purchasing agricultural products as , ,in, Indicates the nth agricultural product purchased by the customer in history. Indicates the price of the nth agricultural product purchased by the customer in history, Indicates the time when the customer purchased the nth agricultural product in history. Indicates the category of the nth agricultural product purchased by the customer in history;

[0019] Step S302: extract the customer's preference characteristics for purchasing agricultural products, and calculate the frequency of the customer purchasing the i-th type of agricultural products. The calculation formula is: ,in, represents the frequency of customers purchasing the i-th type of agricultural products, represents the quantity of agricultural products of category i purchased by customers, Indicates the years between the first and last purchase of agricultural products in the historical behavior data of customers purchasing agricultural products; ([] is the time difference, the data is in date format, xxxx year xx month xx day, Gs function converts the date format data into years, counts the number of days between the first and last purchase of agricultural products, and divides by 365 to get the number of years)

[0020] Step S303: Calculate the customer's category preference for purchasing the i-th category of agricultural products, and the calculation formula is:

[0021] ,

[0022] in, represents the customer's category preference for purchasing the i-th category of agricultural products, Indicates the influence weight of customers browsing agricultural products, represents the number of times customers browse the i-th category of agricultural products, Indicates the total number of times customers browsed all categories of agricultural products. represents the total number of times customers purchase the i-th type of agricultural products, Indicates the number of types of agricultural products purchased by customers;

[0023] Step S304: Calculate the influence factor of the customer's browsing behavior characteristics on the intention to purchase the i-th type of agricultural products, and the calculation formula is:

[0024] ,

[0025] in, represents the influencing factor of customer browsing behavior characteristics on the intention to purchase the i-th type of agricultural products, m represents the total number of times customers browse agricultural products, represents the number of times customers browse the i-th category of agricultural products, Indicates whether the customer purchased the i-th agricultural product that he browsed. is the judgment function. When the user buys the i-th agricultural product that he browses, =1, when the user did not purchase the i-th agricultural product that he browsed, =0.

[0026] Specifically, the specific steps of step S4 are:

[0027] Step S401: selecting a machine learning model based on the extracted customer preference characteristics for purchasing agricultural products, and training the selected machine learning model;

[0028] Step S402: Use the trained machine learning model to predict the customer's intention to purchase agricultural products. The specific prediction formula is:

[0029] ,

[0030] in, represents the customer's intention to purchase the jth agricultural product in the i-th category, represents the deterioration coefficient of the jth agricultural product in the i-th category during transportation, represents the initial quality of the jth agricultural product in the i-th category, represents the sales impact coefficient of the jth agricultural product in the i-th category, represents the sales volume of the jth agricultural product in the i-th category, represents the price impact coefficient of the jth agricultural product in the i-th category, represents the price of the jth agricultural product in the i-th category;

[0031] Step S403: According to , predicting the agricultural products that customers want to buy.

[0032] Specifically, the machine learning model in step S401 includes: any one of the deep learning prediction models.

[0033] The intention prediction system based on AI machine learning includes: historical data collection module, historical data preprocessing module, preference feature extraction module, customer purchase intention prediction module and optimization and improvement module;

[0034] The historical data collection module is used to collect historical behavior data of customers purchasing agricultural products and build a customer database;

[0035] The historical data preprocessing module is used to preprocess the collected historical behavior data of customers purchasing agricultural products;

[0036] The preference feature extraction module is used to extract the preference features of customers purchasing agricultural products from the pre-processed historical behavior data of customers purchasing agricultural products, including: the frequency, price, time, agricultural product category preference and browsing behavior characteristics of customers purchasing agricultural products;

[0037] The customer purchase intention prediction module is used to construct a customer purchase intention machine learning model based on the extracted customer purchase preference characteristics of agricultural products, and predict the customer's intention to purchase agricultural products;

[0038] The optimization and improvement module is used to optimize and improve the model according to actual application conditions and feedback information, so as to improve the accuracy and effectiveness of purchase intention prediction.

[0039] Specifically, the preference feature extraction module includes: a preference feature determination unit and a preference feature calculation unit;

[0040] The preference feature determination unit is used to select the feature type in the historical data of the customer's purchase of agricultural products;

[0041] The preference feature calculation unit is used to calculate the features of the selected customer's historical data on purchasing agricultural products.

[0042] Specifically, the customer purchase intention prediction module includes a machine learning establishment unit and a customer purchase intention prediction unit;

[0043] The machine learning establishment unit is used to train a machine learning model for predicting customer purchase intention;

[0044] The customer purchase intention prediction unit is used to predict the customer's intention to purchase agricultural products.

[0045] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of an intention prediction method based on AI machine learning when executing the computer program.

[0046] A computer-readable storage medium stores computer instructions, which, when executed, execute the steps of an intention prediction method based on AI machine learning.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] 1. The present invention proposes an intention prediction method based on AI machine learning, and optimizes and improves the operating steps and processes. The method has a simple process and solves the current defect of lack of a method for predicting customers' intention to purchase agricultural products, which helps customers to better purchase agricultural products.

[0049] 2. The present invention proposes an intention prediction method based on AI machine learning, which extracts features from historical data of customers purchasing agricultural products, including the frequency, price, time, agricultural product category preferences and browsing behavior characteristics of customers purchasing agricultural products, which helps to improve the accuracy of predicting customer purchasing intentions.

[0050] 3. The present invention proposes an intention prediction method based on AI machine learning, constructs and trains a customer purchase intention machine learning model, takes into account the particularity of agricultural products, adds constraints on the quality of agricultural products in the prediction output, and utilizes the extracted historical data features of agricultural products, combined with sales and price conditions, to effectively predict the customer's intention to purchase agricultural products, thereby improving the prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a flow chart of the intention prediction method based on AI machine learning of the present invention;

[0052] Figure 2 This is an architecture diagram of the intention prediction system based on AI machine learning of the present invention;

[0053] Figure 3 This is a diagram of an electronic device for the intention prediction method based on AI machine learning of the present invention. DETAILED DESCRIPTION

[0054] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, in the description of the present invention, it should be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside" and the like indicate the orientation or position relationship based on the orientation or position relationship shown in the accompanying drawings, which is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "No. 1", "No. 2" and "No. 3" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. The present invention is further explained below in conjunction with specific embodiments.

[0055] Example 1

[0056] See also Figure 1 , an embodiment provided by the present invention: an intention prediction method based on AI machine learning, comprising the following specific steps:

[0057] Step S1: Collect historical behavior data of customers purchasing agricultural products and build a customer database;

[0058] Step S2: pre-processing the collected historical behavior data of customers purchasing agricultural products;

[0059] Step S3: extracting the preference characteristics of customers in purchasing agricultural products from the pre-processed historical behavior data of customers purchasing agricultural products, including: the frequency, price, time, category preference and browsing behavior characteristics of customers purchasing agricultural products;

[0060] Step S4: construct a customer purchase intention machine learning model based on the extracted customer preference characteristics for purchasing agricultural products, and predict the customer's intention to purchase agricultural products;

[0061] Step S5: According to the actual application situation and feedback information, the model is optimized and improved to improve the accuracy and effectiveness of purchase intention prediction to determine the degree of intention.

[0062] The historical behavior data of customers purchasing agricultural products in step S1 includes: historical record data of purchasing agricultural products, historical browsing data of agricultural products and user personal information data.

[0063] The preprocessing in step S2 includes: cleaning, deduplication, missing value processing, removing redundant and repeated data, correcting missing data, and ensuring the quality and integrity of the data.

[0064] The specific steps of step S3 are:

[0065] Step S301: Set the pre-processed historical behavior data set of customers purchasing agricultural products as , ,in, Indicates the nth agricultural product purchased by the customer in history. Indicates the price of the nth agricultural product purchased by the customer in history, Indicates the time when the customer purchased the nth agricultural product in history. Indicates the category of the nth agricultural product purchased by the customer in history;

[0066] Step S302: extract the customer's preference characteristics for purchasing agricultural products, and calculate the frequency of the customer purchasing the i-th type of agricultural products. The calculation formula is: ,in, represents the frequency of customers purchasing the i-th type of agricultural products, represents the quantity of agricultural products of category i purchased by customers, Indicates the years between the first and last purchase of agricultural products in the historical behavior data of customers purchasing agricultural products; ([] is the time difference, the data is in date format, xxxx year xx month xx day, Gs function converts the date format data into years, counts the number of days between the first and last purchase of agricultural products, and divides by 365 to get the number of years)

[0067] Step S303: Calculate the customer's category preference for purchasing the i-th category of agricultural products, and the calculation formula is:

[0068] ,

[0069] in, represents the customer's category preference for purchasing the i-th category of agricultural products, Indicates the influence weight of customers browsing agricultural products, represents the number of times customers browse the i-th category of agricultural products, Indicates the total number of times customers browsed all categories of agricultural products. represents the total number of times customers purchase the i-th type of agricultural products, Indicates the number of types of agricultural products purchased by customers;

[0070] Step S304: Calculate the influence factor of the customer's browsing behavior characteristics on the intention to purchase the i-th type of agricultural products, and the calculation formula is:

[0071] ,

[0072] in, represents the influencing factor of customer browsing behavior characteristics on the intention to purchase the i-th type of agricultural products, m represents the total number of times customers browse agricultural products, represents the number of times customers browse the i-th category of agricultural products, Indicates whether the customer purchased the i-th agricultural product that he browsed. is the judgment function. When the user buys the i-th agricultural product that he browses, =1, when the user did not purchase the i-th agricultural product that he browsed, =0.

[0073] The specific steps of step S4 are:

[0074] Step S401: selecting a machine learning model based on the extracted customer preference characteristics for purchasing agricultural products, and training the selected machine learning model;

[0075] The specific steps to train the selected machine learning model are:

[0076] Step S4011: selecting a public data set of agricultural products purchased by customers, the public data set including data of agricultural products purchased by customers in the background of the shopping website, dividing the public data set of agricultural products purchased by customers into a training set and a test set, with the training set accounting for 70% and the test set accounting for 30%, and using the training set as the input of the machine learning model;

[0077] Step S4012: setting parameters in the machine learning model and selecting a loss function to train a public data set of agricultural products purchased by customers;

[0078] Step S4013: input the test set into the machine learning model for evaluation, and test the machine learning model until the evaluation value of the prediction effect converges and remains unchanged, then stop training;

[0079] Step S402: Use the trained machine learning model to predict the customer's intention to purchase agricultural products. The specific prediction formula is:

[0080] ,

[0081] in, represents the customer's intention to purchase the jth agricultural product in the i-th category, represents the deterioration coefficient of the jth agricultural product in the i-th category during transportation, represents the initial quality of the jth agricultural product in the i-th category, represents the sales impact coefficient of the jth agricultural product in the i-th category, represents the sales volume of the jth agricultural product in the i-th category, represents the price impact coefficient of the jth agricultural product in the i-th category, represents the price of the jth agricultural product in the i-th category;

[0082] Considering that agricultural products may be affected by weather, temperature, transportation tools and other factors during transportation, the initial quality and deterioration coefficient of agricultural products are taken into account. The initial quality is set to 1. The deterioration coefficient is determined based on the three parameters of the deterioration area ratio, hardness ratio and weight loss ratio of agricultural products. The calculation method of the deterioration area is as follows: 1) Using image recognition technology, calculate the surface area of ​​agricultural products and identify the deterioration area and deterioration area of ​​agricultural products; 2) The deterioration area ratio of agricultural products is the ratio of the deterioration area of ​​agricultural products to the surface area of ​​agricultural products; the hardness value is measured by a hardness sensor, and the weight loss ratio is based on the initial weight given by the merchant and the weight of the agricultural products in the hands of the customer. The weights of the three parameters of deterioration area ratio, hardness ratio and weight loss ratio are set, and the deterioration coefficient of agricultural products is obtained after weighted summation;

[0083] Step S403: According to , predicting the agricultural products that customers want to buy.

[0084] The machine learning model in step S401 includes: any one of the deep learning prediction models.

[0085] Through this embodiment: it is possible to obtain the frequency of customers purchasing the i-th agricultural product, the category preference of customers purchasing the i-th agricultural product, and the influencing factors of customers' browsing behavior characteristics on the intention to purchase the i-th agricultural product from the historical data of customers purchasing agricultural products, and then establish a customer purchase intention machine learning model, and train and test the customer purchase intention machine learning model. The trained customer purchase intention machine learning model is used to predict the customer's purchase intention, thereby improving the accuracy of the prediction. At the same time, for the agricultural product cloud platform, it can provide customers with more effective agricultural product recommendations, and arrange them in descending order on the home page of the agricultural product cloud platform according to the customer's intention to purchase each agricultural product. For merchants, they can intuitively feel which agricultural products are popular with users and which agricultural products have defects, and it also helps merchants to actively rectify the agricultural products they sell.

[0086] Example 2

[0087] See also Figure 2 , another embodiment provided by the present invention: an intention prediction system based on AI machine learning, comprising: a historical data collection module, a historical data preprocessing module, a preference feature extraction module, a customer purchase intention prediction module and an optimization and improvement module;

[0088] The historical data collection module is used to collect historical behavior data of customers purchasing agricultural products and build a customer database;

[0089] The historical data preprocessing module is used to preprocess the collected historical behavior data of customers purchasing agricultural products;

[0090] The preference feature extraction module is used to extract the preference features of customers purchasing agricultural products from the pre-processed historical behavior data of customers purchasing agricultural products, including: the frequency, price, time, agricultural product category preference and browsing behavior characteristics of customers purchasing agricultural products;

[0091] The customer purchase intention prediction module is used to construct a customer purchase intention machine learning model based on the extracted customer purchase preference characteristics of agricultural products, and predict the customer's intention to purchase agricultural products;

[0092] The optimization and improvement module is used to optimize and improve the model according to actual application conditions and feedback information, so as to improve the accuracy and effectiveness of purchase intention prediction.

[0093] The preference feature extraction module includes: a preference feature determination unit and a preference feature calculation unit;

[0094] The preference feature determination unit is used to select the feature type in the historical data of the customer's purchase of agricultural products;

[0095] The preference feature calculation unit is used to calculate the features of the selected customer's historical data on purchasing agricultural products.

[0096] The customer purchase intention prediction module includes a machine learning establishment unit and a customer purchase intention prediction unit;

[0097] The machine learning establishment unit is used to train a machine learning model for predicting customer purchase intention;

[0098] The customer purchase intention prediction unit is used to predict the customer's intention to purchase agricultural products.

[0099] Example 3

[0100] See also Figure 3, an electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the intention prediction method based on AI machine learning are implemented.

[0101] A computer-readable storage medium stores computer instructions, which, when executed, execute the steps of an intention prediction method based on AI machine learning.

[0102] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0103] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0104] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0105] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0106] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the purpose of the present invention, which are all within the protection of the present invention.

[0107] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. The intention prediction method based on AI machine learning is characterized by: The specific steps include: Step S1: Collect historical behavior data of customers purchasing agricultural products and build a customer database; Step S2: pre-processing the collected historical behavior data of customers purchasing agricultural products; Step S3: extracting the preference characteristics of customers in purchasing agricultural products from the pre-processed historical behavior data of customers purchasing agricultural products, including: the frequency, price, time, category preference and browsing behavior characteristics of customers purchasing agricultural products; Step S4: construct a customer purchase intention machine learning model based on the extracted customer preference characteristics for purchasing agricultural products, and predict the customer's intention to purchase agricultural products; Step S5: Optimize the model according to the actual application situation and feedback information to determine the degree of intention; The specific steps of step S3 are: Step S301: Set the pre-processed historical behavior data set of customers purchasing agricultural products as , ,in, Indicates the nth agricultural product purchased by the customer in history. Indicates the price of the nth agricultural product purchased by the customer in history, Indicates the time when the customer purchased the nth agricultural product in history. Indicates the category of the nth agricultural product purchased by the customer in history; Step S302: extract the customer's preference characteristics for purchasing agricultural products, and calculate the frequency of the customer purchasing the i-th type of agricultural products. The calculation formula is: ,in, represents the frequency of customers purchasing the i-th type of agricultural products, represents the quantity of agricultural products of category i purchased by customers, Indicates the year between the first and last purchase of agricultural products in the historical behavior data of customers purchasing agricultural products; Step S303: Calculate the customer's category preference for purchasing the i-th category of agricultural products, and the calculation formula is: ， in, represents the customer's category preference for purchasing the i-th category of agricultural products, Indicates the influence weight of customers browsing agricultural products, represents the number of times customers browse the i-th category of agricultural products, Indicates the total number of times customers browsed all categories of agricultural products. represents the total number of times customers purchase the i-th type of agricultural products, Indicates the number of types of agricultural products purchased by customers; Step S304: Calculate the influence factor of the customer's browsing behavior characteristics on the intention to purchase the i-th type of agricultural products, and the calculation formula is: ， in, represents the influencing factor of customer browsing behavior characteristics on the intention to purchase the i-th type of agricultural products, m represents the total number of times customers browse agricultural products, represents the number of times customers browse the i-th category of agricultural products, Indicates whether the customer purchased the i-th agricultural product that he browsed. is the judgment function. When the user buys the i-th agricultural product that he browses, =1, when the user did not purchase the i-th agricultural product that he browsed, =0; The specific steps of step S4 are: Step S401: selecting a machine learning model based on the extracted customer preference characteristics for purchasing agricultural products, and training the selected machine learning model; Step S402: Use the trained machine learning model to predict the customer's intention to purchase agricultural products. The specific prediction formula is: ， in, represents the customer's intention to purchase the jth agricultural product in the i-th category, represents the deterioration coefficient of the jth agricultural product in the i-th category during transportation, represents the initial quality of the jth agricultural product in the i-th category, represents the sales impact coefficient of the jth agricultural product in the i-th category, represents the sales volume of the jth agricultural product in the i-th category, represents the price impact coefficient of the jth agricultural product in the i-th category, represents the price of the jth agricultural product in the i-th category; Step S403: According to , predicting the agricultural products that customers want to buy.

2. An intention prediction system based on AI machine learning, which is implemented based on the intention prediction method based on AI machine learning according to claim 1, characterized in that: include: Historical data collection module, historical data preprocessing module, preference feature extraction module, customer purchase intention prediction module and optimization and improvement module; The historical data collection module is used to collect historical behavior data of customers purchasing agricultural products and build a customer database; The historical data preprocessing module is used to preprocess the collected historical behavior data of customers purchasing agricultural products; The preference feature extraction module is used to extract the preference features of customers purchasing agricultural products from the pre-processed historical behavior data of customers purchasing agricultural products, including: the frequency, price, time, agricultural product category preference and browsing behavior characteristics of customers purchasing agricultural products; The customer purchase intention prediction module is used to construct a customer purchase intention machine learning model based on the extracted customer purchase preference characteristics of agricultural products, and predict the customer's intention to purchase agricultural products; The optimization and improvement module is used to optimize and improve the model according to actual application conditions and feedback information, so as to improve the accuracy and effectiveness of purchase intention prediction.

3. The intention prediction system based on AI machine learning as claimed in claim 2, characterized in that: The preference feature extraction module includes: a preference feature determination unit and a preference feature calculation unit; The preference feature determination unit is used to select the feature type in the historical data of the customer's purchase of agricultural products; The preference feature calculation unit is used to calculate the features of the selected customer's historical data on purchasing agricultural products.

4. The intention prediction system based on AI machine learning as claimed in claim 3, characterized in that: The customer purchase intention prediction module includes a machine learning establishment unit and a customer purchase intention prediction unit; The machine learning establishment unit is used to train a machine learning model for predicting customer purchase intention; The customer purchase intention prediction unit is used to predict the customer's intention to purchase agricultural products.

5. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the intention prediction method based on AI machine learning described in claim 1 are implemented.

6. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the computer instructions are executed by a processor, the steps of the intention prediction method according to claim 1 are implemented.

Images