AI tea product recommendation intelligent system based on large language model

Abstract

The invention relates to the technical field of product intelligent recommendation, in particular to an AI tea product recommendation intelligent system based on a large language model. The system comprises a big data intelligent platform, a product information interaction platform, a multi-style product recommendation module and a selection probability calculation module. Tea product data information in the industry is obtained through the big data intelligent platform, a distribution mechanism is formulated, the collected tea products are classified according to the distribution mechanism, newly-input tea product data are automatically classified, tea product recommendation is re-planned through the multi-style product recommendation module in combination with the recommended tea product recommendation rate, and the tea product recommendation efficiency is improved. Recommendation emphasis is re-marked according to the recommended tea product recommendation rate fed back by the selection probability calculation module, the recommendation adaptation degree is improved, the user purchase rate is improved, meanwhile, the trust degree of the user for the recommended tea products can be increased, a sales channel is provided for later new types of tea leaves, adaptive user requirements are matched, and the situation that due to the fact that the popularity of the new types of tea leaves is too low, the user experience is improved is avoided. And the marketing channel is limited.

Description

Technical Field

[0001] The present invention relates to the technical field of product intelligent recommendation, and specifically, to an AI tea product recommendation intelligent system based on a large language model. Background Art

[0002] As the main beverage source for daily consumption, tea varieties are distributed in various regions. People in different regions have their own unique views on the quality of tea. It is not easy to select tea. To obtain good tea, one needs to master a lot of knowledge, such as the grade standards of various tea varieties, prices and market conditions, as well as the evaluation and inspection methods of tea. The quality of tea is mainly judged from four aspects: color, aroma, taste, and shape. However, for ordinary tea drinkers, when buying tea, they can generally only observe the appearance and color of the dry tea and smell the dry fragrance, making it more difficult to judge the quality of tea.

[0003] Generally, the quality of tea can be identified by sensory evaluation methods. That is, through the visual, sensory, and taste organs of people, grasping the inherent essential characteristics of tea, using the methods of seeing with the eyes, smelling with the nose, touching with the hands, and tasting with the mouth, and finally comprehensively judging the quality of tea. In some regions, people prefer the mellow taste of tea, while in some regions, they prefer the fresh fragrance of tea. Therefore, different people have different views on the quality of tea. With the rise of the e-commerce industry, online tea sales are becoming more and more well-known to people. During the online tea sales process, due to the wide variety of tea varieties, different evaluation criteria for tea quality among people in different regions, and different factors for choosing tea, most users prefer to buy well-known tea varieties during the online product selection process. And if the advantages of each tea and the factors for users to purchase cannot be disassembled and analyzed, it will lead to the inability of sellers to determine the specific tea-drinking habits of current users, resulting in most of the recommended tea products not meeting the user's needs, causing new tea varieties to be unable to adapt to the market and not being able to match users with corresponding needs.

[0004] To address the above problems, there is an urgent need for an AI tea product recommendation intelligent system based on a large language model. Summary of the Invention

[0005] The purpose of the present invention is to provide an AI tea product recommendation intelligent system based on a large language model to solve the problems raised in the above background art.

[0006] To achieve the above purpose, an AI tea product recommendation intelligent system based on a large language model is provided, including a big data intelligent platform, a product information interaction platform, a multi-style product recommendation module, and a selection probability calculation module; The big data intelligent platform is used to obtain tea product data information in the industry, formulate an allocation mechanism, classify the collected tea products according to the allocation mechanism, and formulate the weight of the allocation mechanism, construct a neural learning network, and automatically classify the newly entered tea product data; The product information interaction platform is used to build a tea product data information interaction platform, which shares user tea product data information in real time through the tea product data information interaction platform, and obtains users' tea consumption habits by sharing user tea product data information; The diverse product recommendation module combines users' tea consumption habits to formulate a diverse tea product recommendation mode, and recommends tea products corresponding to the mode to users; The selection probability calculation module collects users' purchase data of recommended tea products, calculates the recommendation rate of recommended tea products, and feeds back the recommendation rate of recommended tea products to the diverse product recommendation module. The diverse product recommendation module re-plans tea product recommendations in combination with the recommendation rate of recommended tea products.

[0007] As a further improvement of this technical solution, the big data intelligent platform includes a bottom-layer logic data acquisition module, an allocation mechanism definition module, and a neural network construction module; Among them, the bottom-layer logic data acquisition module is used to obtain tea product data information in the industry; The output end of the bottom-layer logic data acquisition module is connected to the input end of the allocation mechanism definition module. The allocation mechanism definition module formulates tea classification rules in combination with tea product data information in the industry; The output end of the allocation mechanism definition module is connected to the input end of the neural network construction module. The neural network construction module is used to compare various evaluation criteria of different types of tea, divide the weight of different tea factors of different types, establish a tea classification standard database, compare the various factors of new tea, and provide a reference standard for the price of new tea.

[0008] As a further improvement of this technical solution, the method by which the neural network construction module provides a reference standard for the price of new tea includes the following steps: S1. Compare various factors of new tea through the tea classification standard database; S2. Combine the comparison results to obtain each price level that matches; S3. Make an adaptive recommendation of new tea according to each price level, and obtain the recommendation rate of the corresponding price level; S4. Arrange in descending order according to the recommendation rate values, and select the price level that matches the highest recommendation rate as the recommended price of new tea.

[0009] As a further improvement of this technical solution, the comparison of various factors of new tea in S1 includes the size, color, and shape of the tea.

[0010] As a further improvement of the technical solution, the product information interaction platform includes a tea product keyword extraction module and a chat content acquisition module. The tea product keyword extraction module is used to extract the tea keywords in the chat interaction of the user on the tea product data information interaction platform. The output end of the tea product keyword extraction module is connected to the input end of the chat content acquisition module. The chat content acquisition module combines the tea keywords in the chat interaction to determine the tea type mentioned by the current user.

[0011] As a further improvement of the technical solution, the chat content acquisition module adopts a keyword comparison algorithm, and its algorithm formula is as follows: ; ; ; Where is the set of tea keywords obtained from the chat interaction of the current user, to are the individual tea keywords obtained from the chat interaction of the current user, is the set of tea keywords for comparison, to are the individual tea keywords for comparison, is the tea keyword comparison function, Keyword comparison coincidence rate, is the keyword comparison coincidence rate threshold.

[0012] As a further improvement of the technical solution, the output end of the chat content acquisition module is connected to a tea product data sharing module, and the tea product data sharing module is used to share the tea type information mentioned by the user.

[0013] As a further improvement of the technical solution, the method for the diverse product recommendation module to formulate a diverse tea product recommendation mode includes the following steps: S10. Obtain the tea type mentioned by the current user; S20. Combine the tea classification standard database to determine the various factors of the currently obtained tea; S30. Combine the factor weights corresponding to the various factors, recommend the various teas that match the factor weights as the recommended teas, and establish a diverse tea product recommendation mode.

[0014] Compared with the prior art, the beneficial effects of the present invention: In the AI tea product recommendation intelligent system based on large language models, tea product data information in the industry is obtained through a big data intelligent platform, a distribution mechanism is formulated, the collected tea products are classified according to the distribution mechanism, the newly entered tea product data is automatically classified, and the tea product recommendations are re-planned through a multi-style product recommendation module in combination with the recommendation rate of the recommended tea products. The recommendation emphasis is re-planned according to the recommendation rate of the recommended tea products fed back by the selection probability calculation module, so as to improve the recommendation adaptability, increase the user purchase rate, and at the same time increase the user's trust in the recommended tea products, provide sales channels for new types of tea in the later stage, match the appropriate user needs, and avoid limited sales channels due to the low popularity of new types of tea. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a block diagram of the overall structural process of the present invention.

[0016] The meanings of the various reference numerals in the figure are as follows: 10. Underlying logic data acquisition module; 20. Distribution mechanism definition module; 30. Neural network construction module; 40. Tea product keyword extraction module; 50. Chat content acquisition module; 60. Tea product data sharing module; 70. Multi-style product recommendation module; 80. Selection probability calculation module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0018] Please refer to Figure 1 As shown, an AI tea product recommendation intelligent system based on large language models is provided, including a big data intelligent platform, a product information interaction platform, a multi-style product recommendation module 70, and a selection probability calculation module 80; The big data intelligent platform is used to obtain tea product data information in the industry, formulate a distribution mechanism, classify the collected tea products according to the distribution mechanism, and formulate the weight of the distribution mechanism, construct a neural learning network, and automatically classify the newly entered tea product data; The product information interaction platform is used to build a tea product data information interaction platform, which shares user tea product data information in real time through the tea product data information interaction platform, and obtains users' tea drinking habits by sharing user tea product data information; The multi-style product recommendation module 70 formulates a multi-style tea product recommendation mode in combination with users' tea drinking habits, and recommends tea products corresponding to the mode to users; The selection probability calculation module 80 collects users' purchase data of recommended tea products, calculates the recommendation rate of recommended tea products, and feeds back the recommendation rate of recommended tea products to the multi-style product recommendation module 70. The multi-style product recommendation module 70 re-plans the tea product recommendation in combination with the recommendation rate of recommended tea products.

[0019] In specific use, during the process of tea recommendation, first obtain the tea product data information in the industry through the big data intelligent platform, formulate an allocation mechanism, and classify the collected tea products according to the allocation mechanism. For example, classify the tea according to the price gradient of the tea. Among various teas at the same price level, there will also be differences. For example, the taste of the tea changes due to different production processes. Therefore, among the tea varieties in the same price batch, it is necessary to divide them according to the taste, color and shape of the tea, and formulate the weight of the allocation mechanism, that is, divide the weight of each tea grade, build a neural learning network, match the grade of the newly entered tea product data, and perform automatic classification processing; Users share user tea product data information in real time through the tea product data information interaction platform, and obtain users' tea drinking habits by sharing user tea product data information. For example, obtain keywords related to tea, tea pictures and tea knowledge popularization in users' chats, determine the type of tea being evaluated currently, and transmit the obtained tea type information to the multi-style product recommendation module 70. The multi-style product recommendation module 70 retrieves the classification data of the corresponding tea from the big data intelligent platform, obtains its tea grade distribution and corresponding weights, and conducts multi-style product recommendation according to the grade distribution and corresponding weights. For example, the tea sold at 1000 yuan per catty includes four kinds of teas, A, B, C and D. Among them, the teas belonging to the light fragrance type include tea A and tea C, the teas belonging to the mellow fragrance type include tea B and tea D, the teas belonging to the large leaf type include tea A, tea B and tea C, and the teas belonging to the small leaf type include tea D. If it is obtained that the user's tea belongs to the large leaf type light fragrance tea and the price is similar to the above tea types, and it is impossible to determine whether the user's favorite factor is due to the large leaf type or the light fragrance type or both, at this time, tea A and tea C can be recommended to the user, and the order of recommendation is based on the weight of the types to which tea A and tea C belong. For example, the light fragrance quality of tea A is higher than that of tea C. At this time, among the teas of the light fragrance type, the light fragrance quality of tea A is higher than that of tea C. Therefore, the recommendation order is to recommend tea A first and then tea C; After the tea recommendation is completed, the probability calculation module 80 collects the data of the user's purchase of the recommended tea products, calculates the recommendation rate of the recommended tea products, that is, calculates the ratio of the number of times the tea is recommended to the number of times the user purchases, and feeds back the recommendation rate of the recommended tea products to the multi-style product recommendation module 70. The multi-style product recommendation module 70 re-plans the tea recommendation in combination with the recommendation rate of the recommended tea products. For example, both tea A and tea C belong to the light fragrance type, and the weight of tea A is higher than that of tea C in this type. At the same time, both tea A and tea C belong to the large-leaf type, and the weight of tea C is higher than that of tea A in this type. When the user is in the process of recommending tea products, under the same recommended quantity state, the purchase rate of purchasing tea A is much higher than that of tea C, indicating that the user is more inclined to the light fragrance of tea at the same price. At this time, when making a second recommendation, re-plan the recommendation emphasis to further improve the recommendation adaptability, increase the user's purchase rate, and at the same time be able to increase the user's trust in the recommended tea products, provide a sales channel for new types of tea in the later stage, match the appropriate user needs, and avoid the limitation of the sales channel due to the low popularity of new types of tea.

[0020] In addition, the big data intelligent platform includes a bottom-layer logic data acquisition module 10, a distribution mechanism definition module 20, and a neural network construction module 30; Among them, the bottom-layer logic data acquisition module 10 is used to obtain the tea product data information in the industry; The output end of the bottom-layer logic data acquisition module 10 is connected to the input end of the distribution mechanism definition module 20. The distribution mechanism definition module 20 formulates the tea classification rules in combination with the tea product data information in the industry; The output end of the distribution mechanism definition module 20 is connected to the input end of the neural network construction module 30. The neural network construction module 30 is used to compare the various evaluation criteria of different types of tea, divide the weight of different tea factors of different types, establish a tea classification standard database, compare the various factors of the new tea, and provide a price reference standard for the new tea.

[0021] First, the underlying logic data acquisition module 10 acquires the tea product data information in the industry, that is, the production process, shape, size, color, and taste of tea leaves, etc. Subsequently, the acquired tea product data information is transmitted to the allocation mechanism definition module 20. The allocation mechanism definition module 20 formulates tea classification rules in combination with the tea product data information in the industry, that is, divides the price range, and divides each tea in the same price range into the same grade. Since the advantage points of the teas in the same grade are different, this is because the evaluation criteria for tea quality in each region are different. Therefore, it is necessary to compare the various evaluation criteria of different types of tea through the neural network construction module 30, divide the different tea factor weights of different types, establish a tea classification standard database, compare the various factors of the new tea, and provide a price reference standard for the new tea. For example, for two teas N and M in the same price grade, the color of tea N is the highest in the current price grade through comparison, so its color factor weight belongs to the highest grade, while the leaf size of tea M is the highest in the current price grade through comparison, so its leaf size factor weight belongs to the highest grade, serving as a reference standard for the tea classification of new teas in the later stage; It should be noted that the tea product data information acquired by the underlying logic data acquisition module 10 also includes the following data: User behavior data: including click, browse, search, purchase and other behavior data of users in applications (search engines, apps, web pages, etc.); Transaction data: including users' purchase records, purchase amounts, purchase times, etc.; User feedback data: including users' evaluations, ratings, feedback, etc. of products; Product data: including product pictures, descriptions, prices, inventory and other information; User basic information: including users' age, gender, geographical location and other basic information; Environmental data: including environmental factors such as time, location, and weather that may affect user behavior. After completing the data collection work, it is necessary to deeply process and analyze the collected data to reveal the hidden patterns and trends in the data. This process includes three steps: data cleaning, data transformation, and data normalization to ensure the quality and consistency of the data, remove redundant data therein, and then use statistical methods and machine learning algorithms to analyze the data to determine the correlations and laws between the data. For example: the relationship between purchase trends and weather, the relationship between product ratings and user purchases, build a prediction model, predict future market trends, product sales, user behavior, etc. through the prediction model, and at the same time display the prediction process through visualization technology to intuitively display the prediction results, and then feedback the prediction results to users to help users understand the meaning and possible impacts of the prediction results, and provide decision-making support for users according to the prediction results to help users make better business decisions, provide valuable insights and suggestions for decision-makers in the tea industry, so as to optimize business operations and improve efficiency and benefits.

[0022] Further, the method for the neural network construction module 30 to provide a reference standard for the price of new tea includes the following steps: S1. Compare various factors of the new tea through the tea classification standard database; S2. Combine the comparison results to obtain each price level that matches; S3. Make adaptive recommendations for new tea according to each price level and obtain the recommendation rate of the corresponding price level; S4. Arrange in descending order according to the recommendation rate values, and select the price level matching the highest recommendation rate as the recommended price of the new tea.

[0023] During the process of comparing the prices of new tea, since there are teas with the same factors in different price levels, during the process of matching the prices of new tea, there are often multiple matching prices. For example, the tea sizes are the same and the colors are the same, but due to different evaluation criteria for the taste in different regions, the prices of the two types of tea are different. At this time, if the new tea matches successfully, it will match two different prices. At this time, it is necessary to compare various factors of the new tea through the tea classification standard database, combine the comparison results, exclude the influence of irrelevant prices, and obtain each price level that matches. At this time, it is necessary to make adaptive recommendations for new tea according to each price level and obtain the recommendation rate of the corresponding price level, that is, recommend the new tea in two different price forms to the users matched in the same batch and obtain the recommendation rates of the new tea in different price states. Subsequently, obtain the matching price that best suits the new tea according to the size of the recommendation rate.

[0024] Furthermore, the various factors for comparing the new tea in S1 include the size, color, and shape of the tea. During the comparison process, randomly select new tea samples for the tea classification standard database and divide the comparison difference range. For example, during the comparison of the tea size, measure the average size of each sample and compare it with the tea sizes in each price level, and exclude the tea price levels that exceed the comparison difference range, and select the finally successfully compared price level as the preliminary price of the new tea.

[0025] Specifically, the product information interaction platform includes a tea product keyword extraction module 40 and a chat content acquisition module 50. The tea product keyword extraction module 40 is used to extract the tea keywords in the chat interaction of the user on the tea product data information interaction platform. The output end of the tea product keyword extraction module 40 is connected to the input end of the chat content acquisition module 50. The chat content acquisition module 50 combines the tea keywords in the chat interaction to determine the type of tea mentioned by the current user. During the process of the user interacting with tea information on the tea product data information interaction platform, the tea product keyword extraction module 40 extracts the tea keywords in the chat interaction of the user on the tea product data information interaction platform, such as the shape, size, and color of the tea, etc., as the basis for later identifying the current tea type. Subsequently, the chat content acquisition module 50 combines the tea keywords in the chat interaction to determine the type of tea mentioned by the current user.

[0026] In addition, the chat content acquisition module 50 adopts a keyword comparison algorithm, and its algorithm formula is as follows: ; ; ; Where is the set of tea keywords obtained from the chat interaction of the current user, to are the individual tea keywords obtained from the chat interaction of the current user, is the set of tea keywords to be compared with, to are the individual tea keywords to be compared with, is the tea keyword comparison function, is the keyword comparison coincidence rate, is the keyword comparison coincidence rate threshold. When the keyword comparison coincidence rate is less than the keyword comparison coincidence rate threshold , the tea keyword comparison function outputs 0, indicating that the tea obtained from the chat interaction of the current user does not belong to the tea to be compared with. When the keyword comparison coincidence rate is not less than the keyword comparison coincidence rate threshold , the tea keyword comparison function outputs 1, indicating that the tea obtained from the chat interaction of the current user belongs to the tea to be compared with.

[0027] Since new users don't know enough about tea, they are unable to obtain effective tea-related information and determine the tea needs of this type of user. Further, the output end of the chat content acquisition module 50 is connected to a tea product data sharing module 60. The tea product data sharing module 60 is used to share the tea type information mentioned by the user. By sharing the tea type information mentioned by the user through the tea product data sharing module 60, new users can obtain data information about tea, such as keywords related to tea, providing a data reference for obtaining the tea needs of new users in the later stage.

[0028] Furthermore, the method for the multi-style product recommendation module 70 to formulate a multi-style tea product recommendation mode includes the following steps: S10. Obtain the tea type mentioned by the current user; S20. Combine the tea classification standard database to determine various factors of the currently obtained tea; S30. Combine the factor weights corresponding to each factor, recommend various teas that match the factor weights as recommended teas, and establish a multi-style tea product recommendation mode.

[0029] In the process of making multi-style recommendations for tea products, first, it is necessary to obtain the tea type mentioned by the current user. Subsequently, combine the tea classification standard database to determine various factors of the currently obtained tea. For example, this tea belongs to the light fragrance large-leaf type, and the type of tea that the user likes is uncertain whether it is the light fragrance type or the large-leaf type or both. Therefore, combine the factor weights corresponding to each factor, recommend various teas that match the factor weights as recommended teas, and establish a multi-style tea product recommendation mode. When making recommendations, recommend the light fragrance tea type with the same factor weight as this tea and the large-leaf tea with a larger factor weight. Finally, determine which factor of the tea that the user is more inclined to through the recommended rate feedback, so as to further improve the recommendation adaptability.

[0030] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art of this industry should understand that the present invention is not limited by the above embodiments. The above embodiments and the descriptions in the specification are only preferred examples of the present invention and are not used to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. An AI tea product recommendation system for automatic analysis of tea product data, characterized in that: It includes a user behavior data acquisition module (10), a user transaction data acquisition module (20), a purchased tea product recommendation module (30), a user feedback data acquisition module (40), and a tea product recommendation model construction module (50); Among them, the user behavior data acquisition module (10) is used to acquire the tea product data information concerned by the user on the network; The user transaction data acquisition module (20) is used to acquire the transaction data of the tea products purchased by the user; The purchased tea product recommendation module (30) combines the transaction data of the tea products purchased by the user, acquires the tea product amounts purchased by the user in different time states, and recommends tea products of the same tea product amount but different types to the user according to the current tea product amount; The user feedback data acquisition module (40) is used to acquire the types and feedback data of the recommended tea products purchased by the user, and transmits the types and feedback data of the recommended tea products purchased by the user to the purchased tea product recommendation module (30); The tea product recommendation model construction module (50) combines the tea product recommendation types, the types of tea products purchased by the user, and the corresponding feedback data information, acquires the influence weights of each tea product type on the recommended tea products purchased by the user in the tea product amounts at different stages, and establishes a tea product recommendation model according to the proportion of the influence weights, predicts the tea purchase trends at different purchased tea product amount levels, and plans the corresponding tea product type recommendations according to the tea purchase trends; The method for establishing a tea product recommendation model in the tea product recommendation model construction module (50) includes the following steps: S1. The planning unit counts the time T, determines the recommended tea varieties with the same purchase amount, and counts the quantity of the recommended tea varieties ; S2. Recommend the types of recommended tea products with the same purchase amount to the users in the same tea product purchase amount level; S3. During the unit statistical time T, count the purchase quantities of different recommended tea varieties = the number of purchasing users the number of purchases ; S4. Combine the feedback data of users in the same tea purchase amount class to obtain the praise rates of different recommended tea varieties , the purchase quantity of the recommended tea varieties that received praise, and the total purchase quantity of the recommended tea varieties purchased by users in the same tea purchase amount class within the unit statistical time T; S5. Calculate the weights of different recommended tea varieties , where is the purchase quantity of the currently calculated recommended tea variety, is the time period constant; S6. According to the weights of different recommended tea varieties, sort the recommended tea varieties with the same purchase amount according to the weight size, and use it as the priority recommendation order for the same purchase amount; S7. Repeat the above steps, count the changes in the weight sizes of the tea products with the same purchase amount in different time periods, and combine the environmental changes in different time periods to predict the environmental factors and influence trends affecting the recommended tea products for purchase.

2. The AI tea product recommendation system for automatic analysis of tea product data according to claim 1, wherein: The tea product data information acquired by the user behavior data acquisition module (10) includes application browsing, keyword search, and tea product click-through rate.

3. The AI tea product recommendation system for automatically analyzing tea product data according to claim 1, characterized in that: The transaction data in the user transaction data acquisition module (20) includes purchase records, purchase times, and purchase amounts.

4. The AI tea product recommendation system for automatic analysis of tea product data according to claim 1, characterized in that: The purchased tea product recommendation module (30) includes a tea product amount level positioning unit (310), a tea product data information storage repository (320), and a tea product type positioning and recommendation unit (330); The tea product amount level positioning unit (310) combines the transaction data of the tea products purchased by the user to position the tea product purchase amount level of the current user; The tea product data information storage repository (320) collects the tea product purchase information in the industry, establishes a tea product storage repository, and stores the purchase amount information of each type of tea product; The tea product type positioning and recommendation unit (330) combines the positioned tea product purchase amount level of the user, and retrieves different types of tea products with the same tea product amount from the tea product storage repository as the recommended tea products.

5. The AI tea product recommendation system for automatic analysis of tea product data according to claim 4, characterized in that: The tea product types in the tea product data information storage repository (320) include tea product taste, tea product craftsmanship, and tea product year.

6. The AI tea product recommendation system for automatic analysis of tea product data according to claim 1, characterized in that: The user feedback data acquisition module (40) includes a recommended tea quantity acquisition unit (410), a tea evaluation data acquisition unit (420), and an evaluation difference analysis unit (430). The recommended tea quantity acquisition unit (410) is used to acquire the types of recommended tea products purchased by the user and the corresponding tea purchase quantities. The tea evaluation data acquisition unit (420) is used to acquire the evaluation data after the user purchases the tea products. The evaluation difference analysis unit (430) combines the evaluation data to distinguish between negative reviews and excellent reviews.

7. The AI tea product recommendation system for automatic analysis of tea product data according to claim 1, characterized in that: The output end of the tea product recommendation model construction module (50) is connected to an environmental factor data acquisition module (60). The environmental factor data acquisition module (60) is used to acquire the tea product production environmental factors, including the tea production time, the origin, and the local weather of the origin.

Images