Personalized recommendation method and system for library books

Abstract

The invention relates to the technical field of book recommendation, and discloses a personalized recommendation method and system for library books, and the method comprises the steps: obtaining book RFID tag circulation data and user behavior data in a book borrowing process of a user; performing grid division on the library according to a preset grid size, analyzing the interest degree of a user for books in each grid region, and constructing a book interest thermodynamic diagram; according to the book interest thermodynamic diagram, hot spot areas with interest degree values larger than a preset interest degree threshold value are extracted, recommended books are screened out from the hot spot areas through a preset book recommendation model, and a first recommended book set is obtained; based on the borrowing condition of the user for the first recommended book set, optimizing the recommended books through a preset recommendation optimization model to obtain a second recommended book set; the method can adapt to the dynamic change of the interest of the user in the book recommendation process, improves the accuracy and timeliness of book recommendation, and achieves the personalized recommendation of the user.

Description

Technical Field

[0001] The present invention relates to the technical field of book recommendation, and particularly to a personalized recommendation method and system for library books. Background Art

[0002] With the rapid development of information technology, as an important place for knowledge dissemination, the library needs to recommend books to users based on their interests when providing personalized book recommendations. The interests of users are time-sensitive. Currently, when recommending books to users, it is mainly based on book categories, and the recommendation effect is not good, and personalized recommendation for each user cannot be achieved.

[0003] The prior art has the following problems: analyzing users' interests based on users' borrowing data or the content attributes of books, the data is single, and users' interest preferences cannot be comprehensively and accurately analyzed; only analyzing and recommending based on book categories or content, ignoring the logical arrangement of books in the library, directly analyzing each book for recommendation, the analysis process is complex, and the recommendation efficiency is low; after recommending books, there is a lack of analysis and prediction of the real-time changes in users' interests, and book recommendations cannot be combined with the trend of changes in users' interests. To solve at least one of the above problems, the present invention proposes a personalized recommendation method and system for library books. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the main purpose of the present invention is to provide a personalized recommendation method and system for library books, which can effectively solve the problems in the background art. The specific technical solutions of the present invention are as follows: A personalized recommendation method for library books, comprising: Obtaining the book RFID tag transfer data and user behavior data during the user's book borrowing process; Combining the book RFID tag transfer data and user behavior data, dividing the library into grids according to a preset grid size, analyzing the degree of interest of users in the books in each grid area, and constructing a book interest heat map; According to the book interest heat map, extracting hot areas where the degree of interest value is greater than a preset interest degree threshold, and screening out recommended books through a preset book recommendation model in combination with the hot areas to obtain a first set of recommended books; Based on the borrowing situation of users for the first set of recommended books, optimizing the recommended books through a preset recommendation optimization model to obtain a second set of recommended books, so as to perform personalized recommendation for library books.

[0005] Specifically, by combining the book RFID tag transfer data and user behavior data, the library is divided into grids according to a preset grid size, and the interest level of users in the books within each grid area is analyzed to construct a book interest heat map, including: The library is divided into grids according to a preset grid size to obtain multiple grid areas; By combining the book RFID tag transfer data and user behavior data, the interest level of users in the books within each grid area is analyzed through a preset interest analysis model to obtain the interest level value of each grid area; Taking the grid area where the book borrowed by the user is located as the center, and combining the interest level value to spread to the surrounding grid areas to construct a book interest heat map.

[0006] Specifically, the step of combining the book RFID tag transfer data and user behavior data, analyzing the interest level of users in the books within each grid area through a preset interest analysis model, and obtaining the interest level value of each grid area includes: According to the book RFID tag transfer data, the first grid area where the book borrowed by the user is located is extracted, and the time difference between the borrowing and returning of the book is analyzed to obtain the book borrowing duration; According to the user behavior data, the book access situation of the user is analyzed through a preset interest analysis model to obtain the user behavior analysis result; Combining the book borrowing duration and the user behavior analysis result to determine the interest level value of the first grid area; Through the content similarity between the books in the grid area, and combining the interest level value of the first grid area, the interest level value of each grid area is calculated.

[0007] Specifically, the step of taking the grid area where the book borrowed by the user is located as the center, combining the interest level value to spread to the surrounding grid areas to construct a book interest heat map includes: Taking the first grid area where the book borrowed by the user is located as the central grid area; According to the interest level value of each grid area and the distance from the central grid area, the heat value of each grid area is calculated. When there are multiple central grid areas, the heat values calculated according to each central grid area are superimposed; Taking the central grid area as the center, spreading to the surrounding grid areas according to the heat value to construct a book interest heat map.

[0008] Specifically, according to the book interest heat map, hot areas with interest level values greater than a preset interest level threshold are extracted, and recommended books are screened out through a preset book recommendation model in combination with the hot areas to obtain a first set of recommended books, including: According to the book interest heat map, extract the second grid areas where the interest degree values are greater than the preset interest degree threshold, and merge adjacent second grid areas to obtain multiple hot areas; Within each hot area, analyze the interests of users in different books through a preset book recommendation model to screen and recommend books, and obtain the first area recommended book set; By analyzing the book relevance between the hot areas and non-hot areas, screen and recommend books within the non-hot areas to obtain the second area recommended book set; Combine the first area recommended book set and the second area recommended book set to obtain the first recommended book set.

[0009] Specifically, within each hot area, analyzing the interests of users in different books through a preset book recommendation model to screen and recommend books, and obtaining the first area recommended book set includes: Within each hot area, analyze the attributes of each book to obtain book attribute characteristics; According to the user behavior data, combine the book attribute characteristics, and calculate the interest matching degree of the user in each book through a preset book recommendation model; Screen out the preset first number of recommended books in descending order of the interest matching degree to obtain the first area recommended book set.

[0010] Specifically, by analyzing the book relevance between the hot areas and non-hot areas, screening and recommending books within the non-hot areas to obtain the second area recommended book set includes: Calculate the similarity between the book attribute characteristics of the books within the non-hot areas and the book attribute characteristics of the books within the hot areas to obtain the book similarity; Combine the distance between the non-hot areas and the hot areas and the corresponding book similarity to calculate the relevance between the non-hot areas and the hot areas; Screen out the preset second number of non-hot areas in descending order of the relevance to obtain the candidate area set; In each non-hot area in the candidate area set, screen out the preset third number of recommended books in descending order of the interest matching degree of the user in the books within the non-hot areas to obtain the second area recommended book set.

[0011] Specifically, based on the borrowing situation of the user for the first recommended book set, optimizing the recommended books through a preset recommendation optimization model to obtain the second recommended book set for personalized recommendation of library books includes: According to the borrowing situation of the user for the books in the first recommended book set, analyze the interest change situation of the user to obtain an interest vector; Combined with the interest vector, the recommended books are optimized through a preset recommendation optimization model to obtain a second set of recommended books for personalized recommendation of library books.

[0012] Specifically, combined with the interest vector, the recommended books are optimized through a preset recommendation optimization model to obtain a second set of recommended books for personalized recommendation of library books, including: Based on the interest vector and combined with user behavior data, the interest migration situation of the user is predicted through a preset recommendation optimization model to obtain an interest migration prediction vector; Based on the interest migration prediction vector, the books matching the interest migration prediction vector are selected as the optimized book set; The recommended books are optimized through the optimized book set to obtain a second set of recommended books for personalized recommendation of library books.

[0013] A personalized recommendation system for library books, used to implement the personalized recommendation method for library books described above, includes: A data acquisition module, which acquires the book RFID tag transfer data and user behavior data during the user's book borrowing process; A book interest heat map construction module, which combines the book RFID tag transfer data and user behavior data, divides the library into grids according to a preset grid size, analyzes the user's interest level in the books in each grid area, and constructs a book interest heat map; A book recommendation module, which extracts the hot areas with the interest level value greater than a preset interest level threshold according to the book interest heat map, and combines the hot areas to screen out recommended books through a preset book recommendation model to obtain a first set of recommended books; A recommended book optimization module, which optimizes the recommended books through a preset recommendation optimization model based on the borrowing situation of the user for the first set of recommended books to obtain a second set of recommended books for personalized recommendation of library books.

[0014] Compared with the prior art, the present application has the following beneficial effects: This application combines the circulation data of library RFID tags and user behavior data, divides the library into grids according to a preset grid size, analyzes the interest levels of users in different areas, constructs a book interest heat map, and based on the book interest heat map and the changes in user interests, recommends corresponding books and optimizes them. By combining multi-dimensional data, the activity trajectories and interest preferences of users in the library can be accurately analyzed. Combining user interests with the library space structure and considering the spatial propagation characteristics of user interests and the relevance of book placement in the library can improve the efficiency of book recommendations, expand the recommendation scope, enhance the comprehensiveness of recommended books, and by optimizing the recommended books, it can adapt to the dynamic changes of user interests and improve the accuracy and timeliness of recommended books. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a flowchart of a personalized book recommendation method for a library in Embodiment 1 of the present invention; Figure 2 is a schematic diagram of a book interest heat map in Embodiment 1 of the present invention; Figure 3 is a schematic diagram of the screening of hot spots in the book interest heat map in Embodiment 1 of the present invention; Figure 4 is a schematic diagram of the structure of a personalized book recommendation system for a library in Embodiment 2 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following detailed description of the specific embodiments of the present invention will be made with reference to the accompanying drawings of the specification.

[0017] Many specific details are set forth in the following description in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0018] Secondly, the so-called "one embodiment" or "embodiment" refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that excludes other embodiments.

[0019] Embodiment 1 This embodiment provides a personalized book recommendation method for a library, as Figure 1 shown, a personalized book recommendation method for a library includes: S101. Obtain the book RFID tag transfer data and user behavior data during the user's book borrowing process; S102. Combine the book RFID tag transfer data and user behavior data, divide the library into grids according to the preset grid size, analyze the user's interest level in the books in each grid area, and construct a book interest heat map; S103. According to the book interest heat map, extract the hot areas where the interest level value is greater than the preset interest level threshold, and combine the hot areas to screen out recommended books through the preset book recommendation model to obtain the first recommended book set; S104. Based on the borrowing situation of the user for the first recommended book set, optimize the recommended books through the preset recommendation optimization model to obtain the second recommended book set for personalized recommendation of the library books.

[0020] With the continuous increase in the library's collection and the improvement of users' demand for personalized services, the existing book recommendation methods are difficult to meet the changing needs and interests of users; this technical solution combines the book RFID tag transfer data and user behavior data during the user's book borrowing process, divides the library into grids according to the preset grid size, calculates the interest level value of each grid area based on the obtained data, and constructs a book interest heat map. Based on the book interest heat map, extract the hot areas where the interest level value is greater than the preset threshold, and screen out recommended books from the hot areas and non-hot areas respectively through the preset book recommendation model to form the first recommended book set. Based on the borrowing situation of the user for the first recommended book set, analyze the change of the user's interest, and optimize the recommended books through the preset recommendation optimization model to obtain the second recommended book set for personalized recommendation of the library books.

[0021] This embodiment integrates multi-dimensional data for user interest analysis, can obtain accurate user interest analysis results, recommend books that meet the user's interests, and improve the accuracy and relevance of recommendations; combining the library space area division results and the book interest heat map can quickly screen out the books that the user is interested in, and analyze the change of the user's interest according to the borrowing situation of the user for the recommended books, and optimize the recommended books in real time, which can improve the matching degree between the recommended books and the user's interests, realize real-time personalized recommendation of books, attract users to borrow more books, and improve the circulation of library books.

[0022] In this embodiment, first, the book RFID tag transfer data and user behavior data during the user's book borrowing process are obtained. The book RFID tag transfer data records the location changes and borrowing situations of books in the library, and the user behavior data records the user's activities in the library, including data such as the stay time in each area and retrieval records of the user. By obtaining multi-dimensional data, it is possible to comprehensively understand the interaction between the user and the book, including the borrowing transfer of the book, the activity trajectory and behavior preferences of the user in the library, providing data support for analyzing the books that the user is interested in.

[0023] Exemplarily, RFID readers are installed at positions such as the entrances and exits of the library and on the bookshelves. When a book with an RFID tag passes by the reader, the reader will read the tag information, including the book identification, the book movement time, etc., and transmit the read information to the database of the library management system in real time. For example, when a user borrows a book, the RFID reader at the entrance and exit will record the lending time and the book number of the book; when the book is returned to the bookshelf, the RFID reader on the bookshelf will record the return time and the location of the bookshelf where the book is located.

[0024] Exemplarily, sensors are set in the library, including sensors such as infrared sensors and Wi-Fi probes. The infrared sensor can detect the stay time of the user in a certain area. When the user enters or leaves the area where the infrared sensor is installed, the sensor will record the corresponding time points and calculate the stay duration of the user in this area; the Wi-Fi probe can detect the MAC address of the user's device and analyze the movement trajectory of the user's device in the library. At the same time, the library management system can also record behavior data such as the keywords input by the user when retrieving books and the book pages browsed by the user.

[0025] Specifically, according to the preset grid size, the library space is divided into grids to obtain multiple grid areas. By combining the book RFID tag transfer data and user behavior data, the interest degree of the user in the books in each grid area is analyzed, and a book interest heat map is constructed based on the interest degree. By analyzing indicators such as the borrowing frequency and user stay time of the books in each grid area, the interest degree of the user in the books in this grid area is calculated, and a book interest heat map is constructed according to different interest degree values. The areas with a high interest degree are displayed as areas with a darker color on the heat map, and the areas with a low interest degree have a lighter color. By constructing the book interest heat map, the interest distribution of the user in different areas of the library can be intuitively displayed, and the book area that the user is interested in can be quickly searched when recommending books to the user.

[0026] Specifically, according to the interest degree value of each grid area, grid areas with interest degree values greater than a preset interest degree threshold are extracted from the book interest heat map as hot areas. Combining the hot areas, recommended books are screened out through a preset book recommendation model to obtain a first set of recommended books; first, books related to the books borrowed by the user are analyzed in the hot areas, and recommended books are screened out from the hot areas. By analyzing the distance and book association between the hot areas and non-hot areas, recommended books are screened out from the non-hot areas, and the recommended books in the hot areas and non-hot areas are combined to form a first set of recommended books; by extracting hot areas and screening out recommended books in the hot areas, the high-interest areas of the user can be quickly located. Extracting recommended books in the high-interest areas can improve the efficiency of the book recommendation process. Screening out recommended books in the non-hot areas can improve the comprehensiveness of the recommended books, increase the categories of the recommended books, and improve the matching degree between the recommended books and the user's interests.

[0027] Preferably, according to the borrowing situation of the user for the books in the first set of recommended books, the recommended books are optimized through a preset recommendation optimization model to obtain a second set of recommended books; by analyzing the borrowing situation of the user for the recommended books, the change of the user's interest is analyzed. According to the change of the user's interest, the books in the first set of recommended books are adjusted and optimized, the books with lower user interest are removed, and books that meet the user's real-time interest are added in combination with the change of the user's interest to obtain a second set of recommended books, realizing personalized book recommendation; by analyzing the borrowing feedback situation of the user for the recommended books and optimizing the recommended books, the recommendation strategy can be continuously improved in combination with the user's real-time interest, the recommended books can be adjusted in time, and the accuracy and effectiveness of the recommendation can be improved, providing the user with a more personalized recommendation that meets the current needs.

[0028] This application combines the book RFID tag transfer data and user behavior data, divides the library into grids according to a preset grid size, analyzes the interest degree of users in different areas, constructs a book interest heat map, and recommends corresponding books and optimizes them based on the book interest heat map and the change of the user's interest; combining multi-dimensional data can accurately analyze the user's activity trajectory and interest preference in the library. Combining the user's interest with the library space structure and considering the spatial propagation characteristics of the user's interest and the relevance of the book placement in the library can improve the efficiency of book recommendation, expand the recommendation scope, improve the comprehensiveness of the recommended books, and adapt to the dynamic change of the user's interest by optimizing the recommended books, improving the accuracy and timeliness of the recommended books.

[0029] Further, by combining the circulation data of book RFID tags and user behavior data, the library is divided into grids according to a preset grid size, and the interest degree of users in the books in each grid area is analyzed to construct a book interest heat map, including: S201. Divide the library into multiple grid areas according to a preset grid size; S202. Combine the circulation data of book RFID tags and user behavior data, and analyze the interest degree of users in the books in each grid area through a preset interest analysis model to obtain the interest degree value of each grid area; S203. With the grid area where the borrowed book is located as the center, combine the interest degree value to spread to the surrounding grid areas to construct a book interest heat map.

[0030] In this embodiment, first, the library is divided into multiple grid areas according to a preset grid size; according to the actual area of the library, the bookshelf layout, and the convenience of data processing, the side length size of the grid is determined, and the library space is divided into multiple grid areas according to the corresponding grid size. Through grid division, the books in each grid area can be analyzed, improving the efficiency of book category analysis and book recommendation.

[0031] Specifically, according to the circulation data of book RFID tags and user behavior data, the interest degree of users in the books in each grid area is analyzed through a preset interest analysis model to obtain the interest degree value of each grid area; the collected circulation data of book RFID tags and user behavior data are sorted according to the grid area, and the total number of borrowing times, average borrowing duration of the books in each grid area, as well as the total stay time of users in this area, average number of books browsed, etc. are respectively counted; combined with the content similarity between the grid area where the user borrows the book and the books in other grid areas, the interest degree value of each grid area is calculated; by calculating the interest degree value, the interest situation of users in the books in each grid area can be quickly analyzed, improving the reliability and accuracy of the analysis results, and the calculated interest degree value provides data support for book recommendation.

[0032] Preferably, taking the grid area where the user-borrowed book is located as the central grid area, spreading around according to the interest degree value of each grid area, a book interest heat map is constructed; taking the grid area where the user-borrowed book is located as the center, combining the interest degree value of each grid area and the distance from the central grid area, the heat value of each grid area is calculated. When there is an overlapping area in the spreading area of the central grid area, the heat values of the overlapping areas are superimposed to obtain the heat value of each grid area. Taking the central grid area as the center, spreading to the surrounding grid areas according to the heat value, a book interest heat map is obtained; setting the heat value range, mapping heat values in different ranges to different colors. For example, heat values between 0-10 are set to blue, between 10-20 are set to green, between 20-30 are set to yellow, and above 30 are set to red to generate a book interest heat map. Constructing a book interest heat map through area spreading can comprehensively display the distribution of user interests in the library space, and when recommending books to users, it can quickly locate the areas of user interest.

[0033] Further, combining the book RFID tag transfer data and user behavior data, analyzing the user's interest degree in the books in each grid area through a preset interest analysis model, and obtaining the interest degree value of each grid area, including: S301. According to the book RFID tag transfer data, extract the first grid area where the user-borrowed book is located, analyze the time difference between the book borrowing and returning times, and obtain the book borrowing duration; S302. According to the user behavior data, analyze the user's book access situation through a preset interest analysis model to obtain the user behavior analysis result; S303. Combine the book borrowing duration and the user behavior analysis result to determine the interest degree value of the first grid area; S304. Calculate the interest degree value of each grid area through the content similarity between the books in the grid area, combined with the interest degree value of the first grid area.

[0034] In this embodiment, according to the book RFID tag transfer data, taking the grid area where the user-borrowed book is located as the first grid area, by calculating the time difference between the operation times recorded by the RFID reading and writing device when the book is borrowed and returned, the book borrowing duration is obtained; calculating the borrowing duration can reflect the user's investment in the book, and the longer the borrowing time, the higher the user's interest in the book, providing a data basis for analyzing the user's interest degree in the books in the first grid area.

[0035] Specifically, analyze the user's book access situation based on user behavior data, analyze the user's book retrieval keywords, browsing records, etc. through a preset interest analysis model to obtain the user behavior analysis results, and determine the types of books that the user is interested in; according to the business requirements of the library and the understanding of user behavior, set corresponding user interest judgment rules in the preset behavior analysis model. For example, if the user retrieves books on the same topic more than 3 times, it indicates that the user is interested in books on that topic; input the sorted user behavior data into the behavior analysis model and analyze it according to the preset interest judgment rules to generate corresponding user behavior analysis results for each user; by analyzing the user behavior data, the user's interested books can be analyzed from the user's operation behavior. Compared with analyzing only based on book borrowing data, this solution can more comprehensively understand the user's activities and interest needs in the library, thereby improving the accuracy and pertinence of book recommendations.

[0036] Specifically, combine the book borrowing duration and the user behavior analysis results to calculate the interest degree value of the first grid area; set corresponding weights for the book borrowing duration and the user behavior analysis results according to the book interest analysis requirements, perform weighted summation on the book borrowing duration and the user behavior analysis results, and calculate the interest degree value of the first grid area; the book borrowing duration reflects the user's interest in books from the time perspective, and the user behavior analysis results reflect the user's interest from the behavior perspective. Combining the two can more comprehensively and accurately analyze the user's interest degree in the books of the first grid area.

[0037] Preferably, after calculating the interest degree value of the first grid area, by calculating the content similarity between the books in the grid area, the interest degree value of the first grid area can be used to analyze the interest degree of the books in other grid areas; use text analysis techniques, including algorithms such as TF-IDF algorithm and cosine similarity algorithm, to analyze the content of the books between each grid area, calculate the content similarity between the books in different grid areas. Specifically, vectorize the text information such as the title, abstract, and keywords of the books, and then calculate the cosine value of the angle between the book vectors in the grid area through the cosine similarity algorithm. The closer this value is to 1, the higher the content similarity of the books; at the same time, set the weight of the content similarity in the propagation of the interest degree value according to the distance of the library grid area; for each grid area, perform weighted summation on the content similarity and the set propagation weight to calculate the interest degree value of each grid area; by analyzing the content similarity between the books in the grid area to calculate the interest degree value of each grid area, comprehensively consider the relevance of the book content and the distance of the grid area, avoid the isolated analysis of each grid area, and can comprehensively evaluate the user's interest distribution in the books of each grid area in the library, providing an accurate reference for book recommendations.

[0038] Furthermore, centered on the grid area where the book borrowed by the user is located, combined with the interest degree value, it spreads to the surrounding grid areas to construct a book interest heat map, including: S401. Take the first grid area where the book borrowed by the user is located as the central grid area; S402. Calculate the heat value of each grid area according to the interest degree value of each grid area and the distance from the central grid area. When there are multiple central grid areas, the heat values calculated according to each central grid area are superimposed; S403. Centered on the central grid area, spread to the surrounding grid areas according to the heat value to construct a book interest heat map.

[0039] In this embodiment, the first grid area where the book borrowed by the user is located is taken as the central grid area. The first grid area centrally reflects the user's current reading interest. Setting the first grid area as the central grid area can spread to the surrounding areas with this interesting book as the core. When the library books are placed, the books in adjacent grid areas have a certain relevance in terms of theme and category. Spreading from the first grid area as the core to the surrounding areas can make the analysis of the user's interesting books more targeted, avoid blindly analyzing the entire library grid area, be able to associate with the areas related to the user's current interest, improve the efficiency of user interest analysis, and provide a basis for constructing the heat map.

[0040] Specifically, according to the interest degree value of each grid area and the distance between each grid area and the central grid area, calculate the heat value of each grid area. When a grid area is affected by multiple central grid areas, superimpose the heat values calculated for each central grid area to calculate the heat value of the corresponding grid area; the interest degree value reflects the user's interest in the books in the grid area, and the distance between the grid area and the central grid area reflects the spatial correlation degree of different areas. When the distance between the grid area and the central grid area is closer, it means that the user is more interested in the books in this area and the browsing times of the books in this area are more during the process of borrowing books. By comprehensively considering the interest degree value and the distance from the central grid area, after taking the inverse of the distance and performing weighted summation with the interest degree value, calculate the heat value of each grid area. When there are multiple central grid areas, add the heat values calculated for each central grid area by the grid area to obtain the heat value of each grid area; by comprehensively calculating the heat value considering the interest degree value and the distance, taking into account the user interest intensity and spatial correlation factors, the calculated heat value can more accurately reflect the distribution of user interest in the library space, improve the comprehensiveness and accuracy of heat value calculation, and provide a data basis for constructing a book interest heat map.

[0041] Preferably, the thermal value calculated based on each grid area is diffused to the surrounding grid areas with the central grid area as the center, and a corresponding relationship between the thermal value and the color is established; for example, the thermal value is set to blue in the range of 0-20, green in the range of 21-40, yellow in the range of 41-60, orange in the range of 61-80, and red in the range of 81-100, and a book interest heat map is constructed based on the thermal value and the corresponding color; by constructing a book interest heat map, the distribution of user interests can be intuitively displayed, and based on the book interest heat map, the book area of interest can be quickly screened out, thereby improving the efficiency of book recommendation.

[0042] like Figure 2 As shown in the figure, the thickness of the area box represents the depth of the heat map color, where the thicker the area box, the darker the color in the corresponding area. In the figure, with the central grid area as the center, red area, orange area and yellow area are drawn as an example to represent the regional diffusion of the book interest heat map.

[0043] Furthermore, based on the book interest heat map, hot spots with interest levels greater than a preset interest level threshold are extracted, and recommended books are screened out using a preset book recommendation model based on the hot spots to obtain a first recommended book set, including: S501: extracting second grid areas whose interest levels are greater than a preset interest level threshold according to the book interest heat map, and merging adjacent second grid areas to obtain multiple hotspot areas; S502: In each hotspot area, analyzing the user's interest in different books by using a preset book recommendation model to filter recommended books and obtain a first area recommended book set; S503, by analyzing the book relevance between the hotspot area and the non-hotspot area, screening recommended books in the non-hotspot area to obtain a second area recommended book set; S504: Combine the first region recommended book set and the second region recommended book set to obtain a first recommended book set.

[0044] like Figure 3As shown in the figure, in this embodiment, according to the book interest heat map, the second grid areas with interest degree values greater than the preset interest degree threshold are extracted to obtain multiple hot areas. The areas within the dashed boxes in the figure are selected as hot areas. In each hot area, using the preset book recommendation model, the interests of users in different books are analyzed, the eligible books are screened out to obtain the first regional recommended book set, the relevance of books between the hot areas and non-hot areas is analyzed, combined with the users' interested books, the books with interest potential are screened out in the non-hot areas to obtain the second regional recommended book set, and the first regional recommended book set and the second regional recommended book set are de-duplicated and combined to obtain the first recommended book set; by combining the hot areas and non-hot areas to screen the recommended books, the book interest preferences of users can be accurately grasped, books highly matching the users' interests can be recommended, the recommendation of irrelevant books can be reduced, and the accuracy of the recommendation can be improved. At the same time, by analyzing the relevance of books between regions to screen potential interested books in non-hot areas, the types and scopes of the recommended books are enriched, diverse reading options are provided for users, and the reading interests of users are stimulated.

[0045] In this embodiment, first, in the book interest heat map, the second grid areas with interest degree values greater than the preset interest degree threshold are screened out, and the adjacent second grid areas are merged to obtain the hot areas. The higher the interest degree value, the higher the interest degree of the user in the books in the grid area. By screening out the areas with high user interest degree, the areas where the user interest is concentrated can be quickly located, making the book recommendation process more targeted and improving the efficiency and accuracy of book recommendation.

[0046] Specifically, in each hot area, the interests of users in the books in the hot area are analyzed through the preset book recommendation model. The model comprehensively considers interest factors such as the borrowing history of the books, the historical borrowing behavior of the users, the ratings of the books, and the content tags, analyzes the matching degree of each book with the users' interests, and screens out the books with the highest degree of user interest to form the first regional recommended book set; by screening books in the hot areas, the interest needs of users in the areas where the interests are concentrated can be fully analyzed, and the matching degree of the recommended books with the users' interests can be improved. Compared with random recommendation or non-discriminatory recommendation, this solution can more accurately meet the reading interest needs of users and improve the acceptance rate and borrowing rate of users for the recommended books.

[0047] Preferably, the books in the non-hotspot area are relevant to those in the hotspot area in terms of content. By analyzing the relevance of books between the hotspot area and the non-hotspot area, recommended books are screened out in the non-hotspot area to obtain the recommended book set for the second area; by screening recommended books in the non-hotspot area, considering that there are various association relationships such as content association, theme association, and author association among the books in the library, there will also be associations between books in areas with different popularity levels. By analyzing the relevance of books between the hotspot area and the non-hotspot area, potential books related to the user's interests in the non-hotspot area can be screened out; for example, if a certain type of book is popular in the hotspot area, books by the same author or of the same theme in the non-hotspot area will also meet the user's interests. Screening out these books to form the recommended book set for the second area can enrich the recommended content, expand the user's reading selection range, and increase the borrowing probability of the recommended books being borrowed by the user.

[0048] Specifically, the recommended book set for the first area and the recommended book set for the second area are combined to obtain the first recommended book set. The recommended book set for the first area includes books in the hotspot area that highly match the user's interests, and the recommended book set for the second area includes books in the non-hotspot area that the user may be interested in. Combining the two can integrate the book resources in the user interest concentration area and the potential association area to obtain a richer and more comprehensive first recommended book set, expanding the breadth of recommendations while ensuring the accuracy of the recommended books, and providing users with diverse reading choices; after removing duplicates and merging the books in the recommended book set for the first area and the recommended book set for the second area, the recommended books are sorted from high to low according to the interest degree value of the grid area where the recommended books are located to obtain the first recommended book set. By combining the recommended books in the hotspot area and the recommended books in the non-hotspot area, it can not only meet the user's current interest needs but also guide the user to discover potential reading interests, enhance the user's reading experience in the library, and achieve personalized book recommendations for the user.

[0049] Furthermore, within each hotspot area, recommended books are screened by analyzing the user's interests in different books through a preset book recommendation model to obtain the recommended book set for the first area, including: S601. Analyze the attributes of each book within each hotspot area to obtain the book attribute characteristics; S602. According to the user behavior data, combined with the book attribute characteristics, calculate the interest matching degree of the user for each book through a preset book recommendation model; S603. Screen out the preset first quantity of recommended books from high to low according to the interest matching degree to obtain the recommended book set for the first area.

[0050] In this embodiment, within each hot spot area, the attributes of each book are analyzed respectively to obtain book attribute features. The book attributes include information such as book category, author, publication year, content summary, keywords, reader ratings, borrowing times, etc. The book attribute information of each book is quantified according to a preset quantization standard to obtain the book attribute features of each book. By analyzing the book attributes to obtain the book attribute features and performing quantitative analysis on the book information, the book characteristics can be analyzed from multiple dimensions, improving the accuracy and relevance of book recommendations.

[0051] Specifically, according to the user behavior data and book attribute features, the interest matching degree of the user for each book is calculated through a preset book recommendation model. The user behavior data is sorted out and analyzed. The book recommendation model includes, but is not limited to, content-based recommendation algorithms. Corresponding weights are set for the user behavior data and book attribute features according to the user's interest preferences. For example, the borrowing history weight is set to 0.4, the search keyword weight is set to 0.3, the book rating weight is set to 0.2, and the borrowing times weight is set to 0.1. The sorted user behavior data and book attribute features are input into the recommendation model. The model calculates the interest matching degree of the user for each book by analyzing the association between the user's interest and the book attributes. For example, for user A and the book *Ordinary World*, user A has borrowed the book *Ordinary World*. According to user A's borrowing history and search keywords, combined with the attribute features of the book *Ordinary World*, the similarity scores between other books and *Ordinary World* are calculated according to the set weights, so as to calculate the interest matching degree of the user for each book. By calculating the interest matching degree by combining the user behavior data and book attribute features, the user's interest preferences can be deeply analyzed, and the books that match the user's interests can be screened out. Compared with the single-dimensional recommendation method, this solution combines multiple dimensions, improving the accuracy of book recommendations and the user's acceptance of the recommended books, and helping to enhance the user's reading experience.

[0052] Preferably, according to the calculated interest matching degree, a ranking is performed from high to low. Among the sorted books, the top-ranked books are selected according to a preset first quantity to form a first regional recommended book set. For example, the system sets the preset first quantity to 5 books according to the user's borrowing situation. According to the ranking result, the top 5 books with the highest interest matching degree are selected as the first regional recommended book set. Screening and recommending books according to the interest matching degree can quickly focus on the books that match the user's interests, make the book recommendation process more personalized, improve the efficiency and practicality of the recommended books, and enhance the user's satisfaction with the recommendation service.

[0053] Furthermore, by analyzing the book relevance between the hot spot area and the non-hot spot area, recommended books are screened in the non-hot spot area to obtain a second regional recommended book set, including: S701. Obtain the book similarity by calculating the similarity between the book attribute features of the books in the non - hot - spot area and those of the books in the hot - spot area. S702. Calculate the correlation degree between the non - hot - spot area and the hot - spot area by combining the distance between the non - hot - spot area and the hot - spot area and the corresponding book similarity. S703. Screen out a preset second number of non - hot - spot areas in descending order of the correlation degree to obtain a candidate area set. S704. In each non - hot - spot area in the candidate area set, screen out a preset third number of recommended books in descending order of the user's interest matching degree for the books in the non - hot - spot area to obtain a second - area recommended book set.

[0054] In this embodiment, first, by analyzing the similarity between the book attribute features of the books in the hot - spot area and the non - hot - spot area, calculate the book similarity. Extract the book attribute features of the books from the hot - spot area and the non - hot - spot area respectively. By calculating the cosine similarity between the book attribute features, obtain the book similarity. For each book in the non - hot - spot area, calculate its similarity with all the books in the hot - spot area respectively and take the average value to obtain the book similarity between each book in the non - hot - spot area and the books in the hot - spot area. By calculating the book similarity, it is possible to analyze the connection between the books in the non - hot - spot area and those in the hot - spot area from the book attributes, avoiding the situation of only focusing on the hot - spot area and ignoring the valuable books in the non - hot - spot area, and broadening the scope of recommended books.

[0055] Preferably, analyze the correlation degree between the non - hot - spot area and the hot - spot area according to the calculated book similarity and the distance between the non - hot - spot area and the hot - spot area. Considering that in the spatial layout of the library, the distance between the non - hot - spot area and the hot - spot area affects the user's attention to the books in the non - hot - spot area to a certain extent. The closer the distance, the greater the probability that the user will come into contact with the books in this area. At the same time, the book similarity reflects the degree of connection of the book content. Take the average of the book similarities of each book in the non - hot - spot area, and after taking the inverse of the distance between the non - hot - spot area and the hot - spot area, perform weighted summation with the averaged book similarity to obtain the correlation degree. This can comprehensively consider the spatial factor and the content factor, more comprehensively analyze the correlation degree between the non - hot - spot area and the hot - spot area, quickly screen out the non - hot - spot areas that are closely related to the hot - spot area, and improve the pertinence and accuracy of book recommendation.

[0056] Specifically, according to the calculated correlation degree, the non-hot regions are sorted from high to low according to the correlation degree, and the top-ranked non-hot regions are selected according to a preset second quantity to obtain a candidate region set. By screening the candidate region set, the recommendation range of non-hot regions can be quickly narrowed, avoiding blindly screening books among a large number of non-hot regions, improving the recommendation efficiency. By screening out regions with high correlation degrees, the recommended books are more in line with the user's interests when recommending books, improving the accuracy and effectiveness of book recommendations. In the candidate region set, the interest matching degree of each book for the user is calculated through a book recommendation model and sorted from high to low, and a preset third quantity of books is screened out. For example, the system sets the preset third quantity to 2 books according to the user's book borrowing situation, and selects the top 2 books with the highest interest matching degrees in the candidate non-hot regions to form a second region recommended book set. By screening and recommending books within the candidate regions, the user's reading choices are improved.

[0057] Further, based on the borrowing situation of the user for the first recommended book set, the recommended books are optimized through a preset recommendation optimization model to obtain a second recommended book set for personalized recommendation of library books, including: S801. Analyze the user's interest change situation based on the borrowing situation of the books in the first recommended book set for the user to obtain an interest vector; S802. Combine the interest vector and optimize the recommended books through a preset recommendation optimization model to obtain a second recommended book set for personalized recommendation of library books.

[0058] In this embodiment, based on the borrowing situation of the books in the first recommended book set for the user, the user's interest change situation is analyzed to obtain an interest vector; corresponding quantization criteria are set by analyzing the borrowing situation of the books, corresponding values are set for each interest dimension. When the user borrows a certain type of book and the borrowing duration is long and the borrowing times are many, the interest value corresponding to this type of book is high; based on the user's book borrowing situation, the user's interest change situation is analyzed to obtain an interest vector.

[0059] Exemplarily, user B borrowed a literature book with a borrowing duration of 15 days and it was the second time to borrow this type of book. The value of the literature dimension can be set to 8 (with a full score of 10); when the user does not borrow a certain type of book, the value of this dimension is set to 0, and the interest vector of user B is obtained as [8, 0, 0, 0, 0], indicating that user B has a higher interest in literature books and a lower interest in other categories; by constructing an interest vector, the user's abstract interest changes are transformed into quantitative data, providing a data basis for optimizing book recommendations. According to the interest vector, the book recommendation strategy can be adjusted in a timely manner, improving the pertinence and timeliness of book recommendations and meeting the user's ever-changing reading needs.

[0060] Specifically, according to the interest vector, the books in the first recommended book set are adjusted and optimized through a preset recommendation optimization model in combination with the book resource information of the library. By comprehensively considering factors such as the matching degree between the book and the interest vector, the popularity of the book, and the inventory situation, the books that are more in line with the current interest changes of the user are selected, and the books that do not match the user's interests are removed to obtain the second recommended book set; the book recommendation results are dynamically adjusted according to the changes in the user's interests, improving the matching degree between the recommended books and the user's interests, providing more personalized recommendations that match the user's interests, and helping to improve the borrowing rate of books and the utilization efficiency of library resources.

[0061] Further, in combination with the interest vector, the recommended books are optimized through a preset recommendation optimization model to obtain the second recommended book set for personalized recommendation of library books, including: S901. According to the interest vector and combined with the user behavior data, the interest migration situation of the user is predicted through a preset recommendation optimization model to obtain an interest migration prediction vector; S902. Based on the interest migration prediction vector, the books that match the interest migration prediction vector are selected as the optimized book set; S903. The recommended books are optimized through the optimized book set to obtain the second recommended book set for personalized recommendation of library books.

[0062] In this embodiment, according to the interest vector and combined with the user behavior data, the interest migration situation of the user is predicted through a preset recommendation optimization model to obtain an interest migration prediction vector; considering that the reading interests of users are not fixed and will change with factors such as reading behavior and life experiences, the user's interest vector and user behavior data are integrated. The recommendation optimization model includes models such as time series analysis algorithms, association rule algorithms, or prediction algorithms based on deep learning. The recommendation optimization model in this embodiment is a recurrent neural network model. A large amount of historical data is used to train the recurrent neural network model to obtain a pre-trained recurrent neural network model. The integrated data is input into the pre-trained recurrent neural network model, and the model outputs the predicted interest migration prediction vector; by predicting the user's interest migration prediction vector, the development trend of the user's interests can be analyzed in advance, books that the user is interested in can be recommended in advance, the accuracy and forward-looking of the recommendation can be improved, and new reading interests of the user can also be stimulated.

[0063] Specifically, according to the calculated interest transfer prediction vector, by matching the attribute features of the books with the interest transfer prediction vector, the books that meet the interest transfer prediction vector are screened out. First, the matching rules between the book attribute features and the interest transfer prediction vector are determined, and the book categories and book keywords are matched separately. Books with the coincidence degree of the book keywords and the keywords related to the recently retrieved keywords of the user or the interest book categories in the interest vector reaching a preset ratio (such as 30%) are screened out as the optimized book set. Optimizing the book screening based on the interest transfer prediction vector can screen out books that meet the user's interest trend, improve the recommendation efficiency, and meet the user's reading interest needs.

[0064] Preferably, the books in the first recommended book set are adjusted and optimized using the screened optimized book set to obtain the second recommended book set; calculate the matching degree between each book in the first recommended book set and the interest transfer prediction vector, retain the books with a matching degree greater than the preset ratio in the first recommended book set, and remove the books with a matching degree less than or equal to the preset ratio. At the same time, the books in the optimized book set that are not in the first recommended book set are supplemented to the recommendation list in descending order of the matching degree to obtain the second recommended book set, and the books in the second recommended book set are sorted in descending order of the matching degree, so that the user can see the books that best meet their interests first; optimizing the recommended books through the optimized book set realizes the dynamic update and personalized adjustment of the recommended books. The optimized recommended books better meet the constantly changing interest needs of the user, improving the quality and effect of book recommendations.

[0065] Embodiment 2 In this embodiment, as Figure 4 , a personalized recommendation system for library books is provided to implement a personalized recommendation method for library books, including: A data acquisition module that acquires the book RFID tag transfer data and user behavior data during the user's book borrowing process; A book interest heat map construction module that combines the book RFID tag transfer data and user behavior data, divides the library into grids according to a preset grid size, analyzes the user's interest in the books in each grid area, and constructs a book interest heat map; A book recommendation module that, according to the book interest heat map, extracts the hot areas with an interest degree value greater than a preset interest degree threshold, and combines the hot areas to screen out recommended books through a preset book recommendation model to obtain the first recommended book set; A recommended book optimization module that optimizes the recommended books through a preset recommendation optimization model based on the user's borrowing situation of the first recommended book set to obtain the second recommended book set for personalized recommendation of library books.

[0066] In this embodiment, the data acquisition module collects the data generated during the user's book borrowing process from the library management system and related devices, including the book RFID tag transfer data and user behavior data. Among them, the book RFID tag transfer data records information such as the borrowing, returning, and movement trajectories of books, and the user behavior data includes behavior data such as the user's stay duration in the library, browsing habits, and retrieval operations; data collection provides data support for user interest analysis and book recommendation; the book interest heat map construction module divides the library space according to a preset grid size based on the data collected by the data acquisition module, divides the space into multiple grid areas, combines the book RFID tag transfer data and user behavior data, analyzes the user's interest level in the books in each grid area through a preset interest analysis model, calculates the corresponding interest level value, and spreads from the grid area where the user borrows the book as the center, combining the interest level value to the surrounding areas to construct a book interest heat map, which can intuitively present the user interest distribution trend.

[0067] Specifically, based on the book interest heat map, the book recommendation module filters out the hot areas where the user's interests are concentrated through a set interest level threshold. In the hot areas, it analyzes the user's interest preferences for different books through a preset book recommendation model, filters out the books that meet the user's interests, forms the first recommended book set, and analyzes the correlation between the books in the hot areas and the non-hot areas, filters out potential recommended books in the non-hot areas, and supplements them to the recommended book set to provide a more abundant, comprehensive, and interest-fitting book recommendation list for the user; the recommended book optimization module analyzes the dynamic changes in the user's interests based on the borrowing feedback of the user on the first recommended book set, obtains the user's interest vector, combines the user behavior data, predicts the user's interest migration trend through a preset recommendation optimization model to obtain the interest migration prediction vector, and based on the interest migration prediction vector, filters out the matching books to form the optimized book set, optimizes the first recommended book set, removes the unmatched books, and supplements the books that match the user's interest change trend to obtain the second recommended book set, realizing personalized library book recommendation and improving the user's borrowing experience and the utilization efficiency of library resources.

[0068] 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. What is described in the above embodiments and the specification only illustrates the principles of 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. A personalized recommendation method for library books, characterized in that, Including: Obtaining the book RFID tag transfer data and user behavior data during the user's book borrowing process; Combining the book RFID tag transfer data and user behavior data, dividing the library into grids according to a preset grid size, analyzing the user's interest level in the books in each grid area, and constructing a book interest heat map; According to the book interest heat map, extracting the hot areas where the interest level value is greater than a preset interest level threshold, and screening out recommended books through a preset book recommendation model in combination with the hot areas to obtain a first set of recommended books; Based on the borrowing situation of the user for the first set of recommended books, optimizing the recommended books through a preset recommendation optimization model to obtain a second set of recommended books for personalized recommendation of the library books.

2. The personalized recommendation method for library books according to claim 1, characterized in that, Combining the book RFID tag transfer data and user behavior data, dividing the library into grids according to a preset grid size, analyzing the user's interest level in the books in each grid area, and constructing a book interest heat map, including: Dividing the library into grids according to a preset grid size to obtain multiple grid areas; Combining the book RFID tag transfer data and user behavior data, analyzing the user's interest level in the books in each grid area through a preset interest analysis model to obtain the interest level value of each grid area; Taking the grid area where the user-borrowed book is located as the center, and spreading the area to the surrounding grid areas in combination with the interest level value to construct a book interest heat map.

3. The personalized recommendation method for library books according to claim 2, characterized in that, The combining the book RFID tag transfer data and user behavior data, analyzing the user's interest level in the books in each grid area through a preset interest analysis model to obtain the interest level value of each grid area includes: According to the book RFID tag transfer data, extracting the first grid area where the user-borrowed book is located, analyzing the time difference between the borrowing and returning of the book, and obtaining the book borrowing duration; According to the user behavior data, analyzing the user's book viewing situation through a preset interest analysis model to obtain the user behavior analysis result; Combining the book borrowing duration and the user behavior analysis result to determine the interest level value of the first grid area; Calculating the interest level value of each grid area through the content similarity between the books in the grid area in combination with the interest level value of the first grid area.

4. The personalized recommendation method for library books according to claim 2, wherein, The taking the grid area where the user-borrowed book is located as the center, and spreading the area to the surrounding grid areas in combination with the interest level value to construct a book interest heat map includes: Taking the first grid area where the user-borrowed book is located as the central grid area; Calculating the heat value of each grid area according to the interest level value of each grid area and the distance from the central grid area, where when there are multiple central grid areas, the heat values calculated according to each central grid area are superimposed; Taking the central grid area as the center, spreading the area to the surrounding grid areas according to the heat value to construct a book interest heat map.

5. The personalized recommendation method for library books according to claim 1, characterized in that, According to the book interest heat map, extract the hot regions where the interest degree value is greater than the preset interest degree threshold, and combine the hot regions to screen out recommended books through a preset book recommendation model to obtain a first recommended book set, including: According to the book interest heat map, extract the second grid regions where the interest degree value is greater than the preset interest degree threshold, and merge adjacent second grid regions to obtain multiple hot regions; Within each hot region, analyze the user's interest in different books through a preset book recommendation model to screen out recommended books, and obtain a first regional recommended book set; By analyzing the book relevance between the hot regions and non-hot regions, screen out recommended books within the non-hot regions to obtain a second regional recommended book set; Combine the first regional recommended book set and the second regional recommended book set to obtain a first recommended book set.

6. The personalized recommendation method for library books according to claim 5, characterized in that, The step of, within each hot region, analyzing the user's interest in different books through a preset book recommendation model to screen out recommended books to obtain a first regional recommended book set includes: Within each hot region, analyze the attributes of each book to obtain book attribute characteristics; According to the user behavior data, combine the book attribute characteristics, and calculate the interest matching degree of the user for each book through a preset book recommendation model; Screen out the preset first number of recommended books according to the interest matching degree from high to low to obtain a first regional recommended book set.

7. The personalized recommendation method for library books according to claim 5, characterized in that, The step of, by analyzing the book relevance between the hot regions and non-hot regions, screening out recommended books within the non-hot regions to obtain a second regional recommended book set includes: Calculate the book similarity by calculating the similarity between the book attribute characteristics of the books within the non-hot regions and the book attribute characteristics of the books within the hot regions; Combine the distance between the non-hot regions and the hot regions and the corresponding book similarity to calculate the relevance between the non-hot regions and the hot regions; Screen out the preset second number of non-hot regions according to the relevance from high to low to obtain a candidate region set; Within each non-hot region in the candidate region set, screen out the preset third number of recommended books according to the interest matching degree of the user for the books within the non-hot regions from high to low to obtain a second regional recommended book set.

8. The personalized recommendation method for library books according to claim 1, characterized in that, Based on the borrowing situation of the user for the first recommended book set, optimize the recommended books through a preset recommendation optimization model to obtain a second recommended book set for personalized recommendation of library books, including: According to the borrowing situation of the user for the books in the first recommended book set, analyze the interest change situation of the user to obtain an interest vector; Combine the interest vector and optimize the recommended books through a preset recommendation optimization model to obtain a second recommended book set for personalized recommendation of library books.

9. The personalized recommendation method for library books according to claim 8, wherein Combine the interest vector and optimize the recommended books through a preset recommendation optimization model to obtain a second recommended book set for personalized recommendation of library books, including: According to the interest vector combined with the user behavior data, predict the interest migration situation of the user through a preset recommendation optimization model to obtain an interest migration prediction vector; Based on the interest transfer prediction vector, books that match the interest transfer prediction vector are selected as the optimized book set; The recommended books are optimized by the optimized book set to obtain a second recommended book set for personalized recommendation of library books.

10. A personalized recommendation system for library books, characterized in that, A device for implementing the personalized recommendation method of library books according to any one of claims 1 to 9, comprising: a data acquisition module that acquires book RFID tag transfer data and user behavior data during the user's book borrowing process; a book interest heat map construction module that combines the book RFID tag transfer data and user behavior data, divides the library into grids according to a preset grid size, analyzes the user's interest in books in each grid area, and constructs a book interest heat map; a book recommendation module that extracts hot regions with an interest degree value greater than a preset interest degree threshold according to the book interest heat map, and combines the hot regions to screen out recommended books through a preset book recommendation model to obtain a first recommended book set; a recommended book optimization module that optimizes the recommended books through a preset recommendation optimization model based on the user's borrowing situation of the first recommended book set to obtain a second recommended book set for personalized recommendation of library books.

Images