Book recommendation method and device, electronic equipment and computer storage medium

Abstract

The application provides a book recommendation method and device, electronic equipment and computer storage medium, and belongs to the technical field of big data. In the method, the borrowing probability of each book in each region of a target reader is determined according to the historical borrowing record of the target reader, then the borrowing probability is corrected according to the heat of each book in each region and the popularity of each book in each region, to obtain a book recommendation vector of the target reader in each region. In addition, the book recommendation vector of the target reader in each region is further corrected by the recommendation degree of each regional book management subsystem, to obtain a final target book recommendation vector of the target reader. The obtained target book recommendation vector of the target reader is more accurate, can accurately recommend books for the target reader, and achieves a good book recommendation effect.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of big data, and in particular to a book recommendation method and device, electronic equipment and computer storage medium. BACKGROUND

[0002] At present, when borrowing books, book recommendations often originate from the user's historical borrowing records, and books similar to the books borrowed by the user are recommended to the user.

[0003] The books recommended by the above method have poor accuracy and are difficult to achieve good recommendation results. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a book recommendation method, device, electronic equipment and computer storage medium to alleviate the technical problem of poor accuracy of books recommended by existing book recommendation methods and the difficulty of achieving good recommendation results.

[0005] In a first aspect, an embodiment of the present application provides a book recommendation method applied to a book management system connected with a plurality of regional book management subsystems, the method comprising:

[0006] obtaining book information uploaded by each regional book management subsystem and a recommendation degree of each regional book management subsystem, wherein the book information comprises: a browsing frequency of each book in each region, a discussion frequency of each book in each region, a borrowing frequency of each book in each region, and feature information of each book in each region;

[0007] determining the heat of each book in each region based on the browsing frequency of each book in each region and the discussion frequency of each book in each region;

[0008] determining the popularity of each book in each region based on the borrowing frequency of each book in each region and the browsing frequency of each book in each region;

[0009] obtaining a recommendation request for book recommendation to target readers in a target region initiated by a target regional book management subsystem, wherein the recommendation request carries reader information of the target readers, and the reader information comprises: personal information of the target readers and historical borrowing records of the target readers;

[0010] using a book recommendation model to recommend books based on the reader information of the target readers and the feature information of each book in each region, to obtain a borrowing probability matrix of each book in each region for the target readers;

[0011] According to the borrowing probability matrix of each book in each area of the target reader, the heat of each book in each area, and the popularity of each book in each area, a book recommendation vector of the target reader in each area is determined;

[0012] Based on the book recommendation vector of the target reader in each area and the recommendation degree of the book management subsystem in each area, a target book recommendation vector of the target reader is determined, and then books are recommended to the target reader according to the target book recommendation vector.

[0013] Further, the heat of each book in each area is determined based on the number of views of each book in each area and the number of discussions of each book in each area, including:

[0014] According to the heat calculation formula The heat of each book in each area is calculated, wherein hot ij represents the heat of book i in area j, scan_num ij represents the number of views of book i in area j, dis_num ij represents the number of discussions of book i in area j, danand_rate ij represents the demand degree of book i in area j.

[0015] Further, the popularity of each book in each area is determined based on the number of borrowings of each book in each area and the number of views of each book in each area, including:

[0016] According to the borrowing-viewing ratio calculation formula The borrowing rate of each book in each area is calculated, wherein borrow_scan_rate ij represents the borrowing rate of book i in area j, borrow_num ij represents the number of borrowings of book i in area j, scan_num ij represents the number of views of book i in area j;

[0017] According to the popularity calculation formula The popularity of each book in each area is calculated, wherein welcome_rate ij represents the popularity of book i in area j, borrow_scan_rate ij represents the borrowing rate of book i in area j, damand_rate ij represents the demand degree of book i in area j.

[0018] Further, the book recommendation vector of the target reader in each area is a score vector of each book in each area of the target reader, and the book recommendation vector of the target reader in each area is determined according to a borrowing probability matrix of each book in each area of the target reader, a heat of each book in each area, and a popularity of each book in each area, and the method comprises the following steps of:

[0019] According to a score calculation formula score ij = w ij *hot ij *welcome_rate ij The score of each book in each area of the target reader is calculated, and then a score vector of each book in each area of the target reader is obtained, wherein wcore ij represents the score of book i in area j of the target reader, w ij represents the borrowing probability of book i in area j of the target reader, hot ij represents the heat of book i in area j, and welcome_rate ij represents the popularity of book i in area j.

[0020] Further, the target book recommendation vector of the target reader is determined based on the book recommendation vector of the target reader in each area and a recommendation degree of the book management subsystem in each area, and the method comprises the following steps of:

[0021] According to a target book recommendation vector calculation formula The target book recommendation vector of the target reader is calculated, wherein U represents the target book recommendation vector of the target reader, z represents the number of the area book management subsystems, w k represents the recommendation degree of area book management subsystem k, and S k represents the book recommendation vector of the target reader in the area corresponding to area book management subsystem k.

[0022] Further, after recommending books to the target reader according to the target book recommendation vector, the method further comprises the following steps of:

[0023] Obtaining an actual borrowing message of the target reader;

[0024] If the target reader does not borrow, the recommendation degrees of the area book management subsystems remain unchanged;

[0025] If the target reader borrows books, the updated recommendation degrees of the area book management subsystems are calculated according to a recommendation degree update formula , wherein w' k represents the updated recommendation degree of area book management subsystem k, and w krepresents the recommendation degree of the target regional library management subsystem, represents the recommendation degree of the target regional library management subsystem, and θ is a positive number less than 1, β represents whether the target reader borrows the book recommended by the target regional library management subsystem, β = 0 represents that the target reader borrows the book recommended by the target regional library management subsystem, β = 1 represents that the target reader does not borrow the book recommended by the target regional library management subsystem, U represents a target book recommendation vector of the target reader, and a represents a book vector actually borrowed by the target reader.

[0026] Further, the book information is obtained based on user scanning of a book code, and the book code comprises a publishing agency identification code, a book category code, a publishing batch code and a serial number, and the book code has uniqueness.

[0027] In a second aspect, the embodiment of the present application further provides a book recommendation device applied to a library management system, wherein the library management system is connected with a plurality of regional library management subsystems, and the device comprises:

[0028] A first acquisition unit is configured to acquire book information uploaded by each regional library management subsystem and recommendation degrees of the regional library management subsystems, wherein the book information comprises a browsing frequency of each book in each region, a discussion frequency of each book in each region, a borrowing frequency of each book in each region, and characteristic information of each book in each region.

[0029] A first determination unit is configured to determine a heat of each book in each region based on the browsing frequency of each book in each region and the discussion frequency of each book in each region.

[0030] A second determination unit is configured to determine a popularity of each book in each region based on the borrowing frequency of each book in each region and the browsing frequency of each book in each region.

[0031] A second acquisition unit is configured to acquire a recommendation request for book recommendation to a target reader in a target region initiated by a target regional library management subsystem, wherein the recommendation request carries reader information of the target reader, and the reader information comprises personal information of the target reader and historical borrowing records of the target reader.

[0032] A book recommendation unit is configured to perform book recommendation on the reader information of the target reader and the characteristic information of each book in each region by using a book recommendation model to obtain a borrowing probability matrix of each book in each region for the target reader.

[0033] The third determining unit is configured to determine a book recommendation vector of the target reader in each region according to the borrowing probability matrix of each book in each region, the heat of each book in each region and the popularity of each book in each region.

[0034] The fourth determining unit is configured to determine a target book recommendation vector of the target reader based on the book recommendation vector of the target reader in each region and the recommendation degree of the book management sub-system in each region, and further recommend books for the target reader according to the target book recommendation vector.

[0035] In a third aspect, an electronic device is provided, which includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and the processor executes the computer program to implement the steps of the method in any of the above first aspect.

[0036] In a fourth aspect, a computer storage medium is provided, which stores a computer program, and the computer runs the computer program to implement the steps of the method in any of the above first aspect.

[0037] In the embodiment of the present application, a book recommendation method is provided, which is applied to a book management system connected with a plurality of regional book management subsystems. The method comprises: obtaining book information uploaded by each regional book management subsystem and recommendation degrees of each regional book management subsystem, wherein the book information comprises: browsing times of each book in each region, discussion times of each book in each region, borrowing times of each book in each region, and characteristic information of each book in each region; determining a heat degree of each book in each region based on the browsing times of each book in each region and the discussion times of each book in each region; determining a popularity degree of each book in each region based on the borrowing times of each book in each region and the browsing times of each book in each region; obtaining a recommendation request for book recommendation to a target reader in a target region initiated by a target regional book management subsystem, wherein the recommendation request carries reader information of the target reader, and the reader information comprises: personal information of the target reader and historical borrowing records of the target reader; performing book recommendation on the reader information of the target reader and the characteristic information of each book in each region by using a book recommendation model to obtain a borrowing probability matrix of each book in each region for the target reader; determining a book recommendation vector of each region for the target reader according to the borrowing probability matrix of each book in each region for the target reader, the heat degree of each book in each region, and the popularity degree of each book in each region; determining a target book recommendation vector of the target reader based on the book recommendation vector of each region for the target reader and the recommendation degrees of the regional book management subsystems, and then recommending books for the target reader according to the target book recommendation vector. As can be seen from the above description, in the book recommendation method of the present application, the book recommendation model is used to perform book recommendation on the reader information of the target reader and the characteristic information of each book in each region to obtain the borrowing probability matrix of each book in each region for the target reader. In this process, the historical borrowing records of the target reader are considered, that is, the borrowing probability of each book in each region for the target reader is first determined according to the historical borrowing records of the target reader, and then the borrowing probability is corrected according to the heat degree of each book in each region and the popularity degree of each book in each region to obtain the book recommendation vector of each region for the target reader. In addition, the book recommendation vector of each region for the target reader is further corrected by the recommendation degrees of the regional book management subsystems to obtain the final target book recommendation vector of the target reader. Finally, books are recommended for the target reader according to the target book recommendation vector. The above process takes the heat degree of each book in each region and the popularity degree of each book in each region as consideration indexes for book recommendation, which can improve the accuracy of book recommendation. In addition, the target book recommendation vector of the target reader determined by the joint recommendation of the plurality of regional book management subsystems will be more accurate, which can accurately recommend books for the target reader and achieve good book recommendation effect, thereby solving the technical problem that the existing book recommendation method has poor accuracy in recommending books and is difficult to achieve good recommendation effect. BRIEF DESCRIPTION OF DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the specific embodiments or the prior art of the present application, the following will briefly introduce the drawings needed to be used in the specific embodiments or the prior art description. Obviously, the drawings described below are some embodiments of the present application, and the ordinary skilled in the art can obtain other drawings according to these drawings without any creative effort.

[0039] Figure 1 A flowchart of a book recommendation method provided by an embodiment of the present application;

[0040] Figure 2 A schematic diagram of a book recommendation device provided by an embodiment of the present application;

[0041] Figure 3 A schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0042] The technical solutions of the present application will be described clearly and completely below in conjunction with the embodiments. Obviously, the described embodiments are some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by the ordinary skilled in the art without any creative effort are within the scope of protection of the present application.

[0043] The traditional book recommendation method has poor accuracy of recommended books, and it is difficult to achieve good recommendation effect.

[0044] Therefore, in the book recommendation method of the present application, a book recommendation model is used to recommend books for the reader information of the target reader and the feature information of each book in each area, to obtain a borrowing probability matrix of each book in each area for the target reader. In this process, the historical borrowing records of the target reader are considered, that is, the borrowing probability of each book in each area for the target reader is determined according to the historical borrowing records of the target reader, and then the above borrowing probability is corrected according to the heat of each book in each area and the popularity of each book in each area, to obtain a book recommendation vector of the target reader in each area. In addition, the above book recommendation vector of the target reader in each area is further corrected by the recommendation degree of the area book management subsystem, to obtain a final target book recommendation vector of the target reader. Finally, books are recommended for the target reader according to the target book recommendation vector. The above process takes the heat of each book in each area and the popularity of each book in each area as the consideration index of book recommendation, which can improve the accuracy of book recommendation. In addition, the target book recommendation vector of the target reader determined by the joint recommendation of the multiple area book management subsystems will be more accurate, which can accurately recommend books for the target reader and achieve good book recommendation effect.

[0045] For the convenience of understanding the present embodiment, first, a book recommendation method disclosed by the embodiment of the present application is introduced in detail.

[0046] Embodiment one:

[0047] For the convenience of understanding the present embodiment, first, a book recommendation method disclosed by the embodiment of the present application is introduced in detail, referring to a flowchart of a book recommendation method shown in FIG. 1, mainly including the following steps: Figure 1

[0048] In step S102, book information uploaded by each regional book management subsystem and the recommendation degree of each regional book management subsystem are acquired, wherein the book information includes the browsing times of each book in each region, the discussion times of each book in each region, the borrowing times of each book in each region, and the characteristic information of each book in each region.

[0049] In the embodiment of the present application, the execution subject of the method can be a book management system, and the book management system is connected with multiple regional book management subsystems. The above-mentioned regional book management subsystems will upload the latest book information at irregular times. For example, when the regional book management subsystems obtain that the book information has changed, such as the browsing times of each book in the current region has changed, at this time, the regional book management subsystem of the current region can upload the latest book information. For another example, the discussion times of each book in the current region has changed, at this time, the regional book management subsystem of the current region can upload the latest book information. For another example, the borrowing times of each book in the current region has changed, at this time, the regional book management subsystem of the current region can upload the latest book information. Of course, the latest book information can also be uploaded at intervals of a preset time length, and the present embodiment does not specifically limit the above-mentioned uploading time.

[0050] ​The book information includes: the browsing times of each book in each area, the discussion times of each book in each area, the borrowing times of each book in each area, the characteristic information of each book in each area, the browsing times of each book in each area are obtained by the regional book management subsystem in the corresponding area counting the browsing times of each book in the corresponding area, when a reader borrows a book, the reader needs to select a book first, when selecting a book, the reader needs to browse each book simply, when browsing, the reader needs to scan the book code on the book first, so that the regional book management subsystem can count the browsing times of each book, the discussion times of each book are also obtained by the reader scanning the book code on the book on the corresponding book scanning software of the reader, and then operating the reader interaction platform / community module, entering the reader interaction page, so that the regional book management subsystem can count the discussion information of the book uploaded by the reader to obtain the discussion times of the book, the borrowing times of each book are obtained by counting the borrowing records of each book, and the characteristic information of each book includes: the category of the book, the author of the book, the abstract of the book, the book name and the like.

[0051] The initial value of the recommendation degree of each regional book management subsystem is 1, and the recommendation degree of each regional book management subsystem is updated according to the actual borrowing information of the target reader.

[0052] In addition, it should be noted that the browsing times of each book, the discussion times of each book and the borrowing times of each book are related to time, and can be flexibly set according to specific needs, for example, the browsing times of each book, the discussion times of each book and the borrowing times of each book in the past week before the current time can be set, and the browsing times of each book, the discussion times of each book and the borrowing times of each book in the past 10 days before the current time can also be set, when the time is relatively long, the basis data can be weighted and summed according to needs, for example, the browsing times of each book, the discussion times of each book and the borrowing times of each book in the past 10 days before the current time are obtained, the weight of the browsing times of each book, the discussion times of each book and the borrowing times of each book in the past 1 day before the current time is a1, the weight of the browsing times of each book, the discussion times of each book and the borrowing times of each book in the past 2 days before the current time is a2, the weight of the browsing times of each book, the discussion times of each book and the borrowing times of each book in the past 3 days before the current time is a3, and so on, the sum of all weights is 1, then the browsing times of each book are weighted and summed according to the weight, the discussion times of each book are weighted and summed according to the weight, and so on, to obtain the final browsing times of each book in each region, the discussion times of each book in each region, and the borrowing times of each book in each region, in this way, the recent relevant data can be obtained for subsequent calculation of the heat and welcome degree, which is more accurate and scientific.

[0053] In step S104, the heat of each book in each region is determined based on the browsing times of each book in each region and the discussion times of each book in each region.

[0054] Specifically, the inventors consider that the higher the heat of a book is, the greater the probability of borrowing by a reader is, therefore, in order to accurately recommend books to target readers, the heat of each book in each region is also used as an index for book recommendation, which is used to assist subsequent book recommendation, so as to improve the accuracy of book recommendation.

[0055] The process of determining the heat of each book in each region will be described in detail below, and will not be described here.

[0056] In step S106, the welcome degree of each book in each region is determined based on the borrowing times of each book in each region and the browsing times of each book in each region.

[0057] Specifically, the inventors consider that the higher the welcome degree of a book is, the greater the probability of borrowing by a reader is, therefore, in order to accurately recommend books to target readers, the welcome degree of each book in each region is also used as an index for book recommendation, which is used to assist subsequent book recommendation, so as to improve the accuracy of book recommendation.

[0058] The process of determining the popularity of each book in each region will be described in detail below and will not be repeated here.

[0059] In step S108, a recommendation request for recommending books to the target reader in the target region initiated by the target regional book management subsystem is obtained, wherein the recommendation request carries reader information of the target reader, and the reader information includes: personal information of the target reader, historical borrowing records of the target reader;

[0060] Specifically, the recommendation request can be initiated by the target reader through the book recommendation module of the reading code scanning software, and the reading code scanning software is an application program in the target regional book management subsystem. The personal information of the target reader specifically includes: age, gender, education, book preferences, etc. of the target reader, and the historical borrowing records of the target reader specifically include: books borrowed by the target reader, borrowing time, etc.

[0061] In step S110, a book recommendation model is used to recommend books to the target reader based on the reader information of the target reader and the feature information of each book in each region, and a borrowing probability matrix of the target reader for each book in each region is obtained.

[0062] Specifically, the book recommendation model is a model pre-trained for predicting the borrowing probability of books.

[0063] The training process can be: obtaining the borrowing records of the reader borrowing books, the reader information of the reader, and the feature information of the books, inputting the reader information of the reader and the feature information of the books into the original book recommendation model, and outputting the borrowing probability of the reader borrowing each book. The borrowing records of the reader borrowing books are used as the true value, and the loss is calculated with the predicted value of the model. The parameters of the model are adjusted and trained through the loss.

[0064] In step S112, a book recommendation vector of the target reader in each region is determined based on the borrowing probability matrix of the target reader for each book in each region, the heat of each book in each region, and the popularity of each book in each region.

[0065] Specifically, the book recommendation vector of the target reader in each region is the recommendation value (i.e. score value) of all books contained in each region. The above process specifically corrects the borrowing probability matrix of the target reader for each book in each region based on the heat of each book in each region and the popularity of each book in each region, and obtains the book recommendation vector of the target reader in each region.

[0066] In step S114, a target book recommendation vector of the target reader is determined based on the book recommendation vector of the target reader in each region and the recommendation degree of the regional book management subsystem, and then books are recommended to the target reader according to the target book recommendation vector.

[0067] Specifically, the target book recommendation vector contains recommendation values of all books (each book corresponds to a recommendation value), and books are recommended to the target reader according to the size of the recommendation values (the larger the recommendation value, the higher the priority of the corresponding book).

[0068] In the embodiment of the present application, a book recommendation method is provided, which is applied to a book management system connected with a plurality of regional book management subsystems. The method comprises: obtaining book information uploaded by each regional book management subsystem and recommendation degrees of each regional book management subsystem, wherein the book information comprises: browsing times of each book in each region, discussion times of each book in each region, borrowing times of each book in each region, and characteristic information of each book in each region; determining a heat degree of each book in each region based on the browsing times of each book in each region and the discussion times of each book in each region; determining a popularity degree of each book in each region based on the borrowing times of each book in each region and the browsing times of each book in each region; obtaining a recommendation request for book recommendation to a target reader in a target region initiated by a target regional book management subsystem, wherein the recommendation request carries reader information of the target reader, and the reader information comprises: personal information of the target reader and historical borrowing records of the target reader; performing book recommendation on the reader information of the target reader and the characteristic information of each book in each region by using a book recommendation model to obtain a borrowing probability matrix of each book in each region for the target reader; determining a book recommendation vector of each region for the target reader according to the borrowing probability matrix of each book in each region for the target reader, the heat degree of each book in each region, and the popularity degree of each book in each region; determining a target book recommendation vector of the target reader based on the book recommendation vector of each region for the target reader and the recommendation degrees of the regional book management subsystems, and then recommending books for the target reader according to the target book recommendation vector. As can be seen from the above description, in the book recommendation method of the present application, the book recommendation model is used to perform book recommendation on the reader information of the target reader and the characteristic information of each book in each region to obtain the borrowing probability matrix of each book in each region for the target reader. In this process, the historical borrowing records of the target reader are considered, that is, the borrowing probability of each book in each region for the target reader is first determined according to the historical borrowing records of the target reader, and then the borrowing probability is corrected according to the heat degree of each book in each region and the popularity degree of each book in each region to obtain the book recommendation vector of each region for the target reader. In addition, the book recommendation vector of each region for the target reader is further corrected by the recommendation degrees of the regional book management subsystems to obtain the final target book recommendation vector of the target reader. Finally, books are recommended for the target reader according to the target book recommendation vector. The above process takes the heat degree of each book in each region and the popularity degree of each book in each region as consideration indexes for book recommendation, which can improve the accuracy of book recommendation. In addition, the target book recommendation vector of the target reader determined by the joint recommendation of the plurality of regional book management subsystems will be more accurate, which can accurately recommend books for the target reader and achieve good book recommendation effect, thereby solving the technical problem that the existing book recommendation method has poor accuracy in recommending books and is difficult to achieve good recommendation effect.

[0069] The above briefly introduces the book recommendation method of the present application, and the following will describe the specific contents involved in detail.

[0070] In an optional embodiment of the present application, the popularity of each book in each area is determined based on the browsing times of each book in each area and the discussion times of each book in each area, and specifically includes the following steps:

[0071] According to the hotness calculation formula The hotness of each book in each area is calculated, wherein hot ij represents the hotness of book i in area j, scan_num ij represents the browsing times of book i in area j, dis_num ij represents the discussion times of book i in area j, and demand_rate ij represents the demand degree of book i in area j.

[0072] Specifically, the demand degree of a book is pre-set, and its value range is 0 to 1, and the demand degree of a book is used to represent the characteristics of the book. The more the browsing times of a book in a certain area, the greater the hotness of the book in the area. However, when calculating the hotness, the noise of the demand degree of the book needs to be excluded, so it needs to be divided by the demand degree of the book.

[0073] In an optional embodiment of the present application, the popularity of each book in each area is determined based on the borrowing times of each book in each area and the browsing times of each book in each area, and specifically includes the following steps:

[0074] (1) According to the borrowing rate calculation formula The borrowing rate of each book in each area is calculated, wherein borrow_scan_rate ij represents the borrowing rate of book i in area j, borrow_num ij represents the borrowing times of book i in area j, scan_num ij represents the browsing times of book i in area j.

[0075] (2) According to the popularity calculation formula The popularity of each book in each area is calculated, wherein welcome_rate ij represents the popularity of book i in area j, borrow_scan_rate ij represents the borrowing rate of book i in area j, and demand_rate ij represents the demand degree of book i in area j.

[0076] Specifically, the high borrowing rate of a certain book does not necessarily mean that the book is popular, because some books are essential books, and their purchase rate should be high, not because readers like them. Therefore, when calculating the popularity of books in each region, the borrowing rate of books in each region should be divided by the essentiality of books in the region.

[0077] In an optional embodiment of the present application, the book recommendation vector of the target reader in each region is a score vector of the target reader for each book in each region, and the book recommendation vector of the target reader in each region is determined according to the borrowing probability matrix of the target reader for each book in each region, the heat of each book in each region, and the popularity of each book in each region, and specifically includes the following steps:

[0078] According to the score calculation formula score ij = w ij *hot ij *welcome_rate ij The score of the target reader for each book in each region is calculated, and then the score vector of the target reader for each book in each region is obtained, wherein score ij represents the score of the target reader for book i in region j, w ij represents the borrowing probability of the target reader for book i in region j, hot ij represents the heat of book i in region j, and welcome_rate ij represents the popularity of book i in region j.

[0079] In an optional embodiment of the present application, the target book recommendation vector of the target reader is determined based on the book recommendation vector of the target reader in each region and the recommendation degree of each regional book management subsystem, and specifically includes the following steps:

[0080] According to the target book recommendation vector calculation formula The target book recommendation vector of the target reader is calculated, wherein U represents the target book recommendation vector of the target reader, z represents the number of regional book management subsystems, w k represents the recommendation degree of the regional book management subsystem k, and S k represents the book recommendation vector of the target reader in the region corresponding to the regional book management subsystem k.

[0081] In an optional embodiment of the present application, after recommending books to the target reader according to the target book recommendation vector, the method further includes the following steps:

[0082] (1) Obtain the actual borrowing message of the target reader;

[0083] (2) If the target reader does not borrow, the recommendation degree of each regional book management subsystem remains unchanged;

[0084] (3) If the target reader borrows the book, the recommended degree is updated according to the updating formula The recommended degree of each regional book management subsystem after updating is calculated, wherein w k represents the recommended degree of the regional book management subsystem k after updating, w k represents the recommended degree of the regional book management subsystem k, represents the recommended degree of the target regional book management subsystem, and θ is a positive number less than 1, β represents whether the target reader borrows the book recommended by the target regional book management subsystem, β = 0 represents that the target reader borrows the book recommended by the target regional book management subsystem, β = 1 represents that the target reader does not borrow the book recommended by the target regional book management subsystem, U represents the target book recommendation vector of the target reader, and a represents the book vector actually borrowed by the target reader.

[0085] In an optional embodiment of the present application, the book information is obtained based on user scanning of a book code, and the book code comprises a publishing organization identification code, a book category code, a publishing batch code, and a serial number, and the book code has uniqueness.

[0086] Specifically, the user can be a reader.

[0087] The book management system of the present application can also realize various functions based on the strategy of one book one code, and the other functions realized thereby will be introduced as follows:

[0088] The book code of the present application is generated by using a globally unique identifier (GUID) or an algorithm with sufficient randomness and uniqueness. The book code should comprise information segments such as a publishing organization identification code, a book category code, a publishing batch code, and a serial number, so as to ensure that the book code of each book is unique in the global scope and has identifiable and traceable properties.

[0089] For example, a typical code structure can be: [publishing organization ID (4 bits)]-[book category ID (3 bits)]-[publishing year (4 bits)]-[batch number (2 bits)]-[serial number (6 bits)]. Such a book code can distinguish different publishing organizations and book categories, and can also reflect important information such as publishing time and batch sequence.

[0090] In practical applications, a special coding database is constructed to store all generated book codes and the corresponding book details such as book title, author, ISBN, edition, price, publication date, and print quantity. The database should have high-efficiency data storage, query, and update capabilities, support multi-thread concurrent access, and meet the needs of large-scale book coding management. The above coding database can be a distributed database, specifically constructed by a ClickHouse cluster. When a user initiates an operation request for related data, the operation request is parsed by the database model to obtain a SQL statement for operating the ClickHouse cluster. The SQL statement is sent to the ClickHouse cluster to execute the corresponding database operation, and the user (specifically the system) can obtain the database operation results returned by the ClickHouse cluster.

[0091] At the same time, the database should use secure and reliable encryption technology to encrypt the storage of book codes and book information, prevent data leakage and illegal tampering, and ensure the integrity and confidentiality of the data.

[0092] Develop coding generation software or modules and integrate them with the book management system of publishing enterprises. Before the book is typeset and ready for printing, a batch of book codes is automatically generated according to the predetermined coding rules, and these book codes and the corresponding book information are recorded in the coding database.

[0093] Then, in the printing process, through the interface with the printing equipment, the book codes are printed in the form of QR codes, barcodes, or invisible codes in specific locations of the book, such as the back cover, copyright page, or specific corners of the inner pages. At the same time, ensure the clarity, accuracy, and readability of the book code printing, and avoid problems such as blurring, errors, or repeated printing.

[0094] Develop a reading code scanning software (APP) for mobile devices such as smartphones and tablets, supporting mainstream operating systems such as iOS and Android. The APP should have powerful code scanning and recognition capabilities, allowing quick and accurate identification of QR codes or barcodes on books and parsing of the code information (i.e., book codes).

[0095] The scanning application interface design should be simple and friendly, making it easy for users to operate. After successful scanning, the basic information of the book, such as the title, author, publisher, and publication date, can be displayed immediately, and relevant operation buttons can be provided, such as viewing detailed information, participating in interactive activities, purchasing related books or peripheral products, etc.

[0096] Data Collection and Transmission: When a reader scans the book code using the book scanning software, in addition to parsing the book code, some other related data should also be collected, such as the scanning time, scanning location (obtained through the GPS positioning function of the mobile device), user device information (device model, operating system version, etc.). These data will be packaged together with the book code and transmitted to the code database for storage through a secure network communication protocol, and then analyzed through the system.

[0097] To ensure the stability and reliability of data transmission, network conditions can be detected. In this case, the current network parameters are obtained, the network condition prediction model is used to predict the network condition of the current network parameters, the network condition encountered in the future preset time is obtained, and the target communication mode corresponding to the network condition is obtained. The target communication mode includes any one of the following: MQTT protocol communication mode, short message platform communication mode; when reaching the future preset time, communication is carried out according to the target communication mode.

[0098] Data buffering and breakpoint resuming technologies can also be used to prevent data loss due to network fluctuations or interruptions. At the same time, the transmitted data is encrypted to prevent data from being stolen or tampered with during transmission.

[0099] Data Storage and Analysis System: A high-performance, scalable data storage platform is established to store a large amount of book codes and related information transmitted from the code scanning and data collection system. The data storage platform can use ClickHouse clusters for efficient querying and analysis; some unstructured data such as pictures, audio, and videos (such as promotional materials related to books, author audio explanations, etc.) are stored in a distributed file system.

[0100] At the same time, the data storage platform should have data backup and recovery functions, regularly backup data, and store backup data in off-site data centers to prevent data loss due to local data center failures.

[0101] Data Analysis and Mining: Using big data analysis techniques and data mining algorithms, the book codes and related information stored in the data storage platform are analyzed in depth. For example, the books of the present application can be recommended, the scanning frequency and purchase intention of different regions, different time periods, and different reader groups for various books can be analyzed to understand the market demand and popular trends of books; by analyzing the scanning behavior and interaction data of readers, the interests and reading habits of readers can be mined to provide accurate reader portraits and personalized marketing suggestions for publishing institutions.

[0102] The data analysis results can be presented to the management and marketing personnel of the publishing agency in the form of visual reports, charts, etc., so that they can intuitively understand the market dynamics and reader feedback of the book, and timely adjust the publishing strategy and marketing plan to improve the sales performance and market competitiveness of the book.

[0103] Marketing activity planning and execution: Based on the one-book-one-code, the publishing agency can plan various forms of marketing activities, such as code scanning for prize drawing, points exchange, book recommendation reward, etc. Through setting up corresponding activity entrance and rules in the code scanning application, readers are attracted to participate in the activities, and the popularity and sales of the book are improved.

[0104] For example, when a new book is launched, a code scanning for prize drawing activity can be carried out, and readers can participate in the drawing by scanning the book code, with the opportunity to win book coupons, physical prizes, or the opportunity to interact with the book author, etc. At the same time, the publishing agency can set different prize levels and winning probabilities according to the target and budget of the activity, to ensure the attractiveness and feasibility of the activity.

[0105] Reader interaction and community building - using the code scanning application to build a reader interaction platform and community, encouraging readers to post book reviews, experiences, insights, etc. after reading the book, and interacting with other readers. The publishing agency can arrange special personnel to manage and reply to the interaction content of the readers, enhance communication and contact with the readers, and create a good reading atmosphere and community culture.

[0106] In addition, the publishing agency can also collect feedback and suggestions from readers through the reader interaction platform, understand the satisfaction and improvement needs of readers on the content, layout, printing, etc. of the book, and provide reference for the reprinting and optimization of the book.

[0107] Security and privacy protection system:

[0108] I. Data security protection measures

[0109] Various security protection technologies are adopted to ensure the security of data in the one-book-one-code related system. For example, at the network level, firewalls, intrusion detection systems (IDS) and intrusion prevention systems (IPS) are deployed to prevent external network attacks and illegal access; at the server side of the system, operating system security reinforcement, database access control, data encryption storage and other technologies are adopted to ensure the security of the server and the confidentiality of the data.

[0110] At the same time, the code scanning application is subjected to security vulnerability scanning and repair to prevent the application program from being attacked by hackers or malicious tampering, and to ensure the information security of users when using the code scanning application.

[0111] II. Privacy protection policy formulation and implementation

[0112] Establish a clear privacy protection policy to inform readers about the types of data collected, uses, storage methods, and data sharing and protection measures during the use of the Book One Code related system. Ensure that the personal information and privacy of readers are fully respected and protected, and do not use their personal information for other commercial purposes or share it with third parties without their consent.

[0113] Establish a privacy complaint handling mechanism to timely receive and properly handle readers' complaints and suggestions when they have doubts or dissatisfaction about personal information processing, and to protect the legitimate rights and interests of readers.

[0114] Through the above technical solutions of Book One Code, the whole life cycle management of books can be realized, including coding generation, printing and issuing, code scanning and identification, data collection, analysis and mining, marketing interaction, and security and privacy protection, etc. It provides strong technical support and innovation power for the development of the book industry.

[0115] Based on Book One Code, the following functions can be realized:

[0116] 1. Precise inventory management

[0117] Real-time inventory tracking: Book One Code technology enables each book to have a unique identity, and by scanning the code, the library or bookstore can record the in and out of the book, and can real-time inventory quantity. Whether it is a new book, book lending, return or book transfer, inventory information can be updated immediately. For example, when books are transferred between branches of a large library, scanning the book code during the handover process allows the central library's inventory management system to immediately know the flow of books and the actual inventory of each branch, avoiding errors and delays that may occur during manual inventory.

[0118] Precise inventory positioning: In addition to quantity management, Book One Code can also help staff quickly locate the specific location of a book. In a large library or warehouse with complex shelving layout, scanning the book code can display the book's shelf area, layer, and specific location number. This greatly improves the efficiency of book searching and organizing, reduces the workload of staff, and also facilitates readers to quickly find the books they need.

[0119] 2. Effectively combat piracy and illegal circulation

[0120] Convenient authentication: Consumers or library administrators can quickly verify the authenticity of a book by scanning the unique code on the book. Because the codes of genuine books are uniformly managed and distributed by publishers or copyright owners, it is difficult for pirate books to replicate the same code information. For example, in the book sales channel, bookstore staff scan the book code when they receive the goods, and the system can automatically determine whether the book is genuine, and timely discover and prevent pirate books from entering the sales link. When the book information obtained by scanning the code does not match the actual book information, it is determined that the current book is fake.

[0121] Tracking circulation path: One book one code can record the complete circulation path of a book from the publisher to various levels of sales channels and to the hands of the final consumer. If illegal sales or infringement is found, the circulation record of the book can be traced to quickly locate the problem link and determine the responsible party. This is of great significance to maintaining the normal order of the book market and protecting the legitimate rights and interests of copyright owners.

[0122] 3. Enhance reader interaction and service experience

[0123] Personalized reading service: Through scanning, readers can enter the exclusive online platform of the book, and the platform can provide personalized reading recommendations, author interview videos, and reading experience sharing services for readers according to their scanning history and reading preferences, according to the method of the present application. For example, a reader often scans the code of science fiction books, and the system will recommend more popular science fiction works, related science fiction movies or science fiction theme activity information to him, improving the reading experience and participation of the reader.

[0124] Interactive community building: One book one code builds a bridge for communication between readers. After scanning the code, readers can enter the interactive community of the book and discuss the content of the book, share reading insights, and carry out reading club activities. This interaction not only enhances the reader's understanding and love for the book, but also cultivates the reader's reading habits and a sense of community belonging. At the same time, authors and publishers can collect feedback from readers through the interactive community to improve book content and publishing strategies.

[0125] Optimize marketing and sales strategies

[0126] Precise marketing data collection: Publishers and bookstores can use one book one code to collect marketing data for books, including the geographic location of readers who scan the code, the time of scanning the code, and the purchase intention (whether the reader scans the code or the reader gives the purchase intention on the software). By analyzing these data, we can understand the level of reader interest in books in different regions and time periods, and develop precise marketing activities. For example, if it is found that the scanning rate of a certain new book in a certain city is particularly high, but the purchase conversion rate is low, targeted promotion activities such as discounts and gifts can be carried out in that city,

[0127] 4. Increase the sales volume of books

[0128] Optimization of sales channel management: For the sales channel of books, One Book One Code can help publishers better manage and motivate distributors at all levels. By monitoring the sales of books in different channels, publishers can adjust channel strategies, such as giving more support and rewards to distributors with good sales performance, and optimizing or eliminating channels with poor sales performance. At the same time, it can prevent the behavior of distributors from running between channels, and maintain the stability of market prices. When implementing, the book code of the book is associated with the book code, and the receiving and sending of the book, the distribution record can be tracked throughout the process. Market inspectors can view the flow direction of the book by scanning the code, the system can also automatically record the location information of the reader scanning the code, and compare it with the distribution range of the distributor (for example, the location information of the reader scanning the code is Hebei, and the distribution range of the recorded book distributor is Beijing, it is determined that the book flow direction is abnormal), once the book flow direction is found to be abnormal, a suspected smuggling report can be generated in time, so as to effectively control the smuggling behavior and maintain the normal order of the market.

[0129] 5. Quality traceability

[0130] When the book has quality problems, such as printing errors, binding problems, etc., due to the lack of effective traceability means, it is difficult to quickly determine the source and scope of the problem, resulting in low efficiency in handling problems and increasing costs and risks.

[0131] Solution of One Book One Code: The coding tracking and control function of One Book One Code can comprehensively master the production process and circulation record of the book. When there is a quality problem, scanning the code can quickly track the source and scope of the problem, determine the batch, processing machine, workshop, warehouse management and delivery record, circulation range, etc. of the problem product, and realize quick response and processing, improve the efficiency and accuracy of quality control.

[0132] It should be noted that different personnel correspondingly use systems with different permissions, publishers have the highest permission, and readers only have the permission to provide basic data. The function modules of the systems / software used by different personnel are also different.

[0133] Embodiment two

[0134] The embodiment of the application also provides a book recommendation device, which is mainly used to execute the book recommendation method provided by the above-mentioned content of the embodiment of the application. The book recommendation device provided by the embodiment of the application will be specifically introduced below.

[0135] Figure 2 is a schematic diagram of a book recommendation device according to an embodiment of the application, such as Figure 2As shown, the book recommendation device of the book mainly includes: a first acquisition unit 10, a first determination unit 20, a second determination unit 30, a second acquisition unit 40, a book recommendation unit 50, a third determination unit 60 and a fourth determination unit 70, wherein:

[0136] The first acquisition unit is configured to acquire book information uploaded by each regional book management subsystem and a recommendation degree of each regional book management subsystem, wherein the book information includes: a browsing frequency of each book in each region, a discussion frequency of each book in each region, a borrowing frequency of each book in each region, and feature information of each book in each region.

[0137] The first determination unit is configured to determine a heat of each book in each region based on the browsing frequency of each book in each region and the discussion frequency of each book in each region.

[0138] The second determination unit is configured to determine a popularity of each book in each region based on the borrowing frequency of each book in each region and the browsing frequency of each book in each region.

[0139] The second acquisition unit is configured to acquire a recommendation request for book recommendation to a target reader in a target region initiated by a target regional book management subsystem, wherein the recommendation request carries reader information of the target reader, and the reader information includes: personal information of the target reader and historical borrowing records of the target reader.

[0140] The book recommendation unit is configured to perform book recommendation on the reader information of the target reader and the feature information of each book in each region by using a book recommendation model to obtain a borrowing probability matrix of each book in each region for the target reader.

[0141] The third determination unit is configured to determine a book recommendation vector of the target reader in each region according to the borrowing probability matrix of each book in each region for the target reader, the heat of each book in each region and the popularity of each book in each region.

[0142] The fourth determination unit is configured to determine a target book recommendation vector of the target reader based on the book recommendation vector of the target reader in each region and the recommendation degree of each regional book management subsystem, and further recommend books for the target reader according to the target book recommendation vector.

[0143] In the embodiment of the present application, a book recommendation device is provided, which is applied to a book management system connected with a plurality of regional book management subsystems. The device comprises: obtaining book information uploaded by each regional book management subsystem and recommendation degrees of each regional book management subsystem, wherein the book information comprises: browsing times of each book in each region, discussion times of each book in each region, borrowing times of each book in each region, and characteristic information of each book in each region; determining the heat of each book in each region based on the browsing times of each book in each region and the discussion times of each book in each region; determining the popularity of each book in each region based on the borrowing times of each book in each region and the browsing times of each book in each region; obtaining a recommendation request for book recommendation to a target reader in a target region initiated by a target regional book management subsystem, wherein the recommendation request carries reader information of the target reader, and the reader information comprises: personal information of the target reader and historical borrowing records of the target reader; performing book recommendation on the reader information of the target reader and the characteristic information of each book in each region by using a book recommendation model to obtain a borrowing probability matrix of each book in each region for the target reader; determining a book recommendation vector of each region for the target reader according to the borrowing probability matrix of each book in each region for the target reader, the heat of each book in each region, and the popularity of each book in each region; determining a target book recommendation vector of the target reader based on the book recommendation vector of each region for the target reader and the recommendation degrees of the regional book management subsystems, and then recommending books for the target reader according to the target book recommendation vector. As can be seen from the above description, in the book recommendation device of the present application, the book recommendation model is used to perform book recommendation on the reader information of the target reader and the characteristic information of each book in each region to obtain a borrowing probability matrix of each book in each region for the target reader. In this process, the historical borrowing records of the target reader are considered, that is, the borrowing probability of each book in each region for the target reader is first determined according to the historical borrowing records of the target reader, and then the borrowing probability is corrected according to the heat of each book in each region and the popularity of each book in each region to obtain a book recommendation vector of each region for the target reader. In addition, the book recommendation vector of each region for the target reader is further corrected by the recommendation degrees of the regional book management subsystems to obtain a final target book recommendation vector of the target reader. Finally, books are recommended for the target reader according to the target book recommendation vector. The above process takes the heat of each book in each region and the popularity of each book in each region as consideration indexes for book recommendation, which can improve the accuracy of book recommendation. In addition, the target book recommendation vector of the target reader determined by the joint recommendation of the plurality of regional book management subsystems will be more accurate, which can accurately recommend books for the target reader and achieve good book recommendation effect, thereby solving the technical problem that the existing book recommendation method has poor accuracy in recommending books and is difficult to achieve good recommendation effect.

[0144] Optionally, the first determining unit is further configured to calculate the hotness of each book in each area according to a hotness calculation formula The hotness of each book in each area is calculated, wherein hot ij represents the hotness of book i in area j, scan_num ij represents the number of scans of book i in area j, dis_num ij represents the number of discussions of book i in area j, damand_rate ij represents the demand degree of book i in area j.

[0145] Optionally, the second determining unit is further configured to calculate the borrowing rate of each book in each area according to a borrowing rate calculation formula The borrowing rate of each book in each area is calculated, wherein borrow_scan_rate ij represents the borrowing rate of book i in area j, borrow_num ij represents the number of borrows of book i in area j, scan_num ij represents the number of scans of book i in area j; according to a popularity calculation formula The popularity of each book in each area is calculated, wherein welcome_rate ij represents the popularity of book i in area j, borrow_scan_rate ij represents the borrowing rate of book i in area j, damand_rate ij represents the demand degree of book i in area j.

[0146] Optionally, the book recommendation vector of the target reader in each area is a score vector of each book of each area of the target reader, and the third determining unit is further configured to calculate the score of each book of each area of the target reader according to a score calculation formula score ij = w ij * hot ij * welcome_rate ij The score of each book of each area of the target reader is calculated, and further the score vector of each book of each area of the target reader is obtained, wherein score ij represents the score of book i in area j of the target reader, w ij represents the borrowing probability of book i in area j of the target reader, hot ij represents the hotness of book i in area j, welcome_rate ij represents the popularity of book i in area j.

[0147] Optionally, the fourth determining unit is further configured to calculate the target book recommendation vector according to a target book recommendation vector calculation formula The target book recommendation vector of the target reader is calculated, wherein U represents the target book recommendation vector of the target reader, z represents the number of regional book management subsystems, w k represents the recommendation degree of the regional book management subsystem k, S k represents the book recommendation vector of the target reader in the region corresponding to the regional book management subsystem k.

[0148] Optionally, the device is further used to: acquire the actual borrowing message of the target reader; if the target reader does not borrow, the recommendation degrees of the regional book management subsystems remain unchanged; if the target reader borrows books, the recommendation degrees are updated according to the recommendation degree updating formula The updated recommendation degrees of the regional book management subsystems are calculated, wherein w k represents the updated recommendation degree of the regional book management subsystem k, w k represents the recommendation degree of the regional book management subsystem k, represents the recommendation degree of the target regional book management subsystem, and θ is a positive number less than 1, β represents whether the target reader borrows the books recommended by the target regional book management subsystem, β = 0 represents that the target reader borrows the books recommended by the target regional book management subsystem, β = 1 represents that the target reader does not borrow the books recommended by the target regional book management subsystem, U represents the target book recommendation vector of the target reader, and a represents the book vector actually borrowed by the target reader.

[0149] Optionally, the book information is obtained based on scanning of the book code by the user, and the book code comprises a publishing organization identification code, a book category code, a publishing batch code and a serial number, and the book code has uniqueness.

[0150] The device provided by the embodiment of the application has the same implementation principle and technical effects as the foregoing method embodiment, and for brevity of description, the part not mentioned in the device embodiment can refer to the corresponding content in the foregoing method embodiment.

[0151] As shown in Figure 3 The electronic device 600 provided by the embodiment of the application comprises a processor 601, a memory 602 and a bus, the memory 602 stores machine readable instructions executable by the processor 601, the processor 601 and the memory 602 communicate through the bus when the electronic device is running, and the processor 601 executes the machine readable instructions to perform the steps of the book recommendation method described above.

[0152] Specifically, the memory 602 and the processor 601 can be general memory and processor, which are not specifically limited here, and when the processor 601 runs the computer program stored in the memory 602, the book recommendation method described above can be executed.

[0153] The processor 601 can be an integrated circuit chip having a processing capability of signals. In the implementation process, each step of the above method can be completed by the integrated logic circuit of hardware in the processor 601 or the instruction in the form of software. The processor 601 described above can be a general processor, including a central processing unit (CPU), a network processor (NP), etc.; can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component. Each method, step and logic block diagram disclosed in the embodiments of the present application can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as a hardware code processor for execution, or a combination of hardware and software modules in the code processor for execution. The software module can be located in a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium in the art. The storage medium is located in the memory 602, and the processor 601 reads the information in the memory 602, and combines the hardware to complete the steps of the above method.

[0154] Corresponding to the above book recommendation method, the embodiments of the present application also provide a computer readable storage medium, the computer readable storage medium stores machine executable instructions, when the processor calls and runs the computer executable instructions, the computer executable instructions make the processor run the steps of the above book recommendation method.

[0155] The book recommendation device provided by the embodiments of the present application can be specific hardware on the device or software or firmware installed on the device. The device provided by the embodiments of the present application has the same implementation principle and technical effects as the above method embodiments. For the sake of brevity, the part of the device embodiment not mentioned in the description can refer to the corresponding content in the above method embodiments. Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system, device and unit described above can refer to the corresponding process in the above method embodiments, which will not be repeated here.

[0156] In the embodiments of the present application, it should be understood that the disclosed apparatus and method can be implemented in other manners. The embodiments described above are merely specific implementation manners of the present application, and for example, the division of the units is only a logical function division, and there can be another division manner in actual implementation; for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, or the among different units, can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electric, mechanical or other forms.

[0157] For another example, the flowcharts and block diagrams in the drawings show the possible implementation architectures, functions and operations of the apparatus, method and computer program product according to the embodiments of the present application. In this regard, each block in the flowcharts or block diagrams can represent a module, a program segment or a part of code, which contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in different orders from that shown in the drawings. For example, two consecutive blocks can actually be executed substantially in parallel, and sometimes they can be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and the combination of blocks in the block diagrams and / or flowcharts, can be implemented by a dedicated hardware-based system for executing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0158] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e. can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the present embodiment.

[0159] In addition, each functional unit in the embodiments of the present application can be integrated into one processing unit, or each unit can exist physically as a separate unit, or two or more units can be integrated into one unit.

[0160] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the present application that essentially contribute to the prior art or the parts of the technical solutions of the present application can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing an electronic device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the book recommendation method described in various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0161] It should be noted that: similar reference numbers and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings, in addition, the terms "first", "second", "third" and the like are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0162] Finally, it should be noted that: the above-described embodiments are only specific embodiments of the present application, which are used to illustrate the technical solutions of the present application, but not to limit them, the protection scope of the present application is not limited thereto, although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art within the technical scope disclosed by the present application can still modify or easily think of changes to the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to some technical features thereof; and these modifications, changes or replacements do not make the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application. All should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for recommending books, characterized in that, Applied to a library management system, which is connected to multiple regional library management subsystems, the method includes: The system obtains book information uploaded by the regional book management subsystems and the recommendation rate of the regional book management subsystems. The book information includes: the number of times each book in each region has been viewed, the number of times each book in each region has been discussed, the number of times each book in each region has been borrowed, and the feature information of each book in each region. The popularity of each book in each region is determined based on the number of times each book is viewed and the number of times each book is discussed in each region. The popularity of each book in each region is determined based on the number of times each book is borrowed and viewed in each region. The system obtains a recommendation request initiated by the target area library management subsystem to recommend books to target readers in the target area. The recommendation request carries the reader information of the target reader, which includes: the target reader's personal information and the target reader's historical borrowing records. A book recommendation model is used to recommend books based on the reader information of the target readers and the feature information of each book in each region, thereby obtaining the borrowing probability matrix of each book in each region for the target readers; The book recommendation vector for the target readers in each region is determined based on the borrowing probability matrix of each book in each region, the popularity of each book in each region, and the popularity of each book in each region. Based on the book recommendation vector of the target reader in each region and the recommendation degree of the book management subsystem in each region, the target book recommendation vector of the target reader is determined, and then books are recommended to the target reader according to the target book recommendation vector.

2. The method according to claim 1, characterized in that, The popularity of each book in each region is determined based on the number of views and discussions of each book in each region, including: According to the formula for calculating heat The popularity of each book in each region was calculated, where hot ij This represents the popularity of book i in region j, where scan_num ij dis_num represents the number of times book i in region j has been viewed. ij Damand_rate represents the number of times book i in region j is discussed. ij This indicates the degree of urgency for book i in region j.

3. The method according to claim 1, characterized in that, The popularity of each book in each region is determined based on the number of times each book is borrowed and viewed in each region, including: Calculation formula based on borrowing ratio Calculate the borrowing rate of each book in each region, where borrow_scan_rate ij Borrow_num represents the borrowing rate of book i in region j. ij scan_num represents the number of times book i in region j has been borrowed. ij This represents the number of times book i in region j has been viewed; Calculation formula based on popularity The popularity of each book in each region was calculated, where welcome_rate ij Borrow_scan_rate represents the popularity of book i in region j. ij Damand_rate represents the borrowing rate of book i in region j. ij This indicates the degree of urgency for book i in region j.

4. The method according to claim 1, characterized in that, The book recommendation vector for the target reader in each region is the rating vector of the target reader for each book in each region. The book recommendation vector for the target reader in each region is determined based on the borrowing probability matrix of each book in each region, the popularity of each book in each region, and the overall popularity of each book in each region. This includes: The score is calculated using the scoring formula. ij =w ij *hot ij *welcome_rate ij Calculate the target readers' ratings for each book in each region, and then obtain the target readers' rating vectors for each book in each region, where score ij w represents the rating of book i in region j by the target reader. ij Hot represents the probability that the target reader will borrow book i in region j. ij The welcome_rate represents the popularity of book i in region j. ij This indicates the popularity of book i in region j.

5. The method according to claim 1, characterized in that, The target book recommendation vector for the target reader is determined based on the book recommendation vector of the target reader in each region and the recommendation degree of the book management subsystem in each region, including: Calculation formula based on target book recommendation vector Calculate the target book recommendation vector for the target reader, where U represents the target book recommendation vector for the target reader, z represents the number of regional book management subsystems, and w k S represents the recommendation level of the regional library management subsystem k. k This represents the book recommendation vector for the target reader in the region corresponding to region k in the regional library management subsystem.

6. The method according to claim 1, characterized in that, After recommending books to the target reader based on the target book recommendation vector, the method further includes: Obtain the actual borrowing information of the target reader; If the target reader does not borrow the book, the recommendation level of each regional library management subsystem remains unchanged. If the target reader borrows the book, the algorithm is updated based on the recommendation level. Calculate the recommendation score of each regional library management subsystem after the update, where w' k w represents the updated recommendation level of the regional library management subsystem k. k This represents the recommendation level of the regional library management subsystem k. This represents the recommendation score of the target area's library management subsystem, where θ is a positive number less than 1. β represents whether the target reader has borrowed a book recommended by the target area book management subsystem. β = 0 indicates that the target reader has borrowed a book recommended by the target area book management subsystem, and β = 1 indicates that the target reader has not borrowed a book recommended by the target area book management subsystem. U represents the target book recommendation vector of the target reader, and a represents the vector of books actually borrowed by the target reader.

7. The method according to claim 1, characterized in that, The book information is obtained based on statistics obtained after users scan the book codes. The book codes include: publisher identification code, book category code, publication batch code, and serial number. The book codes are unique.

8. A book recommendation device, characterized in that, The device is applied to a library management system, which is connected to multiple regional library management subsystems, and includes: The first acquisition unit is used to acquire book information uploaded by the book management subsystems of each region and the recommendation level of the book management subsystems of each region. The book information includes: the number of times each book in each region has been viewed, the number of times each book in each region has been discussed, the number of times each book in each region has been borrowed, and the feature information of each book in each region. The first determining unit is used to determine the popularity of each book in each region based on the number of times each book is viewed and the number of times each book is discussed in each region. The second determining unit is used to determine the popularity of each book in each region based on the number of times each book is borrowed and the number of times each book is viewed in each region. The second acquisition unit is used to acquire a recommendation request initiated by the target area library management subsystem to recommend books to target readers in the target area. The recommendation request carries the reader information of the target reader, which includes: the target reader's personal information and the target reader's historical borrowing records. The book recommendation unit is used to recommend books based on the reader information of the target reader and the feature information of each book in each region using a book recommendation model, and to obtain the borrowing probability matrix of each book in each region for the target reader; The third determining unit is used to determine the book recommendation vector for the target reader in each region based on the borrowing probability matrix of each book in each region, the popularity of each book in each region, and the popularity of each book in each region. The fourth determining unit is used to determine the target book recommendation vector of the target reader based on the book recommendation vector of the target reader in each region and the recommendation degree of the book management subsystem in each region, and then recommend books to the target reader according to the target book recommendation vector.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer storage medium, characterized in that, It stores a computer program thereon, and when the computer runs the computer program, it performs the steps of the method according to any one of claims 1 to 7.

Images