A preschool children picture book intelligent recommendation method and a picture book management system

Abstract

The present application belongs to the field of data analysis, and particularly relates to a pre-school children picture book intelligent recommendation method, a picture book management system and a reading device thereof. The recommendation method recommends new books suitable for and meeting the current user's preferences to the current user according to all the user's personal information and reading records. The recommendation method comprises the following steps: S1: obtaining all the picture books and score records of the picture books that the current user has seen or shared, to obtain a seen list and a score table; S2: obtaining all the picture books and score records of the picture books that all the similar users have seen or shared, to obtain a reference list and a score table; S3: constructing a tag similarity matrix; S4: generating a user preference matrix; S5: calculating the final preference score of the current user for each picture book in the reference list; and S6: reordering the reference list to obtain a recommended book bag. The present application solves the problem that the existing picture book recommendation method is not suitable for pre-school children reading recommendation.

Description

Technical Field

[0001] This invention belongs to the field of data analysis, specifically relating to an intelligent recommendation method for preschool children's picture books, a picture book management system, and a reading device thereof. Background Technology

[0002] Studies have shown that cultivating early reading habits in preschool children is crucial for their subsequent learning and development. From the perspective of protecting children's eyesight, it is generally recommended that preschool children read printed picture books, avoiding early exposure to e-books. At the same time, preschool picture books are typically short, requiring fast reading speeds and necessitating a large volume of reading material. Parents need to frequently provide children with new picture books for early education, and how to choose picture books suitable for early childhood development is a common concern for most parents of infants and toddlers.

[0003] There are many existing methods for picture book recommendation, mainly falling into two categories: user preference-based methods and picture book content-based methods. User preference-based methods typically utilize collaborative filtering algorithms, which can be further divided into user-based and picture book-based collaborative filtering recommendation methods. User-based methods focus on recommending picture books to a specific user based on known user preferences and the similarity of preferences among different users. Picture book-based collaborative filtering methods typically evaluate the similarity between multiple picture books based on user preferences or inherent attributes, recommending more relevant picture books based on the specified user's existing preferences. Content-based methods, on the other hand, typically rely on the picture book's own features, attributes, or textual and other characteristics to identify other picture books with similar properties.

[0004] Existing book recommendation algorithms and systems can effectively recommend useful or interesting books to adult users to a certain extent, but their focus is usually on text-based book recommendations. Preschool picture books differ from general books. Preschool children are generally more sensitive to pictures, and during parental guidance in reading, text can more effectively convey the theme and content of the picture book, thus helping parents better assist children in reading. Furthermore, the selection and recommendation of preschool picture books must be combined with children's developmental characteristics. For example, in the early stages of infant development, children are not sensitive to color in books and therefore cannot read colored picture books; at certain periods, they develop a greater interest in facial structures; in addition, picture books can be used to guide children in different stages of preschool development, helping them to understand themselves and their environment, cultivate emotions and feelings, and develop good habits. This invention focuses on these characteristics of children's picture book recommendations when constructing the system, thus improving recommendation performance. Summary of the Invention

[0005] To address the problem that existing book recommendation methods are not suitable for recommending books for preschool children, this invention provides an intelligent recommendation method for preschool children's picture books, a picture book management system, and a reading device thereof.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solution:

[0007] A method for intelligently recommending picture books for preschool children is proposed. This method recommends suitable new books that meet the current user's preferences based on the reading records of the current user and other similar users in a picture book database.

[0008] The intelligent recommendation method includes the following steps:

[0009] S1: Retrieve all picture books that the current user has viewed or shared, along with their rating records. This results in a viewed list A = {A...} i}, i = 1...m, and its rating list S = {S i}, i = 1...m.

[0010] Among them, A i This indicates the picture books that the current user has viewed or shared. i This indicates that the current user is interested in picture book A. i The rating; m represents the number of picture books in the read list.

[0011] S2: Retrieve all picture books viewed or shared by all similar users of the current user, along with their rating records, and remove picture books that the current user has already viewed or shared. This results in a reference list B = {B...} j}, j = 1...n, and its rating list V = {V j}, j = 1...n.

[0012] Among them, B j This indicates picture books that similar users have viewed or shared but that the current user has not viewed. (V) j This indicates all similar users' views on picture book B. j The average rating; n represents the number of picture books in reference list B.

[0013] S3: Obtain the tags of all picture books in the read list A and the reference list B, and construct a tag similarity matrix E based on the tag content of each picture book.

[0014] The label similarity matrix E is used to characterize the degree of type similarity between the viewed list A and the reference list B; the label similarity E is as follows:

[0015]

[0016] Among them, the elements L in the label similarity matrix E ij Picture Book A i and B j The sum of the tag weights of the shared tags.

[0017] S4: Combine the tag similarity matrix E with the user rating data of all picture books in the read list A and the reference list B to generate a user preference matrix F.

[0018] The user preference matrix F is used to characterize the current user's preference for the picture books in reference list B; the user preference matrix F is as follows:

[0019]

[0020] Among them, the elements W in the user preference matrix F ij This represents the predicted rating value of the current user for the picture books in reference list B, calculated by combining the ratings from different users and the tag similarity E.

[0021] S5: Calculate the current user's preference for each picture book in reference list B based on the user preference matrix F. j The final preference score G j G j The calculation formula is as follows:

[0022] G j =max i＝1,...,m {W ij}

[0023] S6: Based on the final preference score G for each picture book in reference list B. j The picture books are reordered, and the top K picture books with the highest final preference scores are selected and pushed to the current user in order to form the required recommended book bag.

[0024] As a further improvement of the present invention, in step S1, picture books that the current user has already viewed refer to picture books that are included in the current picture book database and have been read by the current user. Picture books that the current user has already shared refer to picture books that are not included in the current picture book database but have been viewed and shared by the current user. When any user shares a picture book, they must upload the basic information of the picture book and the user's rating of the picture book to the current picture book database. The basic information of the picture book includes: name, author, picture book number, publication information, tags, and cover image.

[0025] As a further improvement of this invention, a label is used to mark the type information of picture books. The label for each picture book includes the category of the picture book and user-defined attribute descriptions related to the content of the picture book. Among them, the categories of picture books include science exploration, wordless, bilingual, life habits, character development, humanities and arts, and black and white picture books.

[0026] User-defined attribute descriptions may include:

[0027] (1) Keywords that characterize the length of text content in picture books.

[0028] (2) Keywords that describe the content of the illustrations in the picture book.

[0029] (3) Keywords that characterize the difficulty of reading picture books.

[0030] (4) Keywords that characterize the attributes of characters or plots in picture books.

[0031] (5) Keywords describing picture book awards, such as: Hans Christian Andersen Award, Caldecott Medal, Feng Zikai Picture Book Award, etc.

[0032] (6) Keywords describing information about picture book publishers;

[0033] (7) Keywords that characterize the material style of picture books, such as cloth books, pop-up books, etc.

[0034] As a further improvement of the present invention, in step S2, similar users refer to users who are the same age / age group and gender as the current user. The user's age and gender are actively uploaded by the user when registering a user account used to access the current picture book database.

[0035] As a further improvement of the present invention, in step S3, in the label similarity matrix E, the label weights and L ij The calculation formula is as follows:

[0036]

[0037] In the above formula, c represents picture book A. i Or picture book B j The label contains: C represents picture book A. i And picture book B j The collection of all tags. It is a tool used to determine whether label c is a picture book A. i And picture book B j The discriminant function for common labels; N c p represents the number of picture books that share the common label c; cThis indicates the influence weight of tag c. The more picture books that have tag c, the smaller the influence of tag c.

[0038] As a further improvement of the present invention, in step S4, the predicted rating value W is calculated in the user preference matrix F. ij The calculation formula is as follows:

[0039] W ij =S i *V j *L ij .

[0040] As a further improvement to the present invention, in step S6, during the reordering stage of the reference sequence B, the final preference score G for each book is determined. j Using the reference sort as the sorting benchmark, any one of the following sorting methods—bubble sort, selection sort, insertion sort, shell sort, merge sort, quick sort, heap sort, counting sort, bucket sort, and radix sort—is used to arrange the reference sequence B in descending order of the final preference score.

[0041] The present invention also includes a preschool children's picture book management system, which includes a preschool children's picture book intelligent recommendation system comprising two parts: a backend server and a user terminal.

[0042] The backend server contains a picture book database. This database includes the content and basic information of all included picture books, as well as each registered user's reading history and ratings. The database also stores basic information and user reviews of picture books uploaded by any registered user that are not yet included in the database. The backend server is also used to respond to data access requests from verified registered users.

[0043] The client-side interface communicates with the backend server. The client-side sends data requests to the server based on the registered user's instructions, thereby providing registered users with picture book reading, historical browsing statistics, and picture book recommendation services.

[0044] The user-side functional modules include: personal information management, picture book reading, picture book retrieval, picture book sharing, and picture book recommendation, totaling five parts.

[0045] The Personal Information Management page displays the current user's name, age, gender, and other user information; users can also edit their information on this page. The page also displays the user's reading history, picture book ratings, and shared books. The Picture Book Viewing page displays picture books selected by registered users and provides full-text reading services. The Picture Book Search page supports user queries for all picture books in the database, displaying indexed and shared but not yet indexed picture books separately. The Picture Book Sharing page displays a detailed list of picture books shared by users and allows users to share books they have read and recommend to other users in the picture book database. When a user shares a picture book not yet indexed, the user is required to upload basic information about the book and their user rating. Basic book information includes: title, author, picture book number, publication information, tags, and cover image. The picture book recommendation page displays picture books recommended to the current user based on their reading and rating history, as well as that of similar users. The recommended book bag, comprised of these picture books, is automatically generated by the preschool picture book management system using the aforementioned intelligent recommendation method and is proactively displayed when the user enters the picture book recommendation page.

[0046] As a further improvement to this invention, users are required to register an account and upload user information when experiencing the functions of the preschool children's picture book management system. The client only provides related services after a registered user logs in.

[0047] The backend server has dedicated administrators who manually review information uploaded and shared by users that is not included in the picture book database. Once the manual review is passed, the relevant data is stored and displayed.

[0048] The present invention also includes a preschool children's picture book reading device, which runs an application that implements the user-end functions of the preschool children's picture book management system described above, and interacts with a background server located in the cloud during program operation, thereby providing users with picture book reading, picture book sharing or picture book recommendation services.

[0049] The present invention provides a method for intelligent recommendation of picture books for preschool children, a picture book management system, and a reading device thereof, which have the following beneficial effects:

[0050] The picture book recommendation method provided by this invention uses the tags and ratings of picture books that the current user has already read and other users with similar attributes have already read as basic data. It creatively constructs a new quantitative evaluation tool that accurately predicts the current user's evaluation of other unread picture books and uses them as the basis for recommending new picture books.

[0051] This invention provides a picture book recommendation method that can automatically filter out abnormal or unsuitable picture books, ensuring that all recommended picture books are compliant. Furthermore, the method integrates user reviews and content evaluation into the assessment metrics, while also considering multiple user attributes such as age and gender, ensuring that the recommended picture books are those that users like or find useful.

[0052] In particular, the recommendation method provided by this invention can be applied to both online picture book reading and sales, as well as to the management of physical picture books. It is also highly suitable as a recommendation method for picture books for young children or students. Attached Figure Description

[0053] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0054] Figure 1 This is a flowchart of the steps of an intelligent recommendation method for preschool children's picture books provided in Embodiment 1 of the present invention.

[0055] Figure 2 This is a data flowchart illustrating the implementation process of the intelligent recommendation method for picture books for preschool children in Embodiment 1 of the present invention. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0057] Example 1

[0058] This embodiment provides an intelligent recommendation method for preschool children's picture books. This method is used to recommend suitable new books that meet the current user's preferences based on the reading records of the current user and other similar users in the picture book database.

[0059] like Figure 1 As shown, this intelligent recommendation method includes the following steps:

[0060] S1: Retrieve all picture books that the current user has viewed or shared, along with their rating records. This results in a viewed list A = {A...} i}, i = 1...m, and its rating list S = {S i}, i = 1...m.

[0061] Among them, A i This indicates the picture books that the current user has viewed or shared. i This indicates that the current user is interested in picture book A. i The rating; m represents the number of picture books in the read list.

[0062] Picture books currently viewed by the user refer to picture books that are included in the current picture book database and have been read by the current user. Picture books currently shared by the user refer to picture books that are not included in the current picture book database but have been viewed and shared by the current user.

[0063] When any user shares a picture book, they must upload the book's basic information and their user rating to the current picture book database. The basic information includes: title, author, book number, publication information, tags, and cover image.

[0064] The tags in the basic information section are primarily used to mark the type of picture books. Each picture book's tags include its category and user-defined attribute descriptions related to the book's content. These categories include science exploration, wordless, bilingual, lifestyle habits, character development, humanities and arts, and black and white picture books.

[0065] User-defined attribute descriptions may include:

[0066] (1) Keywords that characterize the length of text content in picture books.

[0067] (2) Keywords that describe the content of the illustrations in the picture book.

[0068] (3) Keywords that characterize the difficulty of reading picture books.

[0069] (4) Keywords that characterize the attributes of characters or plots in picture books.

[0070] (5) Keywords describing picture book awards, such as: Hans Christian Andersen Award, Caldecott Medal, Feng Zikai Picture Book Award, etc.

[0071] (6) Keywords describing information about picture book publishers;

[0072] (7) Keywords that characterize the material style of picture books, such as cloth books, pop-up books, etc.

[0073] It is important to emphasize that each picture book can have more than one label, and since the categories of picture books are not distinguished from a single dimension, each picture book's label can also contain more than one category. For example, a purely image-based comic book that mainly presents fables can be classified as both a textless book and a humanities / arts book, thus having two category labels.

[0074] Furthermore, user-defined attribute tags offer greater freedom and richness compared to type tags. These tags can be either commonly used tags selected from the tag library or new tags submitted entirely by the user. For example, users can add tags such as "nursery rhymes," "fairy tales," "poems," "fables," "comics," "Pleasant Goat and Big Big Wolf," "Peter Pan," and "Andersen" based on the theme and content of the picture book.

[0075] S2: Retrieve all picture books viewed or shared by all similar users of the current user, along with their rating records, and remove picture books that the current user has already viewed or shared. This results in a reference list B = {B...} j}, j = 1...n, and its rating list V = {V j}, j = 1...n.

[0076] Among them, B j This indicates picture books that similar users have viewed or shared but that the current user has not viewed. (V) j This indicates all similar users' views on picture book B. j The average rating; n represents the number of picture books in reference list B.

[0077] Similar users refer to users who are the same age / age group and gender as the current user. A user's age and gender are actively uploaded when registering their user account to access the current picture book database.

[0078] The method in this embodiment uses age and gender as key indicators to differentiate users, and these also serve as strong criteria for determining the recommended reading lists for different users. Therefore, the method in this embodiment is very detailed in its age segmentation. Not only is users categorized by age, but even for younger users, considering the potentially large differences in comprehension abilities, users are further segmented by month. Conversely, for older users, considering the smaller differences in comprehension abilities among children of different ages, segmentation is not based on month, but rather on age group. This improves the precision and matching accuracy of the recommended reading lists for different users in this embodiment.

[0079] S3: Obtain the tags of all picture books in the read list A and the reference list B, and construct a tag similarity matrix E based on the tag content of each picture book.

[0080] The label similarity matrix E is used to characterize the degree of type similarity between the viewed list A and the reference list B; the label similarity E is as follows:

[0081]

[0082] Among them, the elements L in the label similarity matrix E ij Picture Book A i and B j The sum of the tag weights of the shared tags.

[0083] Label weight and L ij The calculation formula is as follows:

[0084]

[0085] In the above formula, c represents picture book A. i Or picture book B j The label contains: C represents picture book A. i And picture book B j The collection of all tags. It is a tool used to determine whether label c is a picture book A. i And picture book B j The discriminant function for common labels; N c p represents the number of picture books that share the common label c; c This indicates the influence weight of tag c. The more picture books that have tag c, the smaller the influence of tag c.

[0086] In this embodiment, a newly defined tag similarity matrix is ​​used to quantify the similarity between the content of picture books that the user has read and those that have not. The degree of overlap of tag information between different picture books can be used as one of the feature information in the subsequent picture book recommendation.

[0087] Furthermore, it should be noted that in this embodiment, in addition to the existing picture book tags, the picture books currently shared by the user will also be included as an extended category tag in the calculation of tag similarity, thereby integrating the picture book sharing information of the current user into the similarity calculation process.

[0088] S4: Combine the tag similarity matrix E with the user rating data of all picture books in the read list A and the reference list B to generate a user preference matrix F.

[0089] The user preference matrix F is used to characterize the current user's preference for the picture books in reference list B; the user preference matrix F is as follows:

[0090]

[0091] Among them, the elements W in the user preference matrix F ij This represents the predicted rating value W for the current user's picture books in reference list B, calculated by combining ratings from different users and tag similarity E. Specifically, the predicted rating value W... ij The calculation formula is as follows:

[0092] W ij =S i *V j *L ij .

[0093] In this embodiment, the user preference matrix is ​​actually a rating table used to quantify the current user's preference for picture books in the reference list. The calculation process of this rating table considers both the tag similarity between unread and read picture books (reflecting how "similar" the books are) and the current user's ratings of read books, as well as the ratings of unread books by other similar users (reflecting how "good" the books are). Therefore, the predicted rating values ​​contained in the user preference matrix constructed in this embodiment should be a highly reliable data indicator.

[0094] S5: Calculate the current user's preference for each picture book in reference list B based on the user preference matrix F. j The final preference score G j G j The calculation formula is as follows:

[0095] G j =max i＝1,...,m {W ij}

[0096] In the user preference matrix of this embodiment, each column represents the user's preference assessment of the picture book to be recommended based on one of the picture books the user has already read. The final preference score is calculated by selecting the user's highest preference score for that picture book from the matrix and using that highest preference score as the user's final predicted evaluation of the corresponding picture book.

[0097] S6: Based on the final preference score G for each picture book in reference list B. j The picture books are reordered, and the top K picture books with the highest final preference scores are selected and pushed to the current user in order to form the required recommended book bag.

[0098] During the reordering phase of reference sequence B, the final preference score for each book can be G. jUsing the reference as the sorting benchmark, and employing any one of the following sorting methods—bubble sort, selection sort, insertion sort, shell sort, merge sort, quick sort, heap sort, counting sort, bucket sort, and radix sort—the reference sequence B is arranged in descending order of final preference score.

[0099] Users can customize the number of picture books recommended in the book bag. For example, if a user chooses to recommend 10 picture books at a time, the top 10 picture books from the reordered reference list will be recommended to the user.

[0100] like Figure 2 As shown, the overall working logic of the picture book recommendation method in this embodiment is as follows:

[0101] First, compare the user's already read book list (defined as the read list) with the read book lists of similar users (defined as the unread list). Diverge the two lists and retain picture books that the current user hasn't read but similar users have, forming a reference book list. Then, construct a tag similarity matrix based on the book tags in the reference and read lists. Next, combine the different user ratings from the reference and read lists with the tag similarity matrix to obtain a user preference matrix. Further process the elements in the user preference matrix to obtain the final user rating for each book in the reference book list. Finally, re-rank the reference book list based on the final user ratings for each book, and select the top few books from the re-ranked reference book list as recommended reading materials for the current user.

[0102] The picture book recommendation method provided in this embodiment can be applied to online electronic picture book recommendations as well as offline recommendations of physical picture books in library collections. The only difference from existing physical picture book lending systems is that, to implement this recommendation method, users need to provide a rating of any picture book they have finished reading to the library management. Furthermore, for younger children, this task can be handled by their guardians.

[0103] Example 2

[0104] Building upon Example 1, this example further provides a preschool children's picture book management system. This intelligent recommendation system for preschool children's picture books comprises two parts: a backend server and a user terminal. The backend server contains a picture book database. The database contains the content and basic information of all included picture books, as well as each registered user's reading records and evaluation results for the included picture books. The database also stores basic information and user evaluations of picture books uploaded by any registered user but not included in the database. The backend server is also used to respond to data access requests from verified registered users. The user terminal communicates with the backend server. The user terminal sends data requests to the server according to the registered user's instructions, thereby providing registered users with picture book reading, historical browsing information statistics, and picture book recommendation services.

[0105] The user-side functional modules include: personal information management, picture book reading, picture book retrieval, picture book sharing, and picture book recommendation, totaling five parts.

[0106] The personal information management page displays the current user's name, age, gender, and other user information; users can also edit their information on this page. The page also provides statistics and displays on the user's reading history, picture book ratings, and shared books. However, the information displayed on this page is only brief; for detailed information, users need to click on the relevant item to navigate to another page.

[0107] The picture book reading page displays the picture books selected by registered users and provides them with full-text reading services. This page is equivalent to a local library in other online reading systems or devices; in practical applications, it can be displayed in the form of a bookshelf. Users can also customize or operate the page with features such as "page turning method," "background color," "one-click access to reading history," and "page navigation." Furthermore, the picture books currently being read on this page can also offer options for online download and offline reading.

[0108] The picture book search page supports user queries for all picture books in the database. Considering that the picture book database in this embodiment can support user-uploaded "shared books," the system displays indexed picture books and user-shared but not yet indexed picture books in separate columns during the search process. For indexed picture books, users can view basic information, browse user reviews, and view the full text after the book is retrieved. For unindexed picture books, users can only view the uploaded basic information and the sharing user's evaluation of the book.

[0109] The picture book sharing page displays a detailed list of picture books shared by users and allows users to share books they have read and recommend to other users in the picture book database. When a user shares a picture book not included in the database, the user is also required to upload basic information about the book and a user rating of the book. The basic information about the picture book mentioned in this embodiment includes: name, author, picture book number, publication information, tags, and cover image.

[0110] The picture book recommendation page displays picture books recommended to the current user based on their reading and rating records, as well as those of other similar users. The recommended book bag, comprised of these picture books, is automatically generated by the preschool children's picture book management system using the intelligent recommendation method described in Example 1, and is proactively displayed when the user enters the picture book recommendation page.

[0111] It should be noted that, in order to facilitate the collection of each user's book reading data and thus provide services to other users, the preschool children's picture book management system provided in this embodiment requires all users to register an account and upload user information before experiencing the system's functions. The user interface only provides relevant services after a registered user logs in.

[0112] In the system provided in this embodiment, the backend server has dedicated administrators who manually review information uploaded and shared by users that includes picture books not included in the picture book database. After the manual review is passed, the relevant data is stored and displayed. This is to prevent users from sharing incorrect data or uploading illegal materials, which could negatively impact other users.

[0113] In practical applications, the preschool children's picture book management system provided in this embodiment can be implemented using a web-based or mobile app for online reading, or it can be implemented using an online management platform based on an existing physical picture book library. It can even be implemented using a management system based on online book sales platforms (such as Amazon, Dangdang, JD.com, etc.), thus only providing picture book recommendation services. This embodiment does not limit the specific form of implementing this solution.

[0114] Example 2

[0115] Building upon Embodiments 1 and 2, this embodiment further provides a preschool children's picture book reading device, which is an online reading terminal device, i.e., an e-book reader. It can be a desktop reading device, such as an online reader connected to the internet, similar to those in a picture book library. It can also be a handheld device, such as a tablet computer, mobile phone, e-reader, or other online reading terminal product. This terminal device has the complete capability to implement the user terminal functions described in Embodiment 2.

[0116] Specifically, the online reading terminal device in this embodiment can be a smartphone, tablet computer, laptop computer, desktop computer, rack server, blade server, tower server, or cabinet server (including a standalone server or a server cluster composed of multiple servers), etc., capable of executing programs. The computer equipment in this embodiment includes, but is not limited to, a memory and a processor that can be interconnected via a system bus.

[0117] In this embodiment, the memory (i.e., the readable storage medium) includes flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory can be an internal storage unit of a computer device, such as the hard disk or RAM of the computer device. In other embodiments, the memory can also be an external storage device of the computer device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the computer device. Of course, the memory can also include both internal storage units and external storage devices of the computer device. In this embodiment, the memory is typically used to store the operating system and various application software installed on the computer device. In addition, the memory can also be used to temporarily store various types of data that have been output or will be output.

[0118] In some embodiments, the processor may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip.

[0119] The processor is typically used to control the overall operation of computer devices. In this embodiment, the processor is used to run program code stored in memory or process data, and then executes the application program of the preschool children's picture book management system in Embodiment 1 when the program is running, and interacts with the back-end server located in the cloud when the program is running, thereby providing users with picture book reading, picture book sharing or picture book recommendation services.

[0120] It should be emphasized that the terminal device provided in this embodiment needs to have network connectivity to support communication and data interaction with the backend server. It also needs to have an independent display, or, if it does not have a display, it can connect to other displays.

[0121] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligently recommending picture books for preschool children, which recommends suitable new books that meet the current user's preferences based on the reading records of the current user and other similar users in a picture book database; characterized in that, The intelligent recommendation method includes the following steps: S1: Retrieve all picture books that the current user has viewed or shared, along with their rating records; this will result in a "read list" containing all picture books that the current user has viewed or shared. A={A i }, i=1……m , and its rating list S ={S i }，i =1……m ; in, A i This indicates the picture books that the current user has read or shared. S i This indicates the current user's preference for picture books. A i The rating; m This indicates the number of picture books in the viewed list; S2: Retrieve all picture books viewed or shared by all similar users of the current user, along with their rating records, and remove picture books that the current user has already viewed or shared; thus obtaining a reference list containing all picture books viewed or shared by similar users. B={B j }, j=1……n and its rating list V={V j }, j=1……n ; Similar users refer to users who are the same age / age group and gender as the current user; B j This indicates picture books that similar users have viewed or shared but that the current user has not viewed. V j This indicates all similar users' views on picture books. B j The average score; n Reference list B The number of picture books in the country; S3: Get the list of read items A and reference list B The tags for all picture books in the database are collected, and a tag similarity matrix is ​​constructed based on the tag content of each picture book. E The tag similarity matrix E Used to represent the viewed list A and reference list B The degree of similarity in the types of picture books; Among them, the label similarity matrix E elements in L ij Picture books A i and B j The sum of the tag weights of the shared tags; S4: Fusion Tag Similarity Matrix E and the list of read items A and reference list B Generate a user preference matrix from user rating data for all picture books in the database. F The user preference matrix F Used to characterize the current user's reference list B The degree of preference for Chinese picture books; Among them, the user preference matrix F elements in W ij This indicates a combination of ratings and tag similarity from different users. E Calculated current user reference list B Predicted ratings for Chinese picture books; W ij The calculation formula is as follows: ； S5: Based on the user preference matrix F Calculate the current user's reference list B Each picture book in China B j Final preference score G j , G j The calculation formula is as follows: ； S6: According to the reference list B Final preference score for each picture book G j The picture books were reordered, and those with the highest final preference scores were selected. K This picture book is then pushed to the current user in sequence to form the required recommended book bag.

2. The intelligent recommendation method for preschool children's picture books as described in claim 1, characterized in that: In step S1, the picture books that the current user has already read refer to the picture books that have been included in the current picture book database and have been read by the current user. The picture books currently shared by the user refer to picture books that are not included in the current picture book database but have been viewed and shared by the current user. When any user shares a picture book, they must upload the basic information of the picture book and the user's rating of the picture book to the current picture book database. Basic information about picture books includes: title, author, picture book number, publication information, label, and cover image.

3. The intelligent recommendation method for preschool children's picture books as described in claim 2, characterized in that: The tags are used to mark the type information of picture books. Each picture book's tag includes the category of the picture book and user-defined attribute descriptions related to the picture book's content. The categories of picture books include science exploration, wordless, bilingual, lifestyle habits, character development, humanities and arts, and black and white picture books. User-defined attribute descriptions include: (1) Keywords that characterize the length of text content in picture books; (2) Keywords describing the content of the illustrations in the picture book; (3) Keywords that characterize the difficulty level of picture book reading; (4) Keywords that characterize the attributes of characters or plots in picture books; (5) Keywords describing the picture book's award information; (6) Keywords describing picture book publisher information; (7) Keywords that characterize the material style of picture books.

4. The intelligent recommendation method for preschool children's picture books as described in claim 1, characterized in that: In step S2, the user's age and gender are actively uploaded by the user when registering a user account used to access the current picture book database.

5. The intelligent recommendation method for preschool children's picture books as described in claim 4, characterized in that: In step S3, in the label similarity matrix E In the middle, label weight and L ij The calculation formula is as follows: In the above formula, c Picture books A i or picture book B j The tags contained in it; C Picture books A i and picture books B j The collection of all tags. This is a tool used to determine whether label 'c' is a picture book. A i and picture books B j A discriminant function for common labels; Indicates having common tags c The number of picture books; This represents the influence weight of tag c, when it has the tag c The more picture books there are, the less influence the tag c has.

6. The intelligent recommendation method for preschool children's picture books as described in claim 1, characterized in that, In step S6, in the reference list B The reordering phase, based on the final preference for each book. G j Using the reference list as the sorting criterion, employ any one of the following sorting methods: bubble sort, selection sort, insertion sort, shell sort, merge sort, quick sort, heap sort, counting sort, bucket sort, and radix sort. B Arrange them in descending order of final preference score.

7. A preschool children's picture book management system, characterized in that, The intelligent recommendation system for preschool children's picture books includes: The backend server contains a picture book database, which includes the content and basic information of all the picture books included, as well as each registered user's reading records and evaluation results for the included picture books; the picture book database also stores the basic information and user evaluations of picture books uploaded by any registered user that are not included in the picture book database; the backend server is also used to respond to data access requests from verified registered users. The user client communicates with the backend server and sends data requests to the server based on the registered user's instructions, thereby providing registered users with picture book reading, browsing history statistics, and picture book recommendation services. The user client's functional modules include: personal information management, picture book reading, picture book retrieval, picture book sharing, and picture book recommendation. The personal information management page displays the current user's name, age, gender, and other user information, and shows the user's reading history, picture book ratings, and shared books. The picture book reading page displays the picture books selected by the registered user and provides full-text reading services. The picture book retrieval page supports user queries for all picture books in the database, and during the retrieval process, it categorizes included picture books and picture books shared by users but not yet included. The picture book sharing page displays a detailed list of picture books shared by users and supports users in sharing books they have read and recommending to other users in the picture book database. When a picture book shared by a user is not included in the picture book database, the user is also required to upload basic information about the picture book and a user rating of the picture book. The basic information of the picture book includes: name, author, picture book number, publication information, tags, and cover image. The picture book recommendation page displays picture books recommended to the current user based on the system's intelligent evaluation of the current user's and other similar users' reading and rating records. The recommended book bag, consisting of the recommended picture books, is automatically generated by the preschool children's picture book management system using the preschool children's picture book intelligent recommendation method as described in any one of claims 1-6, and is actively displayed when the user enters the picture book recommendation page.

8. The preschool children's picture book management system as described in claim 7, characterized in that: Users need to register an account and upload user information when experiencing the functions of the preschool children's picture book management system; the user terminal only provides related services after the registered user logs in. The backend server has dedicated administrators who manually review information uploaded and shared by users that is not included in the picture book database. After the manual review is passed, the relevant data is stored and displayed.

9. A preschool children's picture book reading device, characterized in that: The preschool children's picture book reading device runs an application that implements the user-end functions of the preschool children's picture book management system as described in any one of claims 7 or 8, and interacts with the back-end server located in the cloud during program execution, thereby providing users with picture book reading, picture book sharing, or picture book recommendation services.

Images