A school library intelligent book recommendation method and system

Abstract

The present invention discloses an intelligent book recommendation method and system for a school library, including collecting and analyzing the information of readers in a database, reading and writing data, judging the readers' hobbies and the types of knowledge that need to be supplemented; classifying the books by subject and subject difficulty according to the book information; analyzing the reader information and analyzing the books needed by the readers; scoring the value and difficulty of the books through the book evaluation information, and sorting the applicability of the books according to the scores; finding books with high relevance to the subject according to the reader information and the book evaluation, and recommending them to the readers. The system includes a reader information management and analysis module, a book information management module, a book evaluation module, and a book recommendation module.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent book recommendation for libraries, and in particular to an intelligent book recommendation method and system for school libraries. Background Art

[0002] The methods of recommending books to readers in the prior art are mainly based on the readers' borrowing history data and the borrowing data of readers of the same type. The readers' reading interests are inferred based on their borrowing history data, and then books that meet their interest characteristics are recommended to the readers; or readers are divided into several types based on their information, and books that their readers of the same type prefer to borrow are recommended to the readers. In the above recommendation methods, there is a problem that the amount of readers' historical borrowing data is small, resulting in low recommendation accuracy. For new readers or readers who borrow books less frequently, the available historical data is very limited, and the system cannot accurately judge their reading interests, and thus cannot accurately recommend books suitable for such readers; in addition, inaccurate reader information will also affect the system's judgment of the reader type. Readers may only fill in a simple and rough form when registering their personal information, and the system cannot accurately judge the reader type, thereby reducing the accuracy of the recommendation. Summary of the invention

[0003] The purpose of the present invention is to provide an intelligent book recommendation method and system for school libraries, where the readers are students, based on the book evaluation system and combined with the characteristics of student readers. The purpose of the present invention is to make the data volume richer and more accurate in the judgment basis of matching readers with books; to enrich the applicability information of books by using the book evaluation system, and to enrich the personal characteristic information of readers by using the student management data and academic management data of the school, so as to achieve more accurate and intelligent book recommendations.

[0004] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows:

[0005] An intelligent book recommendation method for a school library, comprising:

[0006] Step A collects and analyzes the information of readers in the database, reads and writes data, and determines the readers' hobbies and the types of knowledge that need to be supplemented;

[0007] Step B classifies the books by subject and subject difficulty according to the book information;

[0008] Step C analyzes the reader information and the books the reader needs;

[0009] Step D scores the value and difficulty of the books through the book evaluation information, and sorts the applicability of the books according to the scores;

[0010] Step E finds books that are highly relevant to the topic based on reader information and book reviews and recommends them to readers.

[0011] An intelligent book recommendation system for a school library, comprising:

[0012] The student information management and analysis module is used to collect and analyze reader information to determine readers' interests and hobbies and the types of knowledge they need to supplement;

[0013] Book information management module: book data is based on book acquisition and editing information and reader evaluation information;

[0014] The book evaluation module calculates the value score and difficulty score for the book based on the collected book evaluation information;

[0015] The book recommendation module is used to combine the different characteristic information of each reader to recommend books that may be of interest, and to generate suitable study book lists based on the student readers' staged course mastery, so as to achieve accurate and effective recommendations for student readers.

[0016] The present invention has the following advantages:

[0017] It can provide each reader with more accurate and personalized book recommendations, and for student readers, it can recommend book lists that are helpful for professional learning based on their level of mastery of each subject, providing readers with a reference when they don’t know what books they want to read or which books are suitable for their professional level. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 Flowchart of smart book recommendation method for school libraries;

[0019] Figure 2 This is the structure diagram of the intelligent book recommendation system for the school library;

[0020] Figure 3 It is the software block diagram of the school library intelligent book recommendation system architecture. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below in conjunction with embodiments and drawings.

[0022] like Figure 1 As shown in the figure, the process of the intelligent book recommendation method of the school library includes:

[0023] 1) Collect and analyze the information of readers in the database, read and write data, and determine the readers' hobbies and the types of knowledge that need to be supplemented;

[0024] 2) Classify books by subject and difficulty level based on book information;

[0025] 3) Analyze readers’ information and the books they need;

[0026] Analyzing the books readers need based on their majors, class schedules, subject grades, and elective courses, and recommending books related to their majors will help students master subject knowledge more solidly. By analyzing the readers' class schedules, we can understand the knowledge progress that students need to master at this stage, and thus make more accurate recommendations. For example, last semester, students completed a database course, and this semester the library has advanced database-related books, which may be included in the reader's recommendation list by the system.

[0027] Subject scores are mainly used to analyze students' level of course mastery. For example, if a student's score in a certain course is particularly low, it means that the student may not have a solid grasp of the course, and it is necessary to recommend some books suitable as an introduction to the course to lay a foundation for learning the course. On the contrary, if the student's score is very high, the system will consider recommending books in the collection that belong to the advanced stage of the subject to the student to bring out his or her strengths.

[0028] Analysis of elective courses shows that elective courses represent the interests of student readers to a certain extent. The system pushes high-quality library books based on readers' interest areas.

[0029] 4) Score the value and difficulty of books through book evaluation information, and rank the books according to their suitability;

[0030] The evaluation scores mainly come from external ratings and internal ratings. The external ratings are mainly obtained by crawling readers' ratings on external websites, such as Douban Books. The book value rating can be directly based on external reader ratings, while the difficulty rating is mainly calculated by matching keywords in external readers' comments. The initial score is 5 points. For example, words such as "incomprehensible" and "obscure" mean that the book is difficult to understand. If there are evaluations with these keywords, the difficulty score will be increased by 1 point; words such as "easy", "easy" and "entry-level" mean that the book is easy to understand. If there are evaluations with these keywords, the difficulty score will be reduced by 1 point; the lowest score is 0 points, which means it is very easy to read and suitable for entry; the highest score is 10 points, which means it is very difficult to understand. Therefore, in the external rating, the book value rating comes directly from the reader ratings on external websites (0-10 points), and the difficulty is calculated by matching the reader evaluation keywords (0-10 points).

[0031] Internal reader ratings are mainly evaluations of books borrowed by internal readers. Different readers will have different scoring weights for books. For example, a student who performs particularly well in a certain subject will have a higher weight for the books in the subject he is good at, and on the contrary, the weight for the books in the subject where he performs poorly will be lower. The purpose of this design is to make it easier for the system to identify the difficulty and value of books. For example, if a student who ranks first in a course rates a book related to the subject of the course as "very difficult", then this book is likely to be very difficult for other students in the school. Moreover, the higher the level of mastery of the subject, the stronger the book evaluation ability, and the more representative the rating will be.

[0032] Calculation formula:

[0033] R=(1 / (1+rate))*score (1)

[0034] AvgR=avg(R) (2)

[0035] WR=(v / (v+m))×AvgR+(m / (v+m))×C (3)

[0036] Among them, score is the reader's score (lowest 0, highest 10), rate is the ranking ratio of subject scores, for example, if it is in the top 10%, rate = 0.1. If the reader does not have a course in this subject, rate = 1; R represents the score calculated by weighting the subject score ratio; AvgR represents the average score calculated by weighting the subject score ratio; v is the number of people who have evaluated the book; m is the average number of evaluations of the books in the collection; C is the average score of all books currently; WR represents the comprehensive score of the book.

[0037] The purpose of formula 3 is that the more books that are reviewed, the higher they will be ranked. For example, if 10 people have read book A and all of them have rated it 10 points, and 10,000 people have read book B and the average rating is 9.8 points, then the comprehensive rating of book B is higher than that of book A.

[0038] The scores recommended by internal subject teachers are mainly evaluations of the value and difficulty of the books in the collection by the subject teachers, which is more representative.

[0039] Final score = external score + internal reader score + subject teacher score.

[0040] The suitability of a book is calculated by the value score and difficulty score of the book, and is mainly used to evaluate whether the book is suitable for the reader. Here, WR1 represents the value score of the book, and WR2 represents the difficulty score of the book. The calculation method of these two values ​​can refer to Formula 3. rate is the ranking ratio of the subject score, for example, for the top 10%, rate = 0.1. The reader's subject mastery score is Skill, and the suitability score is set to AR.

[0041] Skill=(1 / (1+rate))*10 (4)

[0042] AR=WR1-|WR2-Skill|+New (5)

[0043] The higher the applicability score AR, the higher the recommendation of the book. New represents the new book coefficient. The new book coefficient is added mainly because new books are just put on the shelves and have no other data, so they are easily ignored. Therefore, a new book coefficient is added. This score is 0 or 5. The new book is 5. After a certain period of time, the value becomes 0.

[0044] 5) Based on reader information and book reviews, find books that are highly relevant to the topic and recommend them to readers.

[0045] Book recommendations include search recommendations and stage recommendations.

[0046] When readers use the book search system to find books on related subjects, the system will find books that are related to the subject and have high applicability based on the readers' personal information and book reviews, and recommend them to the readers.

[0047] Phased recommendation means that at a certain stage, a list of books will be actively pushed to readers via SMS or WeChat mini-program messages. For example, after a reader's subject scores are released during an exam, the system will count the student's mastery of the subject, calculate the applicability of the library's books to the reader, and select books with high applicability to push to the reader. If the reader's subject scores are poor, it may be that his foundation is not solid enough, so books with relatively low subject difficulty will be pushed to the reader as introductory books to lay a good foundation; if the subject scores are particularly good, advanced books will be recommended to help the reader further improve.

[0048] like Figure 2 As shown, this embodiment also provides a school library intelligent book recommendation system, including

[0049] Reader information management and analysis module: The data comes from the school's student management and academic management information. The reader's personal information, major information, course information, performance information, etc. are collected and analyzed to determine the reader's interests and hobbies, the type of knowledge that needs to be supplemented, etc.

[0050] Book information management module: Book data is mainly based on book acquisition and editing information and reader evaluation information.

[0051] Book evaluation module: This module calculates the value score and difficulty score for the book based on the collected book evaluation information.

[0052] Book recommendation module: This module combines the different characteristic information of each reader to recommend books that may be of interest to them, and generates suitable study book lists for them based on the student readers' staged course mastery, so as to achieve accurate and effective recommendations for student readers.

[0053] like Figure 3 As shown, the system is a software that sets up the client and server architecture;

[0054] The client is the reader's access portal and the user access layer, mainly web pages and WeChat applets;

[0055] The server includes a database, a data layer and a business layer; the database is used to store readers' personal information, readers' subject mastery information, readers' course information, basic information of books and book evaluation information; the data layer is used to process the storage and read-write logic of the database; the business layer includes book information management, reader information analysis, book evaluation, book applicability ranking and book recommendation.

[0056] Although the embodiments disclosed in the present invention are as above, the above contents are only embodiments adopted for facilitating the understanding of the present invention and are not intended to limit the present invention. Any technician in the technical field to which the present invention belongs can make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in the present invention, but the patent protection scope of the present invention shall still be subject to the scope defined in the attached claims.

Claims

1. A smart book recommendation method for a school library, characterized in that: The following steps are involved: Step A collects and analyzes the information of readers in the database, reads and writes data, and determines the readers' hobbies and the types of knowledge that need to be supplemented; Step B classifies the books by subject and subject difficulty according to the book information; Step C analyzes the reader information and analyzes the books the reader needs; Step D scores the value and difficulty of the books through the book evaluation information, and sorts the applicability of the books according to the scores; Step E: Find books with high relevance to the topic based on reader information and book reviews, and recommend them to readers; The book value and the book difficulty evaluation scores are derived from external ratings and internal ratings; The internal rating is calculated by using different readers' different rating weights for the book, and the calculation formula is: R = (1 / (1+rate))*score (1) AvgR = avg(R) (2) WR = (v / (v+m))× AvgR + (m / (v+m)) × C (3) Among them, score is the reader's rating, rate is the ranking ratio of subject scores, R represents the score calculated by weighting the subject score ratio; AvgR represents the average score calculated by weighting the subject score ratio; v is the number of reviewers of the book; m is the average number of reviews of the books in the collection; C is the average score of all books at present; WR represents the comprehensive score of the book; The applicability ranking of books is calculated by the value score and difficulty score of the books. The reader's subject mastery score Skill is calculated based on the ranking ratio of the subject scores. The applicability score is AR. Skill = (1 / (1+rate) )* 10 (4) AR = WR1-|WR2-Skill|+New (5) The higher the applicability score AR, the higher the recommendation degree of the book; WR1 is the book value score, WR2 is the book difficulty score, and New represents the new book coefficient.

2. The intelligent book recommendation method for school libraries according to claim 1, characterized in that: The reader information includes the reader's personal information, subject mastery information and course information; the book information includes the basic information of the book and the evaluation information of the book.

3. The intelligent book recommendation method for school libraries according to claim 1, characterized in that: In the step B, the books are classified by subject and subject difficulty according to the abstract, theme, basic information and book evaluation information of the books.

4. The intelligent book recommendation method for school libraries according to claim 1, characterized in that: In step C, the books needed by the reader are analyzed according to the reader's major, class schedule, subject grades and elective courses.

5. The intelligent book recommendation method for school libraries according to claim 1, characterized in that: The external ratings are obtained by crawling readers' ratings from external websites through web crawlers; the internal ratings are evaluations of internal readers' borrowing.

6. The intelligent book recommendation method for school libraries according to claim 1, characterized in that: Book recommendations include search recommendations and stage recommendations.

7. An intelligent book recommendation system for a school library, characterized in that: include The student information management and analysis module is used to collect and analyze reader information to determine readers' interests and hobbies and the types of knowledge they need to supplement; Book information management module: book data is based on book acquisition and editing information and reader evaluation information; The book evaluation module calculates the value score and difficulty score for the book based on the collected book evaluation information; The book recommendation module is used to combine the different characteristic information of each reader to recommend books that may be of interest to them, and to generate a suitable study book list based on the student reader's staged course mastery, so as to achieve accurate and effective recommendations for student readers; The evaluation scores of book value and book difficulty are derived from external and internal ratings; The internal rating is calculated by using different readers' different rating weights for the book, and the calculation formula is: R = (1 / (1+rate))*score (1) AvgR = avg(R) (2) WR = (v / (v+m))× AvgR + (m / (v+m)) × C (3) Among them, score is the reader's rating, rate is the ranking ratio of subject scores, R represents the score calculated by weighting the subject score ratio; AvgR represents the average score calculated by weighting the subject score ratio; v is the number of reviewers of the book; m is the average number of reviews of the books in the collection; C is the average score of all books at present; WR represents the comprehensive score of the book; The ranking of book applicability is calculated by the value score and difficulty score of the book. According to the ranking ratio of subject scores, the reader's subject mastery score Skill is calculated. The applicability score is AR. Skill = (1 / (1+rate) )* 10 (4) AR = WR1-|WR2-Skill|+New (5) The higher the applicability score AR, the higher the recommendation degree of the book; WR1 is the book value score, WR2 is the book difficulty score, and New represents the new book coefficient.

8. The intelligent book recommendation system for school libraries according to claim 7, characterized in that: Set up the client and server in the system architecture; The client is the reader's access portal and the user access layer; The server side includes a database, a data layer and a business layer; The database is used to store readers' personal information, readers' subject mastery information, readers' course information, basic information of books and book evaluation information; the data layer is used to process the storage and read and write logic of the database; the business layer includes book information management, reader information analysis, book evaluation, book suitability ranking and book recommendation.

Images