A book data processing method and a readable storage medium

Abstract

This invention discloses a book data processing method and a readable storage medium. The method includes: acquiring a user's search term; retrieving a corresponding book based on the user's search term; establishing a set of books for a search category based on the search category of the current search book and the books in that search category; sorting the books in the search category based on the information of all the books in the search category to obtain a sequence set of books in the search category; acquiring the user's borrowing records; obtaining recommended books for the user based on the number of books borrowed in the current borrowing records; displaying the search book and the recommended books to the user; and improving the user experience by comprehensively analyzing the user's borrowing records and the search book selections to obtain recommended books that better match the user's preferences.

Description

Technical Field

[0001] This invention relates to the field of book data management, and more particularly to a book data processing method and a readable storage medium. Background Technology

[0002] With the advent of the digital age, paper books are gradually being replaced by digital books, making the processing of digital book data increasingly important.

[0003] Nowadays, online reading is a popular choice for many people. After a user searches for books, a selection of books is recommended. However, these recommendations are often the latest releases or books recommended by authors at a cost, failing to accurately reflect the user's reading preferences. Therefore, how to provide precise and personalized book recommendations to enhance the user experience is a pressing issue that needs to be addressed. Summary of the Invention

[0004] The purpose of this invention is to provide a book data processing method and a readable storage medium, which solves the above-mentioned technical problems pointed out in the prior art.

[0005] This invention provides a method for processing book data, comprising the following steps:

[0006] Obtain the user's search terms, and retrieve a corresponding book based on the current user search terms;

[0007] Based on the search category corresponding to the currently searched book and the books in the search category corresponding to the search category, establish a set of books in the search category corresponding to the current search category. ; where q o This refers to the 0th book in the current search category set.

[0008] Based on the search category book information corresponding to all the search category books, sort the search category books to obtain a search category book sequence set;

[0009] Retrieve the current user's borrowing history; based on the number of books borrowed in the current borrowing history, retrieve recommended books for the current user;

[0010] The searched books and recommended books are displayed to the current user.

[0011] Preferably, obtaining the user's recommended books based on the number of books borrowed in the current borrowing record includes the following steps:

[0012] The preset number of recommended books is N; determine whether the number of borrowed books is equal to 0; if so, determine the first N search category books in the search category book sequence set as recommended books;

[0013] If not, then obtain the borrowing record information;

[0014] The borrowing record information includes the total number of books borrowed z, the borrowing category j corresponding to each borrowed book, and the number of books borrowed S corresponding to each borrowing category. j ;

[0015] Based on the borrowing records and the search category book sequence set, the recommended books for the current user are calculated;

[0016] Preferably, the step of calculating and obtaining the recommended books for the current user based on the borrowing record information and the search category book sequence set includes the following steps:

[0017] Based on the total number of books borrowed by the user, z, and the number of books borrowed for each borrowing category, S. j Calculate and obtain the user's current preference level H for the types of books to be borrowed. j ;

[0018] Based on the information of the borrowing category books, sort all the borrowing category books in each borrowing category to obtain the borrowing category book sequence set corresponding to each borrowing category;

[0019] Preset a weight value 'a' for recommended borrowed books and a weight value 'b' for recommended books in search categories; calculate the number of recommended borrowed books 'n1' and the number of recommended books in search categories 'n2' based on the number of recommended books 'N'.

[0020] Based on the number of books recommended for borrowing, n1, and the user's current preference for book categories, H... j Calculate the number m of recommended books for each borrowing category. j ;

[0021] After filtering out the borrowed books from the borrowing records from the book sequence set of the borrowing category, the recommended borrowing number m is determined according to the borrowing category. j Determine the first m items in the corresponding borrowing category book sequence set. j The books in each borrowing category are selected as recommended borrowing books; and after filtering the searched books from the search category book sequence set, the top n2 search category books in the search category book sequence set are determined as recommended search books.

[0022] The borrowing recommended books and the search recommended books are determined to be the recommended books for the current user;

[0023] The information about the borrowing category books includes multiple rating factors for the borrowing category books and multiple user borrowing reviews for the borrowing category books.

[0024] Preferably, the step of sorting all the books in each borrowing category according to the information of the borrowing category books to obtain the borrowing category book sequence set corresponding to each borrowing category includes the following operation steps:

[0025] Based on multiple rating factors for books in all the aforementioned borrowing categories, construct a set of rating factors for the current borrowing category. ;where x f This is the rating factor for the f-th borrowed book in the current borrowing category;

[0026] Construct a user borrowing review set based on multiple user borrowing reviews of books across all the aforementioned borrowing categories. ; where v d Provide a borrowing review for the d-th user;

[0027] Based on the set of factors for rating borrowed books and the set of user borrowing reviews, obtain a set of single-factor reviews. ; where r ig The g-th user borrowing review is the rating factor for the i-th borrowed book.

[0028] Based on the aforementioned single-factor evaluation set, a comprehensive scoring matrix is ​​constructed. ; where r mn The nth user's borrowing review is the rating factor for the mth borrowed book.

[0029] A weight is preset for each of the borrowed book rating factors, resulting in a set of weights for the borrowed book rating factors. ; where a y Let y be the weight of the rating factor for the y-th borrowed book;

[0030] Based on the comprehensive scoring matrix and the set of weights for the scoring factors of the borrowed books, the score B for each borrowed book in the current borrowing category is calculated and obtained;

[0031] The borrowed books are sorted from highest to lowest according to their ratings to obtain a sequence set of books by borrowing category.

[0032] Preferably, the step of constructing a user borrowing review set based on multiple user borrowing reviews of books across all borrowing categories includes the following steps:

[0033] Obtain all original borrowing reviews corresponding to the books in the borrowing category; preprocess the original borrowing reviews to obtain the data to be extracted;

[0034] Feature extraction is performed on the data to be extracted to obtain multiple feature words; based on all the feature words, a feature word set C={c1, c2, c3, ...ck} is constructed; where ck is the k-th feature word;

[0035] Multiple evaluation terms are preset; based on all the evaluation terms and their corresponding associated evaluation terms, an evaluation term set p = {pt1, pt2, pt3, ..., ptw} is constructed; where ptw is the w-th associated evaluation term of the t-th evaluation term;

[0036] The feature words are combined with all the associated evaluation words to obtain evaluation word pairs; an evaluation word pair matrix is ​​constructed based on all the evaluation word pairs. ; where ckptw is the evaluation word pair obtained by combining the k-th feature word and the w-th associated evaluation word of the t-th evaluation word;

[0037] Traverse each evaluation word pair in the evaluation word pair matrix; calculate the similarity q of each evaluation word pair in the evaluation word pair matrix; filter and obtain multiple user borrowing evaluations according to the preset evaluation word pair similarity threshold Q;

[0038] Based on all the user borrowing reviews mentioned, construct a user borrowing review set.

[0039] The present invention provides a readable storage medium comprising a stored program, wherein the program, when running, controls the device where the readable storage medium is located to execute the above-described book data processing method.

[0040] Compared with the prior art, the embodiments of the present invention have at least the following technical advantages:

[0041] Analysis of the book data processing method and readable storage medium provided by this invention reveals that, in practical applications, a search book is first determined based on the user's search terms. Then, multiple search category books are obtained according to the category of the search book (the search category books here are consistent with the category to which the current user's searched book belongs). Further, the search category books are scored and sorted using the information of each search category book to obtain a search category book sequence set (when recommending books, the top-ranked search category books in the search category book sequence set are selected, because the higher the ranking of a search category book in the sequence set, the higher its overall score, and the more likely the recommended book will be accepted by the user). Further analysis of the user's borrowing records determines whether the number of borrowing records is greater than 0. If not, it proves that the user is a new user, and recommended books are directly selected from the search category book sequence set (since the current new user is searching for this type of book, it can be assumed that the current user likes this type of book, and multiple highly-rated books in this type of book are recommended to the user).

[0042] If the analysis shows that the user's borrowing history is not empty, then the current user is considered a returning user. In this case, recommendations need to be made based on both the user's searched books and their borrowing history. Specifically, the first step in making recommendations is to analyze the borrowed books in the user's history. Based on the category of each borrowed book and the number of books borrowed, the user's preference for each borrowing category is calculated. Books are then selected for recommendation from each borrowing category based on this preference. Furthermore, the books in each borrowing category are comprehensively rated based on information about the books in the user's current borrowing category. The rating may be based on multiple factors. The system incorporates user borrowing reviews and uses fuzzy comprehensive evaluation to calculate ratings for books within each borrowing category. These ratings are then ranked to create a sequence set of books for each category. Based on the user's current preference for each category, recommended books are selected from this sequence set. This ensures that the recommended books better match the user's preferences, resulting in more accurate recommendations and a better user experience. Finally, the system sends the user's search results along with the recommended books, providing a personalized and accurate recommendation process that enhances the user experience. Attached Figure Description

[0043] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0044] Figure 1 This is a schematic diagram illustrating the operation steps of a book data processing method provided in Embodiment 1 of the present invention;

[0045] Figure 2 This is a schematic diagram illustrating the steps for obtaining recommended books in a book data processing method according to Embodiment 1 of the present invention;

[0046] Figure 3 This is a schematic diagram illustrating the steps of obtaining recommended books through comprehensive analysis in a book data processing method provided in Embodiment 1 of the present invention;

[0047] Figure 4 This is a schematic diagram illustrating the operation steps of obtaining a book borrowing category sequence set in a book data processing method provided in Embodiment 1 of the present invention;

[0048] Figure 5 This is a schematic diagram illustrating the steps of obtaining a user borrowing evaluation set in a book data processing method provided in Embodiment 1 of the present invention;

[0049] Figure 6 This is a schematic diagram of the overall operation flow of a book data processing method provided in Embodiment 1 of the present invention. Detailed Implementation

[0050] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings.

[0052] Example 1

[0053] like Figure 1 or Figure 6 As shown, Figure 1 This is a schematic diagram illustrating the operation steps of a book data processing method provided in Embodiment 1 of the present invention; Figure 6 This is a schematic diagram of the overall operation flow of a book data processing method provided in Embodiment 1 of the present invention;

[0054] This invention proposes a method for processing book data, comprising the following steps:

[0055] Step S10: Obtain the user's search terms, and retrieve a corresponding book based on the current user search terms;

[0056] It should be noted that the above embodiment of the present invention obtains the books that the user wants to borrow by entering the search terms for the books to be searched in the text box.

[0057] In this embodiment of the invention, the search books obtained from the user's search term may be one or more books. In this case, the user needs to manually select the corresponding search book that is ultimately needed to be searched.

[0058] Step S20: After obtaining the search category corresponding to the currently searched book and the books in the search category corresponding to the search category, establish a set of books in the search category corresponding to the current search category. , where q o This refers to the 0th book in the current search category set.

[0059] It should be noted that the set of books by search category in the above embodiments of the present invention refers to the set of books consisting of all search books corresponding to the current search category to which the currently searched book belongs.

[0060] Step S30: Obtain the search category book information corresponding to each of the search category books; construct a search category information set based on the search category book information corresponding to all the search category books. , where u n For the nth search category of book information;

[0061] For example, the information for each book in a search category might include rating factors (such as the number of times the book has been borrowed, the publisher, the author, etc.) and user reviews (such as excellent, good, good, average, poor, etc.). In subsequent steps, the information for all books in the current search category is used to calculate the ratings for all books in the current search category and sort them to obtain a set of search category books.

[0062] Step S40: Based on all the book information of the search category, calculate the score of each book in the search category using the fuzzy comprehensive evaluation algorithm, and then sort all the books in the search category from high to low according to the scores of the books in the search category to obtain the search category book sequence set;

[0063] It should be noted that in the technical solution adopted in the above embodiments of the present invention, the method for calculating the rating of each search category book is to obtain the search category book user review set and the search category book rating factor set from the search category book information corresponding to each search category book; use the above search category book user review set and search category book rating factor set to obtain the search category book single factor review set (i.e., the rating result of each user search review for each search category book rating factor); then construct the search category book rating matrix; preset the search category book rating factor weight value for each search category book rating factor to obtain the search category book rating factor weight set; further, calculate the rating of each search category book, and sort all search category books from high to low according to the search category book rating to obtain the search category book sequence set.

[0064] Step S50: Obtain the current user's borrowing records; and obtain the total number of books borrowed corresponding to the borrowing records; obtain the current user's recommended books based on the number of borrowed books;

[0065] Step S60: Display the searched books and recommended books to the current user.

[0066] Specifically, such as Figure 2 As shown, in step S50, the recommended books for the current user are obtained based on the number of borrowed books, including the following steps:

[0067] Step S51: The preset number of recommended books is N; determine whether the number of borrowed books is equal to 0; if so (if the number of borrowed books is equal to 0, it means that the current user's borrowing record is empty, indicating that the current user is a new user, then use the search category book set analysis output results to recommend books to the current user), then determine the first N search category books in the search category book sequence set as recommended books;

[0068] Step S52: If not (if the number of borrowed books is greater than 0, that is, the current user's borrowing record is not empty, indicating that the current user is an old user, then the search category book set and the borrowed books are used for comprehensive analysis, and the comprehensive analysis results are used to recommend books to the user), then obtain the information of the borrowing record;

[0069] The borrowing record information includes the total number of books borrowed z, the borrowing category j corresponding to each borrowed book, and the number of books borrowed S corresponding to each borrowing category. j ;

[0070] Step S53: Calculate and obtain the recommended books for the current user based on the borrowing record information and the search category book sequence set;

[0071] Specifically, such as Figure 3 As shown, in step S53, the recommended books for the current user are calculated based on the information of the borrowing records and the book sequence set of the search category, including the following steps:

[0072] Step S531: Based on the total number of books borrowed by the user z and the number of books borrowed for each borrowing category S... j Calculate and obtain the user's current preference level H for the types of books to be borrowed. j ;

[0073] The user's preference for book categories H j The calculation method is as follows:

[0074] ;

[0075] In the formula, H j S represents the current user's preference for book category j; j z represents the number of books borrowed in borrowing category j; z represents the total number of books borrowed by the current user.

[0076] Step S532: Obtain the borrowing category books corresponding to each borrowing category and the information of the borrowing category books corresponding to the borrowing category books; based on the information of the borrowing category books, calculate the score of each borrowing category book through the fuzzy comprehensive evaluation algorithm, and sort all the borrowing category books in each borrowing category from high to low according to the score of the borrowing category books to obtain the borrowing category book sequence set corresponding to each borrowing category;

[0077] The information about the borrowing category books includes multiple rating factors for the borrowing category books and multiple user borrowing reviews of the borrowing category books;

[0078] Step S533: Preset the recommended weight value a for borrowed books and the recommended weight value b for search category books; calculate the recommended number n1 for borrowed books and the recommended number n2 for search category books based on the number N of recommended books; the recommended number N of books is the sum of the recommended number n1 for borrowed books and the recommended number n2 for search category books (i.e., N = n1 + n2).

[0079] The calculation method for the number of recommended books for borrowing, n1, is as follows:

[0080] n1 = N × a; where n1 is the number of books recommended for borrowing; N is the number of recommended books; and a is the weight value of the book recommendation.

[0081] The calculation method for the number n2 of recommended books in the search category is as follows:

[0082] n2 = N × b; where n2 is the number of books recommended in the search category; N is the number of recommended books; and b is the weight value of the book recommendation in the search category.

[0083] Step S534: Based on the number of books recommended for borrowing n1 and the user's current preference for book categories H j Calculate the number m of recommended books for each borrowing category. j ;

[0084] The number of recommended books for each borrowing category is m. j The calculation method is as follows:

[0085] ;

[0086] In the formula, m j The number of books recommended in book category j, m; H j n1 represents the current user's preference for book category j; n1 represents the number of books recommended for borrowing.

[0087] Step S535: After filtering out the borrowed books from the borrowing records from the book sequence set of the borrowing category, recommend the number of books to be borrowed according to the borrowing category m. j Determine the first m items in the sequence set corresponding to the borrowing category (of the corresponding borrowing category). j The books in each borrowing category are selected as recommended borrowing books; and after filtering the searched books from the search category book sequence set, the top n2 search category books in the search category book sequence set are determined as recommended search books.

[0088] Step S536: Determine that the borrowed recommended books and the search recommended books are the recommended books for the current user;

[0089] For example: the preset number of recommended books, N, is 10;

[0090] There are currently three books available for borrowing, denoted as ①, ②, and ③. Each book belongs to a different borrowing category: book ① is category I, book ② is category II, and book ③ is category III. The total number of books borrowed, N, is 3, and the number of books borrowed for each category, S, is... j If the value is 1, then the user's preference level H for borrowing category I can be calculated. j for The current user's preference level for borrowing category II (H) j for The current user's preference level for borrowing category III (H) j for ;

[0091] Obtain the books corresponding to each borrowing category, i.e., the borrowing category books in category I are A, B, C, D, E, F, G, H; the borrowing category books in category II are a, b, c, d, e, f, g, h; and the borrowing category books in category III are a1, b1, c1, d1, e1, f1, g1, h1. Obtain the information of the borrowing category books corresponding to all borrowing category books. After scoring each borrowing category book according to the information of the borrowing category books in each borrowing category, sort them to obtain the sequence set of borrowing category books corresponding to each borrowing category, represented as: I={A, B, C, D, E, F, G, H}; II={a, b, c, d, e, f, g, h}; III={a1, b1, c1, d1, e1, f1, g1, h1}.

[0092] The preset weight value a for recommended borrowed books is 0.6 and the weight value b for recommended search category books is 0.4; based on the number of recommended books N=10, the number of recommended borrowed books n1 is calculated to be 6 and the number of recommended search category books n2 is 4.

[0093] Based on the number of books recommended for borrowing, n1=6, and the user's current preference for book categories, H... j for Calculate the number m of recommended books for each borrowing category. j ;

[0094] The recommended number of books to borrow in borrowing category I is 6× =2; The recommended number of books to borrow in Borrowing Category II is 6× =2; The recommended number of books to borrow in Borrowing Category III is 6× =2;

[0095] Then, after filtering out ① from the borrowing record from borrowing category I, the first two books of borrowing category I in the book sequence set of borrowing category I are selected as recommended books for borrowing category I, that is, A and B are selected as recommended books for borrowing; after filtering out ② from the borrowing record from borrowing category II, the first two books of borrowing category II in the book sequence set of borrowing category II are selected as recommended books for borrowing category II, that is, a and b are selected as recommended books for borrowing; after filtering out ③ from the borrowing record from borrowing category III, the first two books of borrowing category III in the book sequence set of borrowing category III are selected as recommended books for borrowing category III, that is, a1 and b1 are selected as recommended books for borrowing.

[0096] Furthermore, after filtering out the currently searched books from the search book sequence set, four books from the search category book sequence set are selected as recommended search books.

[0097] Ultimately, the recommended books were determined from those borrowed and those searched.

[0098] Analysis of the above technical solutions shows that the information recommendation scheme adopted in this embodiment of the invention has a fixed total number of recommended books N. After setting weights for borrowed books and search category books respectively, the total number of search recommended books and borrowed recommended books is also fixed. Furthermore, it is necessary to analyze and select borrowed recommended books, that is, to calculate the user's preference for borrowing categories based on the number of borrowed books and the corresponding categories of borrowed books, and then calculate how many borrowing category books should be recommended under each borrowing category based on the preference and the total number of borrowed recommended books.

[0099] In summary, after further filtering out known search and borrowed books, a corresponding number of books are selected from the very beginning of the corresponding sequence set as recommended books.

[0100] Specifically, such as Figure 4 As shown, in step S532, after calculating the score of each borrowing category book using the fuzzy comprehensive evaluation algorithm based on the information of the borrowing category books, all the borrowing category books in each borrowing category are sorted from high to low according to the scores of the borrowing category books to obtain the borrowing category book sequence set corresponding to each borrowing category, including the following operation steps:

[0101] Step S5321: Based on the multiple rating factors of all the books in the borrowing categories, construct a set of rating factors for the books in the current borrowing category. ;where x f This is the rating factor for the f-th borrowed book in the current borrowing category;

[0102] It should be noted that, in this embodiment of the application, obtaining the set of rating factors for borrowed books is a prerequisite for sorting the borrowed books to obtain a sequence set of books by borrowing category. The rating value of each book is obtained by comprehensively scoring each book using various rating factors, thereby allowing the books to be sorted to obtain a sequence set of books by borrowing category.

[0103] Step S5322: Construct a user borrowing review set based on multiple user borrowing reviews of books in all the borrowing categories. ; where v d Provide a borrowing review for the d-th user;

[0104] It should be noted that, in this embodiment of the application, user borrowing evaluation refers to the evaluation that users give to the borrowed books. Thus, for each user's evaluation, a user borrowing evaluation set is formed for each borrowed book. User borrowing evaluation can serve as an indispensable factor in obtaining user ratings, making the book ranking results more objective, thereby making the obtained book sequence set of borrowing categories accurate, and ultimately making the final recommended books more likely to match the current user's preferences.

[0105] Step S5323: Obtain a single-factor evaluation set based on the set of borrowed book rating factors and the set of user borrowing evaluations. ; where r ig The g-th user borrowing review is the rating factor for the i-th borrowed book.

[0106] It should be noted that the single-factor evaluation set in this application embodiment is the rating result of each user's borrowing evaluation for each borrowed book rating factor.

[0107] Step S5324: Based on the single-factor evaluation set, construct a comprehensive scoring matrix. ;

[0108] Where, r mn The nth user's borrowing review is the rating factor for the mth borrowed book.

[0109] Step S5325: Preset weights for each of the borrowed book rating factors to obtain a set of weights for the borrowed book rating factors. ; where a y Let y be the weight of the rating factor for the y-th borrowed book;

[0110] It should be noted that, in this embodiment of the application, the method for setting the weight of the borrowing book rating factors can be the Delphi method, the weighted average method, the crowd evaluation method, etc., and this embodiment of the application does not limit it.

[0111] Step S5326: Calculate and obtain the score B of each borrowed book in the current borrowing category based on the comprehensive scoring matrix and the set of weights of the borrowed book scoring factors;

[0112] The calculation method for the rating B of the borrowed books is as follows:

[0113] ;

[0114] Step S5327: Sort the borrowed books from high to low according to their ratings to obtain a book sequence set by borrowing category.

[0115] It should be noted that, in this embodiment of the application, by obtaining information on the books in each borrowing category, the fuzzy comprehensive evaluation method is used to calculate the score of each borrowing book, and then the ranking of the scores is used to obtain the ranking table of the borrowing books in that category (i.e., the sequence set of borrowing category books). By using multi-factor and multi-user objective evaluation, the score of each borrowing book is calculated to ensure the humanization of the final output of recommended books.

[0116] During the implementation of this application, the technical personnel discovered that user borrowing reviews are usually long texts. It is very difficult to calculate the rating of each borrowed book using such long texts, and the vocabulary in the long texts is different from the pre-set vocabulary for user borrowing reviews. Therefore, it is difficult to ensure the accuracy of the calculated rating of each borrowed book, and consequently, it is impossible to ensure the accuracy of the final obtained set of borrowing category book sequences.

[0117] Specifically, such as Figure 5 As shown, in step S5322, a user borrowing evaluation set is constructed based on multiple user borrowing evaluations of books in all the borrowing categories, including the following steps:

[0118] Step S53221: Obtain all original borrowing reviews corresponding to the books in the borrowing category; preprocess the original borrowing reviews to obtain the data to be extracted;

[0119] It should be noted that the original borrowing evaluation in the above embodiments of the present invention is for each book in the borrowing category. After all users have borrowed the books, borrowers usually evaluate the books in that borrowing category, thus obtaining the original borrowing evaluation. The above-mentioned preprocessing methods for the original borrowing evaluation include, but are not limited to, noise reduction (removing punctuation marks, stop words, and HTML tags) processing of the original borrowing evaluation. By removing these irrelevant information, the data dimensionality can be reduced and the training speed and performance of the algorithm can be improved, thereby obtaining the data to be extracted (the data to be extracted refers to the text data to be extracted for features).

[0120] Step S53222: Perform feature extraction on the data to be extracted to obtain multiple feature words; based on all the feature words, construct a feature word set C={c1, c2, c3, ...ck}; where ck is the kth feature word;

[0121] It should be noted that the feature words extracted in the above-described embodiments of this application can be extracted using methods such as equal weighted statistics, TF-IDF, and word2vec, and this application does not impose any limitations on them. The feature words are used to compare with the associated evaluation words corresponding to the evaluation words set below, so as to determine whether the extracted feature words can be used to calculate the book rating, thereby ensuring the accuracy of the book rating calculation.

[0122] Step S53223: Preset multiple evaluation terms; based on the evaluation terms, use web crawling technology to crawl multiple related evaluation terms respectively; based on all the evaluation terms and the related evaluation terms corresponding to the evaluation terms, construct an evaluation term set p={pt1, pt2, pt3, ..., ptw}; where ptw is the w-th related evaluation term of the t-th evaluation term;

[0123] It should be noted that in the technical solutions adopted in the above embodiments of this application, the preset evaluation terms may include evaluation terms such as "very good" and "average". The related evaluation terms for "very good" obtained by web crawling technology include "excellent", "outstanding", "remarkable", and "outstanding". The related terms for "average" obtained by web crawling technology include "plain", "unremarkable", and "common". The preset evaluation terms and their corresponding related terms together constitute an evaluation term set, which makes it easier to identify the evaluation terms to which the user's borrowing evaluation belongs, thereby better performing the book rating operation.

[0124] Step S53224: Combine the feature words with all the associated evaluation words to obtain evaluation word pairs; construct an evaluation word pair matrix based on all the evaluation word pairs. ; where ckptw is the evaluation word pair obtained by combining the k-th feature word and the w-th associated evaluation word of the t-th evaluation word;

[0125] Step S53225: Traverse each evaluation word pair in the evaluation word pair matrix; calculate and obtain the similarity q of each evaluation word pair in the evaluation word pair matrix; preset the evaluation word pair similarity threshold Q; determine whether the similarity q of the current evaluation word pair is greater than or equal to the evaluation word pair similarity threshold Q; if so, determine that the evaluation word corresponding to the associated evaluation word in the current evaluation word pair is the user borrowing evaluation (repeat the above traversal operation to obtain multiple user borrowing evaluations);

[0126] Step S53226: Construct a user borrowing evaluation set based on all the user borrowing evaluations.

[0127] It should be noted that the technical solution adopted in the above-described embodiments of this application preprocesses the original user borrowing reviews to obtain the data to be extracted, then performs feature extraction to obtain feature words, reducing the difficulty of calculating book ratings; at the same time, it presets evaluation words and obtains the associated evaluation words corresponding to each evaluation word; further, it combines the feature words and associated words to obtain an evaluation word pair matrix; then, it calculates the similarity of each element (evaluation word pair) in the evaluation word pair matrix and presets a similarity threshold; the user borrowing reviews are obtained by judging through the similarity threshold, thereby improving the accuracy of the final calculated book ratings for borrowing categories.

[0128] In summary, the book data processing method and readable storage medium proposed in this invention first determine a search book based on the user's search terms. Then, based on the category of the search book, multiple corresponding search category books are obtained (the search category books here are consistent with the category to which the current user's searched book belongs). Further, the search category books are scored and sorted using the information of each search category book to obtain a search category book sequence set (when recommending books, the top-ranked search category books in the search category book sequence set are selected, because the higher the ranking of a search category book in the sequence set, the higher its overall score, and the more likely the recommended book will be accepted by the user). Further, the user's borrowing records are analyzed to determine if the number of borrowing records is greater than 0. If not, it proves that the user is a new user, and recommended books are directly selected from the search category book sequence set (because the current new user is searching for this type of book, it can be assumed that the current user likes this type of book, so multiple highly-rated books in this type of book are recommended to the user).

[0129] If the analysis shows that the user's borrowing history is not empty, then the current user is considered a returning user. In this case, recommendations need to be made based on both the user's searched books and their borrowing history. Specifically, the first step in making recommendations is to analyze the borrowed books in the user's history. Based on the category of each borrowed book and the number of books borrowed, the user's preference for each category is calculated. Books are then selected for recommendation based on this preference. Furthermore, a comprehensive score is given to the books in each borrowing category based on the information of the books in the user's current borrowing category. The scoring can be based on multiple factors and user borrowing reviews. A fuzzy comprehensive evaluation method is used to calculate the scores for books in each borrowing category. Then, the books in each borrowing category are sorted to obtain a sequence set of books for each category. Recommended books are then selected from this sequence set based on the user's current preference for each borrowing category. This ensures that the recommended books better match the user's preferences, resulting in more accurate recommendations and a better user experience. Finally, the user's search results and recommended books are sent together to the user.

[0130] Furthermore, when processing user reviews of books selected for borrowing, feature evaluation words are extracted from the original user reviews. Then, related evaluation words are obtained using preset evaluation words. These feature evaluation words and related evaluation words are then combined one by one to obtain evaluation word pairs. Evaluation words are then selected based on the similarity of each evaluation word pair, ensuring accurate sorting in the resulting set of borrowed books. This guarantees that the final recommended books truly match user preferences, resulting in more personalized book recommendations and improved user experience.

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; those skilled in the art can modify the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A book data processing method characterized by comprising: The following steps are included: Obtain the user's search terms, and retrieve a corresponding book based on the current user search terms; Based on the search category corresponding to the currently searched book and the books in the search category corresponding to the search category, establish a set of books in the search category corresponding to the current search category. ; where q o This refers to the book in the 0th search category within the current collection of books in that search category. Based on the search category book information corresponding to all the search category books, sort the search category books to obtain a search category book sequence set; Retrieve the current user's borrowing history; based on the number of books borrowed in the current borrowing history, retrieve recommended books for the current user; Display the searched books and recommended books to the current user; The step of obtaining the user's recommended books based on the number of books borrowed in the current borrowing record includes the following steps: The preset number of recommended books is N; determine whether the number of borrowed books is equal to 0; if so, determine the first N search category books in the search category book sequence set as recommended books; If not, then obtain the borrowing record information; Based on the borrowing records and the search category book sequence set, the recommended books for the current user are calculated; The step of calculating and obtaining the recommended books for the current user based on the borrowing record information and the search category book sequence set includes the following steps: Based on the total number of books borrowed by the user, z, and the number of books borrowed for each borrowing category, S. j Calculate and obtain the user's current preference level H for the types of books to be borrowed. j ; Based on the information of the borrowing category books, sort all the borrowing category books in each borrowing category to obtain the borrowing category book sequence set corresponding to each borrowing category; Preset a weight value 'a' for recommended borrowed books and a weight value 'b' for recommended books in search categories; calculate the number of recommended borrowed books 'n1' and the number of recommended books in search categories 'n2' based on the number of recommended books 'N'. Based on the number of books recommended for borrowing, n1, and the user's current preference for book categories, H... j Calculate the number m of recommended books for each borrowing category. j ; After filtering out the borrowed books from the borrowing records from the book sequence set of the borrowing category, the recommended borrowing number m is determined according to the borrowing category. j Determine the first m items in the corresponding borrowing category book sequence set. j The books in each borrowing category are selected as recommended borrowing books; and after filtering the searched books from the search category book sequence set, the top n2 search category books in the search category book sequence set are determined as recommended search books. The borrowing recommended books and the search recommended books are determined to be the recommended books for the current user.

2. The book data processing method according to claim 1, wherein The borrowing record information includes the total number of books borrowed z, the borrowing category j corresponding to each borrowed book, and the number of books borrowed S corresponding to each borrowing category. j .

3. The book data processing method according to claim 2, wherein The number N of recommended books is the sum of the number of borrowed books recommended (n1) and the number of books recommended in the search category (n2).

4. The book data processing method according to claim 3, wherein The number of recommended books N is a constant; the number of books recommended for borrowing n1 is a constant; the number of books recommended for the search category n2 is a constant; and the user's preference for book categories H... j For variables.

5. The method of claim 4, wherein, The information about the borrowing category books includes multiple rating factors for the borrowing category books and multiple user borrowing reviews for the borrowing category books.

6. The book data processing method according to claim 5, characterized in that, The step of sorting all books in each borrowing category according to the information of the borrowing category books to obtain the borrowing category book sequence set corresponding to each borrowing category includes the following operation steps: Based on multiple rating factors for books in all the aforementioned borrowing categories, construct a set of rating factors for the current borrowing category. ;where x f This is the rating factor for the f-th borrowed book in the current borrowing category; constructing a user borrowing evaluation set according to a plurality of user borrowing evaluations of all the borrowing category books ; wherein v d is the dth user borrowing evaluation Based on the set of factors for rating borrowed books and the set of user borrowing reviews, obtain a set of single-factor reviews. ; where r ig The g-th user borrowing review is the rating factor for the i-th borrowed book. Based on the aforementioned single-factor evaluation set, a comprehensive scoring matrix is ​​constructed. ; where r mn The nth user's borrowing review is the rating factor for the mth borrowed book. A weight is preset for each of the borrowed book rating factors, resulting in a set of weights for the borrowed book rating factors. ; where a y Let y be the weight of the rating factor for the y-th borrowed book; Based on the comprehensive scoring matrix and the set of weights for the scoring factors of the borrowed books, the score B for each borrowed book in the current borrowing category is calculated and obtained; The borrowed books are sorted from highest to lowest according to their ratings to obtain a sequence set of books by borrowing category.

7. The book data processing method according to claim 6, wherein The step of constructing a user borrowing review set based on multiple user borrowing reviews of books in all the borrowing categories includes the following steps: Obtain all original borrowing reviews corresponding to the books in the borrowing category; preprocess the original borrowing reviews to obtain the data to be extracted; Feature extraction is performed on the data to be extracted to obtain multiple feature words; based on all the feature words, a feature word set C={c1, c2, c3, ...ck} is constructed; where ck is the k-th feature word; Multiple evaluation terms are preset; based on all the evaluation terms and their corresponding associated evaluation terms, an evaluation term set p = {pt1, pt2, pt3, ..., ptw} is constructed; where ptw is the w-th associated evaluation term of the t-th evaluation term; The feature words are combined with all the associated evaluation words to obtain evaluation word pairs; an evaluation word pair matrix is ​​constructed based on all the evaluation word pairs. ; where ckptw is the evaluation word pair obtained by combining the k-th feature word and the w-th associated evaluation word of the t-th evaluation word; Traverse each evaluation word pair in the evaluation word pair matrix; calculate the similarity q of each evaluation word pair in the evaluation word pair matrix; filter and obtain multiple user borrowing reviews according to the preset evaluation word pair similarity threshold Q; Based on all the user borrowing reviews mentioned, construct a user borrowing review set.

8. A readable storage medium for book data processing, characterized in that, The readable storage medium includes a stored program, wherein, when the program is executed, it controls the device containing the readable storage medium to perform a book data processing method according to any one of claims 1 to 7.

Images