A smart library management system and method based on big data

Abstract

The present invention relates to the field of library management technology, specifically to a smart library management system and method based on big data, wherein the system includes an association substitution analysis module, wherein the association substitution analysis module obtains a library set consisting of the remaining libraries except the library to be tested in all libraries in the metropolitan area, obtains the book borrowing characteristics of library users who follow a public account and the characteristic association values ​​between each library in the library set; and combines the location of each library in the library set with the distance between the library to be tested, to generate a comprehensive association substitution value of the library to be tested for each element in the library set for the user. The present invention takes into account the situation that the user's borrowing and returning of books and the operation of retrieving books cannot be performed during the maintenance of the library management system, realizes the calibration of the library maintenance time and the screening of libraries in the metropolitan area, reduces the impact of library maintenance on users, and increases the user's experience of using the library.

Description

Technical Field

[0001] The present invention relates to the technical field of library management, and in particular to a smart library management system and method based on big data. Background Art

[0002] With the rapid development of the global information industry, the application of computer technology is becoming increasingly popular. In the field of library management, the use of computer technology has effectively improved the management level of books. The process involves the application of software and hardware. By using the library management system, people can quickly query the location and inventory status of the corresponding books by searching for information about the books to be queried on computer devices, which is convenient for users to quickly borrow their favorite books. At the same time, the library management system analyzes the user's borrowing and retrieval behavior, and can accurately understand the user's borrowing and retrieval habits, so as to combine the library's collection and recommend books that meet the user's borrowing and retrieval habits to the user, thereby exploring the user's reading potential and guiding the user to develop good reading habits.

[0003] However, both software and hardware will bring a certain degree of burden and damage to the computer as the usage time and frequency increase, and maintenance is required. The existing library management system usually sets the maintenance cycle manually and announces it before maintenance, but this method has major defects. On the one hand, during the maintenance of the library management system, users cannot perform operations such as borrowing and returning books and retrieving books, which affects users' use of the library. The existing library system often closes the library during maintenance. On the other hand, users often have dependence on the library in the process of using it, and often borrow and search for books in the same library. When the corresponding library cannot operate and the user does not understand the collection of other libraries, the user cannot quickly and effectively find an alternative library to the corresponding library (different users borrow books with different focuses and categories, and there are differences in the collection of different libraries), which affects the user's experience and work progress. Summary of the invention

[0004] The purpose of the present invention is to provide a smart library management system and method based on big data to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a smart library management method based on big data, the method comprising the following steps:

[0006] S1. Obtaining library information of all libraries in the metropolitan area in the most recent unit time, wherein the library information includes book information, book borrowing and returning circulation information, and library status information, and the library to be tested is a library in the corresponding metropolitan area;

[0007] S2. Obtain the operation time set of each user in the tested library in the most recent unit time, combine the preset maintenance interval of the tested library and the average maintenance duration of each maintenance in the historical data, obtain the next best maintenance time of the tested library, generate a maintenance announcement, and push it to the corresponding public account users through the bound public account;

[0008] S3, query the database for the association between the public account user and the library user to be tested, determine the user category of the public account user, the public account user category includes library users who follow the public account and non-library users who follow the public account; obtain the historical book borrowing information of the library users who follow the public account in the library to be tested, and extract the book borrowing characteristics of the library users who follow the public account;

[0009] S4, obtaining a library set consisting of the remaining libraries in all libraries in the metropolitan area except the library to be tested, obtaining the book borrowing characteristics of library users who follow the public account and the characteristic association values ​​between each library in the library set; combining the location of each library in the library set with the distance between the library to be tested and generating a comprehensive association substitution value of the library to be tested for each element in the library set for the user;

[0010] S5. Sort the libraries in the library set in descending order according to the comprehensive associated replacement values ​​of the library to be tested for the user, generate a replacement library sequence for the library users who follow the corresponding public account during the maintenance period of the library to be tested, bind it with the maintenance announcement and push it to the corresponding public account users at the same time. In the push information of different public account users, the replacement library sequences of different library users who follow the public account during the maintenance period of the library to be tested are different. Non-library users who follow the public account do not have a replacement library sequence during the maintenance period of the library to be tested.

[0011] Furthermore, the unit time in S1 is a constant preset in the database, the book information includes the names of the books in the library and the categories to which they belong; the library borrowing and returning circulation information includes the borrowing records and returning records of library users; and the library status information includes the normal operation status and the closed maintenance status.

[0012] Furthermore, when obtaining the operation time set of each user in the library to be tested within the most recent unit time in S2, each user corresponds to an operation time set, and each operation time set includes the borrowing operation time, returning operation time and searching operation time of the corresponding user in the library to be tested;

[0013] The method for obtaining the next optimal maintenance time of the library to be tested in S2 comprises the following steps:

[0014] S21, taking n days as a time period, obtaining an operation proportion information set corresponding to the operation time set of each day in the i1 time periods before the current time in the historical data, wherein the operation proportion information includes the borrowing operation proportion information, the returning operation proportion information and the retrieval operation proportion information, and recording the operation proportion information set corresponding to the operation time set of the jth day in the i-th time period before the current time in the historical data as Aij, wherein 1≤i≤i1, and 1≤j≤n;

[0015] S22. Obtain the impact value of the comprehensive operation proportion corresponding to the jth day in a time period, recorded as Qj,

[0016] Qj=[∑ i＝1 i＝i1 r1·Aij1+r2·Aij2+r3·Aij3] / i1,

[0017] Among them, r1, r2 and r3 are constants preset in the database, Aij1 represents the proportion of borrowing operations in Aij, Aij2 represents the proportion of returning operations in Aij, and Aij3 represents the proportion of searching operations in Aij;

[0018] S23, obtain the preset maintenance interval T of the library to be tested, the average maintenance time T1 of each maintenance in the historical data, and the end time tq of the previous maintenance of the library to be tested, obtain the next maintenance time prediction interval of the library to be tested, recorded as [tq+T, tq+T+T1], and obtain the interval corresponding to [tq+T, tq+T+T1] in the divided time period, recorded as

[0019] [B1, B2], where B1 is the sequence of days to which tq+T belongs in the corresponding time period (i.e., which day in the corresponding time period it belongs to), and B2 is the sequence of days to which tq+T+T1 belongs in the corresponding time period, and the units of tq, T and T1 are all days;

[0020] S24, using C as a time adjustment parameter, adjusting [B1, B2] to obtain an interval [B1+C, B2+C], where C is an integer of 0≤C≤n; when C is calculated to be different values, the corresponding ∑ j1＝B1+C j1＝B2+C In Qj1, ∑ j1＝B1+C j1＝B2+C The C value corresponding to the minimum value of Qj1 is recorded as C1, and the optimal maintenance time of the next library under test is obtained [tq+T+C1, tq+T+T1+C1];

[0021] When obtaining the operation ratio information corresponding to the three types of operations, namely, the borrowing operation ratio information, the returning operation ratio information, and the retrieval operation ratio information, construct the operation time chains corresponding to each type of operation on the same day. The chain node in each operation time chain corresponds to an operation time of the corresponding category. The two adjacent operation times corresponding to the same type of operations with an interval time less than the chain interval time are taken as two adjacent nodes on the same operation time chain, and the two adjacent operation times corresponding to the same type of operations with an interval time greater than or equal to the chain interval time are taken as nodes on different operation time chains.

[0022] The proportion of each type of operation is equal to the ratio of the sum of the lengths of each operation time chain corresponding to the corresponding type of operation on that day to the length of one day. The length of the operation time chain is equal to the difference between the maximum operation time and the minimum operation time in each chain node.

[0023] The interchain time length is t0-β·(AM-d),

[0024] Among them, t0 is the chain spacing duration corresponding to the previous time period, AM is the maximum value of the operation proportion information of all categories in the previous time period, d represents the adjustment threshold and d is a constant preset in the database, β is the chain spacing adjustment conversion coefficient and β is a constant preset in the database.

[0025] The present invention screens the time interval with the least impact on the operation by analyzing the accumulated values ​​of the comprehensive operation proportion influence values ​​corresponding to each time period when different time adjustment parameters are used, thereby realizing the calibration of the maintenance time prediction interval and obtaining the optimal maintenance time; the chain spacing duration is updated when obtaining the operation time chain, considering that the number of library users will gradually change with the change of time. If a fixed chain spacing is simply used, the analysis result will have a certain degree of deviation (if the chain spacing duration is too large, it will make it difficult to effectively distinguish the differences between the operation time chains; and if the chain spacing duration is too small, the operation time chains in the same type of operation will be relatively scattered, and the proportion value obtained will be too small, resulting in poor accuracy in subsequent comparisons). Therefore, the present invention can achieve accuracy control of the obtained time adjustment parameter C1 to a certain extent through adaptive adjustment of the chain spacing duration.

[0026] Further, in S3, when the public account in the database is a library user to be tested, the user category of the public account user is determined to be a library user who follows the public account; when the public account in the database is not a library user to be tested, the user category of the public account user is determined to be a non-library user who follows the public account;

[0027] The historical book borrowing information is a collection of corresponding information of each book borrowed by the corresponding user in the corresponding library, and the corresponding information of each book borrowing includes the name of the book and the category to which the book belongs;

[0028] When extracting the book borrowing features of the library users who follow the official account in S3, the book borrowing features include the book set corresponding to each category of books borrowed by the library users who follow the official account in the library to be tested and the proportion of the borrowing volume of the corresponding category of books in the total borrowing volume of the corresponding user, and the search results of the user during two consecutive book borrowing periods are bound to the books borrowed once after the corresponding search record, and the binding results are used as part of the book borrowing features;

[0029] The search results include the search content and the threshold number of books in stock corresponding to the search content and the actual number of books in stock corresponding to the search content. The number of books in stock represents the number of corresponding books collected by the library, and the actual number of books in stock represents the actual number of corresponding books remaining in the library after being borrowed during the search.

[0030] Further, when the feature association values ​​between the book borrowing features of the library users who follow the public account and the libraries in the library set are obtained in S4, the feature association value between the book borrowing features of the library users who follow the public account and the g-th library in the library set is recorded as Rg.

[0031] Rg=∑ k＝1 k＝k1 Fk·G(g,k)·Hgk

[0032] Among them, Fk represents the proportion of the k-th category of books borrowed in the book borrowing characteristics of library users who follow the official account to the total borrowing volume of the corresponding user,

[0033] Hgk represents the ratio of the total collection of the g-th library in the library set to the total collection of the library to be tested in the book borrowing characteristics of library users who follow the official account.

[0034] G(g, k) represents the preference bias coefficient of the library user who follows the official account relative to the k-th category of books in the g-th library in the library set, and G(g, k) is equal to the quotient of the bias adaptation value of the search record of the library user who follows the official account relative to the g-th library in the library set and the bias adaptation value of the search record of the library user who follows the official account relative to the library to be tested;

[0035] Obtain the number of intersection elements between the valid search forms and invalid search forms of the library users who follow the official account and the books collected in the g-th library in the library set, and divide the obtained number of intersection elements by the total number of valid search forms and invalid search forms of the library users who follow the official account as the biased adaptation value of the search record of the library users who follow the official account relative to the g-th library in the library set;

[0036] The biased adaptation value of the search records of the library users who follow the official account relative to the library to be tested is equal to the quotient of the number of elements in the valid search form of the library users who follow the official account divided by the total number of valid search forms and invalid search forms of the library users who follow the official account;

[0037] The valid search form of the library users who follow the official account is a set consisting of the search content corresponding to the searched book and the bound borrowed book, and the search content corresponding to the book stock threshold number of 0; the form consisting of the remaining search content in the set consisting of all search contents in the book borrowing characteristics corresponding to the library users who follow the official account, excluding the elements in the corresponding valid search form, is recorded as the invalid search form of the library users who follow the official account.

[0038] The present invention takes into account the differences in the book collections of different libraries and the differences in the category preferences of different users in borrowing books. It combines the user's historical search records to comprehensively analyze the degree of adaptation between different libraries and users compared with the library to be tested, thereby providing data support for obtaining the user's alternative library sequence during the maintenance period of the library to be tested in the subsequent steps.

[0039] Furthermore, when generating the comprehensive associated substitute value of the library to be tested for each element in the library set for the user in S4, the comprehensive associated substitute value of the library to be tested for the g-th element in the library set for the user is recorded as Vg.

[0040] Vg=β1·Rg / Lg,

[0041] Wherein, Lg represents the distance between the library to be tested and the g-th library in the library set; β1 represents the conversion coefficient and β1 is a constant preset in the database.

[0042] A smart library management system based on big data, the system includes the following modules:

[0043] A metropolitan area information collection module, wherein the metropolitan area information collection module obtains library information of all libraries in the metropolitan area within the most recent unit time;

[0044] Maintenance time prediction and announcement module, which obtains the operation time set of each user in the tested library within the most recent unit time, combines the preset maintenance interval cycle of the tested library and the average maintenance time of each maintenance in the historical data, obtains the next best maintenance time of the tested library, generates a maintenance announcement, and pushes it to the corresponding public account users through the bound public account;

[0045] A user borrowing feature analysis module, wherein the user borrowing feature analysis module queries the association relationship between the public account users and the library users to be tested in the database, determines the user category of the public account users, and the category of the public account users includes library users who follow the public account and non-library users who follow the public account; obtains the historical book borrowing information of the library users who follow the public account in the library to be tested, and extracts the book borrowing features of the library users who follow the public account;

[0046] A correlation substitution analysis module, wherein the correlation substitution analysis module obtains a library set consisting of the remaining libraries in all libraries in the metropolitan area except the library to be tested, obtains a feature correlation value between the book borrowing characteristics of library users who follow the public account and each library in the library set; combines the location of each library in the library set with the distance between the library to be tested, and generates a comprehensive correlation substitution value of the library to be tested for each element in the library set for the user;

[0047] A push information management module is provided, wherein the push information management module sorts the libraries in the library set in descending order according to the comprehensive associated replacement values ​​of the tested libraries for the users, generates a replacement library sequence for the corresponding library users who follow the public account during the maintenance period of the tested libraries, binds the sequence with the maintenance announcement and pushes it to the corresponding public account users at the same time. Non-library users who follow the public account do not have a replacement library sequence during the maintenance period of the tested libraries.

[0048] Furthermore, the user borrowing feature analysis module includes a user category classification unit and a borrowing feature extraction unit.

[0049] The user category classification unit queries the database for the association relationship between the public account users and the library users to be tested, and determines the user category of the public account users, wherein the category of the public account users includes library users who follow the public account and non-library users who follow the public account;

[0050] The borrowing feature extraction unit obtains historical book borrowing information of library users who follow the official account in the library to be tested, and extracts book borrowing features of library users who follow the official account;

[0051] Furthermore, the association substitution analysis module includes a feature association value acquisition unit and an association substitution value generation unit.

[0052] The feature association value acquisition unit acquires a library set consisting of the remaining libraries in all libraries in the metropolitan area except the library to be tested, and obtains the feature association values ​​between the book borrowing features of the library users who follow the public account and the libraries in the library set;

[0053] The associated substitute value generating unit generates a comprehensive associated substitute value of the library to be tested for each element in the library set for the user, by combining the location of each library in the library set with the distance between the library to be tested and the library to be tested.

[0054] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention takes into account the situation that users' operations of borrowing and returning books and retrieving books cannot be performed during the maintenance of the library management system, combines the user's borrowing and retrieving information of library books and the library's collection information in historical data, realizes the calibration of library maintenance time and the screening of libraries in the metropolitan area, pushes announcement information to users in time for early warning, and outputs a sequence of alternative libraries that conforms to the user's borrowing and retrieving habits during the library maintenance period, reduces the impact of library maintenance on users, and increases users' use experience of the library. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0056] Figure 1 It is a flow chart of a smart library management method based on big data of the present invention;

[0057] Figure 2 It is a structural schematic diagram of a smart library management system based on big data of the present invention. DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] See also Figure 1 The present invention provides a technical solution: a smart library management method based on big data, the method comprising the following steps:

[0060] S1. Obtaining library information of all libraries in the metropolitan area in the most recent unit time, wherein the library information includes book information, book borrowing and returning circulation information, and library status information, and the library to be tested is a library in the corresponding metropolitan area;

[0061] The unit time in S1 is a constant preset in the database; the book information includes the names of the books in the library and the categories to which they belong; the library borrowing and returning circulation information includes the borrowing and returning records of library users; and the library status information includes the normal operation status and the closed maintenance status.

[0062] S2. Obtain the operation time set of each user in the tested library in the most recent unit time, combine the preset maintenance interval of the tested library and the average maintenance duration of each maintenance in the historical data, obtain the next best maintenance time of the tested library, generate a maintenance announcement, and push it to the corresponding public account users through the bound public account;

[0063] When obtaining the operation time set of each user in the tested library within the most recent unit time in S2, each user corresponds to an operation time set, and each operation time set includes the borrowing operation time, returning operation time and searching operation time of the corresponding user in the tested library;

[0064] The method for obtaining the next optimal maintenance time of the library to be tested in S2 comprises the following steps:

[0065] S21, taking n days as a time period (in this embodiment, taking one week as a time period, i.e., n=7), obtaining an operation proportion information set corresponding to the operation time set of each day in the i1 time periods before the current time in the historical data, wherein the operation proportion information includes the borrowing operation proportion information, the returning operation proportion information, and the searching operation proportion information, and recording the operation proportion information set corresponding to the operation time set of the jth day in the i-th time period before the current time in the historical data as Aij, wherein 1≤i≤i1, and 1≤j≤n;

[0066] S22. Obtain the impact value of the comprehensive operation proportion corresponding to the jth day in a time period, recorded as Qj,

[0067] Qj=[∑ i＝1 i＝i1 r1·Aij1+r2·Aij2+r3·Aij3] / i1,

[0068] Among them, r1, r2 and r3 are constants preset in the database, Aij1 represents the proportion of borrowing operations in Aij, Aij2 represents the proportion of returning operations in Aij, and Aij3 represents the proportion of searching operations in Aij;

[0069] S23, obtain the preset maintenance interval T of the library to be tested, the average maintenance time T1 of each maintenance in the historical data, and the end time tq of the previous maintenance of the library to be tested, obtain the next maintenance time prediction interval of the library to be tested, recorded as [tq+T, tq+T+T1], and obtain the interval corresponding to [tq+T, tq+T+T1] in the divided time period, recorded as

[0070] [B1, B2], where B1 is the sequence of days to which tq+T belongs in the corresponding time period (i.e., which day in the corresponding time period it belongs to), and B2 is the sequence of days to which tq+T+T1 belongs in the corresponding time period, and the units of tq, T and T1 are all days;

[0071] S24, using C as a time adjustment parameter, adjusting [B1, B2] to obtain an interval [B1+C, B2+C], where C is an integer of 0≤C≤n; when C is calculated to be different values, the corresponding ∑ j1＝B1+C j1＝B2+C In Qj1, ∑ j1＝B1+C j1＝B2+C The C value corresponding to the minimum value of Qj1 is recorded as C1, and the optimal maintenance time of the next library under test is obtained [tq+T+C1, tq+T+T1+C1];

[0072] When obtaining the operation ratio information corresponding to the three types of operations, namely, the borrowing operation ratio information, the returning operation ratio information, and the retrieval operation ratio information, construct the operation time chains corresponding to each type of operation on the same day. The chain node in each operation time chain corresponds to an operation time of the corresponding category. The two adjacent operation times corresponding to the same type of operations with an interval time less than the chain interval time are taken as two adjacent nodes on the same operation time chain, and the two adjacent operation times corresponding to the same type of operations with an interval time greater than or equal to the chain interval time are taken as nodes on different operation time chains.

[0073] The proportion of each type of operation is equal to the ratio of the sum of the lengths of each operation time chain corresponding to the corresponding type of operation on that day to the length of one day. The length of the operation time chain is equal to the difference between the maximum operation time and the minimum operation time in each chain node.

[0074] The interchain time length is t0-β·(AM-d),

[0075] Among them, t0 is the chain spacing duration corresponding to the previous time period, AM is the maximum value of the operation proportion information of all categories in the previous time period, d represents the adjustment threshold and d is a constant preset in the database, β is the chain spacing adjustment conversion coefficient and β is a constant preset in the database.

[0076] S3, query the database for the association between the public account user and the library user to be tested, determine the user category of the public account user, the public account user category includes library users who follow the public account and non-library users who follow the public account; obtain the historical book borrowing information of the library users who follow the public account in the library to be tested, and extract the book borrowing characteristics of the library users who follow the public account;

[0077] In S3, when the public account in the database is a library user to be tested, the user category of the public account user is determined to be a library user who follows the public account; when the public account in the database is not a library user to be tested, the user category of the public account user is determined to be a non-library user who follows the public account;

[0078] The historical book borrowing information is a collection of corresponding information of each book borrowed by the corresponding user in the corresponding library, and the corresponding information of each book borrowing includes the name of the book and the category to which the book belongs;

[0079] When extracting the book borrowing features of the library users who follow the official account in S3, the book borrowing features include the book set corresponding to each category of books borrowed by the library users who follow the official account in the library to be tested and the proportion of the borrowing volume of the corresponding category of books in the total borrowing volume of the corresponding user, and the search results of the user during two consecutive book borrowing periods are bound to the books borrowed once after the corresponding search record, and the binding results are used as part of the book borrowing features;

[0080] The search results include the search content and the threshold number of books in stock corresponding to the search content and the actual number of books in stock corresponding to the search content. The number of books in stock represents the number of corresponding books collected by the library, and the actual number of books in stock represents the actual number of corresponding books remaining in the library after being borrowed during the search.

[0081] In the library management system corresponding to this embodiment,

[0082] 1) When managing non-searchable content after a user logs into a library account (i.e., managing the period during which a user does not perform a search operation after logging into a library account),

[0083] Acquire a user's historical borrowing record data set, wherein each element of the historical borrowing record data set is a borrowing record subset corresponding to a borrowing category of the corresponding user, and each borrowing record subset includes the names of each book borrowed by the corresponding user in the corresponding category;

[0084] Obtain historical borrowing record data sets corresponding to different users, and obtain a comprehensive book recommendation data set corresponding to the user to be tested, wherein the comprehensive book recommendation data set includes multiple elements, and each element corresponds to a borrowing record subset corresponding to a borrowing category of the user to be tested;

[0085] Compare the historical borrowing record data sets corresponding to the rest of the users in the database with the historical borrowing record data sets corresponding to the user to be tested, and calculate the similarity between the elements (borrowing record subsets) with the same corresponding book categories in the two. The obtained similarity is equal to the ratio of the number of elements in the intersection of the two elements corresponding to the corresponding book categories in the two to the number of elements in the union of the two elements corresponding to the corresponding book categories in the two.

[0086] Extract the historical borrowing record data sets corresponding to the remaining users of the database, and use the element with the greatest similarity with each borrowing record subset in the historical borrowing record data set of the user to be tested as an element in the comprehensive book recommendation data set corresponding to the user to be tested, and when there are multiple elements with the greatest similarity with the same borrowing record subset in the historical borrowing record data set of the user to be tested, then the corresponding element in the comprehensive book recommendation data set corresponding to the user to be tested is the union of the elements with the greatest similarity with the same borrowing record subset in the historical borrowing record data set of the user to be tested;

[0087] The library management system in this implementation obtains the historical borrowing record data set and the comprehensive book recommendation data set corresponding to the user to be tested, and counts the set of book names common in the historical borrowing record data set and the comprehensive book recommendation data set, which is recorded as TSG. The set of remaining book names except the elements in TSG in all the book names in the comprehensive book recommendation data set corresponding to the user to be tested is counted, which is recorded as TSS, and the book names in TSS are arranged in descending order according to the total number of borrowing times in the corresponding library to generate an interest book recommendation sequence, and the interest book recommendation sequence corresponding to the user to be tested is pushed to the user to be tested within the time period after the user to be tested logs in to the library account and does not perform a search operation; the interest book recommendation sequence of the user to be tested is updated and changed in real time, and is comprehensively affected by the borrowing record of the user to be tested and the borrowing records of other users of the library;

[0088] 2) When managing the user's search content,

[0089] When the library under test is not in the maintenance stage, the library management system under test obtains the user's search content in real time. When the book searched by the user belongs to the books collected by the library under test, the borrowing and inventory status of the searched book in the library under test at the current search time (the number of books of the corresponding name remaining after borrowing) is displayed. If the inventory of the searched book corresponding to the current time of the library under test is greater than or equal to 1, it is displayed that the user can borrow the corresponding searched book;

[0090] When the book searched by the user does not belong to the books collected by the library to be tested, or when the book searched by the user belongs to the books collected by the library to be tested and the corresponding stock of the searched book in the library to be tested at the current time is less than 1, the library management system in this embodiment will extract the keywords in the current search content of the user, and combine the extracted keywords to obtain all combinations of the extracted keywords, each combination containing all or part of the extracted keywords; further obtain the names of all the books in the collection of the library to be tested that include the keywords in each combination, and recommend the names of the books whose corresponding books have a stock greater than or equal to 1 and are associated with the user (referring to the case where the category to which the corresponding book belongs belongs to the category set of the user's historical borrowed books, wherein the category set of the user's historical borrowed books is a set consisting of the categories corresponding to the user's historical borrowed books) to the user,

[0091] If the name of the book recommended to the user appears in the first borrowing record after the user's corresponding search content, the user's corresponding search content is determined to be a valid search content. Otherwise, the user's corresponding search content is determined to be an invalid search content.

[0092] S4, obtaining a library set consisting of the remaining libraries in all libraries in the metropolitan area except the library to be tested, obtaining the book borrowing characteristics of library users who follow the public account and the characteristic association values ​​between each library in the library set; combining the location of each library in the library set with the distance between the library to be tested and generating a comprehensive association substitution value of the library to be tested for each element in the library set for the user;

[0093] When obtaining the feature association values ​​between the book borrowing features of the library users who follow the public account and the libraries in the library set in S4, the feature association value between the book borrowing features of the library users who follow the public account and the g-th library in the library set is recorded as Rg.

[0094] Rg=∑ k＝1 k＝k1 Fk·G(g,k)·Hgk

[0095] Among them, Fk represents the proportion of the k-th category of books borrowed in the book borrowing characteristics of library users who follow the official account to the total borrowing volume of the corresponding user.

[0096] Hgk represents the ratio of the total collection of the g-th library in the library set to the total collection of the library to be tested in the book borrowing characteristics of library users who follow the official account.

[0097] G(g, k) represents the preference bias coefficient of the library user who follows the official account relative to the k-th category of books in the g-th library in the library set, and G(g, k) is equal to the quotient of the bias adaptation value of the search record of the library user who follows the official account relative to the g-th library in the library set and the bias adaptation value of the search record of the library user who follows the official account relative to the library to be tested;

[0098] In this embodiment, when k∈[1, k1], the sum of the corresponding Fk is equal to 1, and Fk serves the purpose of weight adjustment in the normalization process, while G(g, k) determines the degree of user adaptation of the books of the g-th library relative to the library to be tested based on the search form (valid search form and invalid search form), and serves the purpose of secondary adjustment;

[0099] Obtain the number of intersection elements between the valid search forms and invalid search forms of the library users who follow the official account and the books collected in the g-th library in the library set, and divide the obtained number of intersection elements by the total number of valid search forms and invalid search forms of the library users who follow the official account as the biased adaptation value of the search record of the library users who follow the official account relative to the g-th library in the library set;

[0100] The biased adaptation value of the search records of the library users who follow the official account relative to the library to be tested is equal to the quotient of the number of elements in the valid search form of the library users who follow the official account divided by the total number of valid search forms and invalid search forms of the library users who follow the official account;

[0101] The valid search form of the library users who follow the official account is a set consisting of the search content corresponding to the searched book and the bound borrowed book, and the search content corresponding to the book stock threshold number of 0; the form consisting of the remaining search content in the set consisting of all search contents in the book borrowing characteristics corresponding to the library users who follow the official account, excluding the elements in the corresponding valid search form, is recorded as the invalid search form of the library users who follow the official account.

[0102] When generating the comprehensive associated substitute value of the library to be tested for each element in the library set for the user in S4, the comprehensive associated substitute value of the library to be tested for the g-th element in the library set for the user is recorded as Vg.

[0103] Vg=β1·Rg / Lg,

[0104] Wherein, Lg represents the distance between the library to be tested and the g-th library in the library set; β1 represents the conversion coefficient and β1 is a constant preset in the database.

[0105] In this embodiment, the comprehensive association substitution value of the g-th element in the library set for the user's tested library is affected by both Rg and Lg. When the value of Rg is larger and the value of Lg is smaller, the corresponding value of Vg is larger.

[0106] That is, Vg is positively correlated with Rg, and Vg is negatively correlated with Lg;

[0107] S5. Sort the libraries in the library set in descending order according to the comprehensive associated replacement values ​​of the library to be tested for the user, generate a replacement library sequence for the library users who follow the corresponding public account during the maintenance period of the library to be tested, bind it with the maintenance announcement and push it to the corresponding public account users at the same time. In the push information of different public account users, the replacement library sequences of different library users who follow the public account during the maintenance period of the library to be tested are different. Non-library users who follow the public account do not have a replacement library sequence during the maintenance period of the library to be tested.

[0108] like Figure 2 As shown, a smart library management system based on big data includes the following modules:

[0109] A metropolitan area information collection module, wherein the metropolitan area information collection module obtains library information of all libraries in the metropolitan area within the most recent unit time;

[0110] Maintenance time prediction and announcement module, which obtains the operation time set of each user in the tested library within the most recent unit time, combines the preset maintenance interval cycle of the tested library and the average maintenance time of each maintenance in the historical data, obtains the next best maintenance time of the tested library, generates a maintenance announcement, and pushes it to the corresponding public account users through the bound public account;

[0111] A user borrowing feature analysis module, wherein the user borrowing feature analysis module queries the association relationship between the public account users and the library users to be tested in the database, determines the user category of the public account users, and the category of the public account users includes library users who follow the public account and non-library users who follow the public account; obtains the historical book borrowing information of the library users who follow the public account in the library to be tested, and extracts the book borrowing features of the library users who follow the public account;

[0112] A correlation substitution analysis module, wherein the correlation substitution analysis module obtains a library set consisting of the remaining libraries in all libraries in the metropolitan area except the library to be tested, obtains a feature correlation value between the book borrowing characteristics of library users who follow the public account and each library in the library set; combines the location of each library in the library set with the distance between the library to be tested, and generates a comprehensive correlation substitution value of the library to be tested for each element in the library set for the user;

[0113] A push information management module is provided, wherein the push information management module sorts the libraries in the library set in descending order according to the comprehensive associated replacement values ​​of the tested libraries for the users, generates a replacement library sequence for the corresponding library users who follow the public account during the maintenance period of the tested libraries, binds the sequence with the maintenance announcement and pushes it to the corresponding public account users at the same time. Non-library users who follow the public account do not have a replacement library sequence during the maintenance period of the tested libraries.

[0114] The user borrowing feature analysis module includes a user category classification unit and a borrowing feature extraction unit.

[0115] The user category classification unit queries the database for the association relationship between the public account users and the library users to be tested, and determines the user category of the public account users, wherein the category of the public account users includes library users who follow the public account and non-library users who follow the public account;

[0116] The borrowing feature extraction unit obtains historical book borrowing information of library users who follow the official account in the library to be tested, and extracts book borrowing features of library users who follow the official account;

[0117] The association substitution analysis module includes a feature association value acquisition unit and an association substitution value generation unit.

[0118] The feature association value acquisition unit acquires a library set consisting of the remaining libraries in all libraries in the metropolitan area except the library to be tested, and obtains the feature association values ​​between the book borrowing features of the library users who follow the public account and the libraries in the library set;

[0119] The associated substitute value generating unit generates a comprehensive associated substitute value of the library to be tested for each element in the library set for the user, by combining the location of each library in the library set with the distance between the library to be tested and the library to be tested.

[0120] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0121] Finally, it should be noted that the above description is only 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 aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A smart library management method based on big data, characterized in that: The method comprises the following steps: S1. Obtaining library information of all libraries in the metropolitan area in the most recent unit time, wherein the library information includes book information, book borrowing and returning circulation information, and library status information. The library to be tested is a library in the corresponding metropolitan area; S2. Obtain the operation time set of each user in the tested library in the most recent unit time, combine the preset maintenance interval of the tested library and the average maintenance duration of each maintenance in the historical data, obtain the next best maintenance time of the tested library, generate a maintenance announcement, and push it to the corresponding public account users through the bound public account; S3, query the database for the association between the public account user and the library user to be tested, determine the user category of the public account user, the public account user category includes library users who follow the public account and non-library users who follow the public account; obtain the historical book borrowing information of the library users who follow the public account in the library to be tested, and extract the book borrowing characteristics of the library users who follow the public account; S4, obtaining a library set consisting of the remaining libraries in all libraries in the metropolitan area except the library to be tested, obtaining the book borrowing characteristics of library users who follow the public account and the characteristic association values ​​between each library in the library set; combining the location of each library in the library set with the distance between the library to be tested and generating a comprehensive association substitution value of the library to be tested for each element in the library set for the user; S5. Sort the libraries in the library set in descending order according to the comprehensive associated replacement values ​​of the library to be tested for the user, generate a replacement library sequence for the library users who follow the corresponding public account during the maintenance period of the library to be tested, bind it with the maintenance announcement and push it to the corresponding public account users at the same time. In the push information of different public account users, the replacement library sequences of different library users who follow the public account during the maintenance period of the library to be tested are different. Non-library users who follow the public account do not have a replacement library sequence during the maintenance period of the library to be tested.

2. The method for managing a smart library based on big data according to claim 1, characterized in that: The unit time in S1 is a constant preset in the database; the book information includes the names of the books in the library and the categories to which they belong; the library borrowing and returning circulation information includes the borrowing and returning records of library users; and the library status information includes the normal operation status and the closed maintenance status.

3. The method for managing a smart library based on big data according to claim 1, characterized in that: When obtaining the operation time set of each user in the tested library within the most recent unit time in S2, each user corresponds to an operation time set, and each operation time set includes the borrowing operation time, returning operation time and searching operation time of the corresponding user in the tested library; The method for obtaining the next optimal maintenance time of the library to be tested in S2 comprises the following steps: S21, taking n days as a time period, obtaining an operation proportion information set corresponding to the operation time set of each day in the i1 time periods before the current time in the historical data, wherein the operation proportion information includes the borrowing operation proportion information, the returning operation proportion information and the retrieval operation proportion information, and recording the operation proportion information set corresponding to the operation time set of the jth day in the i-th time period before the current time in the historical data as Aij, wherein 1≤i≤i1, and 1≤j≤n; S22. Obtain the impact value of the comprehensive operation proportion corresponding to the jth day in a time period, recorded as Qj, ； Among them, r1, r2 and r3 are constants preset in the database, Aij1 represents the proportion of borrowing operations in Aij, Aij2 represents the proportion of returning operations in Aij, and Aij3 represents the proportion of searching operations in Aij; S23, obtaining the preset maintenance interval T of the library to be tested, the average maintenance duration T1 of each maintenance in the historical data, and the end time tq of the previous maintenance of the library to be tested, and obtaining the predicted interval of the next maintenance time of the library to be tested, recorded as [tq+T, tq+T+T1], and obtaining the interval corresponding to [tq+T, tq+T+T1] in the divided time period, recorded as [B1, B2], wherein B1 is the sequence of days to which tq+T belongs in the corresponding time period, and B2 is the sequence of days to which tq+T+T1 belongs in the corresponding time period; S24, using C as a time adjustment parameter, adjusting [B1, B2] to obtain an interval [B1+C, B2+C], where C is an integer of 0≤C≤n; when C is calculated to be different values, the corresponding middle, The C value corresponding to the minimum value is recorded as C1, and the optimal maintenance time of the next library under test is obtained [tq+T+C1, tq+T+T1+C1]; When obtaining the operation ratio information corresponding to the three types of operations, namely, the borrowing operation ratio information, the returning operation ratio information, and the retrieval operation ratio information, construct the operation time chains corresponding to each type of operation on the same day. The chain node in each operation time chain corresponds to an operation time of the corresponding category. The two adjacent operation times corresponding to the same type of operations with an interval time less than the chain interval time are taken as two adjacent nodes on the same operation time chain, and the two adjacent operation times corresponding to the same type of operations with an interval time greater than or equal to the chain interval time are taken as nodes on different operation time chains. The proportion of each type of operation is equal to the ratio of the sum of the lengths of each operation time chain corresponding to the corresponding type of operation on that day to the length of one day. The length of the operation time chain is equal to the difference between the maximum operation time and the minimum operation time in each chain node. The interchain time length is t0-β·(AM-d), Among them, t0 is the chain spacing duration corresponding to the previous time period, AM is the maximum value of the operation proportion information of all categories in the previous time period, d represents the adjustment threshold and d is a constant preset in the database, β is the chain spacing adjustment conversion coefficient and β is a constant preset in the database.

4. The method for managing a smart library based on big data according to claim 1, characterized in that: In S3, when the public account in the database is a library user to be tested, the user category of the public account user is determined to be a library user who follows the public account; when the public account in the database is not a library user to be tested, the user category of the public account user is determined to be a non-library user who follows the public account; The historical book borrowing information is a collection of corresponding information of each book borrowed by the corresponding user in the corresponding library, and the corresponding information of each book borrowing includes the name of the book and the category to which the book belongs; When extracting the book borrowing features of the library users who follow the official account in S3, the book borrowing features include the book set corresponding to each category of books borrowed by the library users who follow the official account in the library to be tested and the proportion of the borrowing volume of the corresponding category of books in the total borrowing volume of the corresponding user, and the search results of the user during two consecutive book borrowing periods are bound to the books borrowed once after the corresponding search record, and the binding results are used as part of the book borrowing features; The search results include the search content and the threshold number of books in stock corresponding to the search content and the actual number of books in stock corresponding to the search content. The number of books in stock represents the number of corresponding books collected by the library, and the actual number of books in stock represents the actual number of corresponding books remaining in the library after being borrowed during the search.

5. The method for managing a smart library based on big data according to claim 4 is characterized by: When obtaining the feature association values ​​between the book borrowing features of the library users who follow the public account and the libraries in the library set in S4, the feature association value between the book borrowing features of the library users who follow the public account and the g-th library in the library set is recorded as Rg. ； Among them, Fk represents the proportion of the k-th category of books borrowed in the book borrowing characteristics of library users who follow the official account to the total borrowing volume of the corresponding user, Hgk represents the ratio of the total collection of the g-th library in the library set to the total collection of the library to be tested in the book borrowing characteristics of library users who follow the official account. G(g, k) represents the preference bias coefficient of the library user who follows the official account relative to the k-th category of books in the g-th library in the library set, and G(g, k) is equal to the quotient of the bias adaptation value of the search record of the library user who follows the official account relative to the g-th library in the library set and the bias adaptation value of the search record of the library user who follows the official account relative to the library to be tested; Obtain the number of intersection elements between the valid search forms and invalid search forms of the library users who follow the official account and the books collected in the g-th library in the library set, and divide the obtained number of intersection elements by the total number of valid search forms and invalid search forms of the library users who follow the official account as the biased adaptation value of the search record of the library users who follow the official account relative to the g-th library in the library set; The biased adaptation value of the search records of the library users who follow the official account relative to the library to be tested is equal to the quotient of the number of elements in the valid search form of the library users who follow the official account divided by the total number of valid search forms and invalid search forms of the library users who follow the official account; The valid search form of the library users who follow the official account is a set consisting of the search content corresponding to the searched book and the bound borrowed book, and the search content corresponding to the book stock threshold number of 0; the form consisting of the remaining search content in the set consisting of all search contents in the book borrowing characteristics corresponding to the library users who follow the official account, excluding the elements in the corresponding valid search form, is recorded as the invalid search form of the library users who follow the official account.

6. The method for managing a smart library based on big data according to claim 5 is characterized by: When generating the comprehensive associated substitute value of the library to be tested for each element in the library set for the user in S4, the comprehensive associated substitute value of the library to be tested for the g-th element in the library set for the user is recorded as Vg. Vg = β1·Rg / Lg, Wherein, Lg represents the distance between the library to be tested and the g-th library in the library set; β1 represents the conversion coefficient and β1 is a constant preset in the database.

7. A smart library management system based on big data, the system is implemented by a smart library management method based on big data as described in any one of claims 1 to 6, characterized in that: The system includes the following modules: A metropolitan area information collection module, wherein the metropolitan area information collection module obtains library information of all libraries in the metropolitan area within the most recent unit time; Maintenance time prediction and announcement module, which obtains the operation time set of each user in the tested library within the most recent unit time, combines the preset maintenance interval cycle of the tested library and the average maintenance time of each maintenance in the historical data, obtains the next best maintenance time of the tested library, generates a maintenance announcement, and pushes it to the corresponding public account users through the bound public account; A user borrowing feature analysis module, wherein the user borrowing feature analysis module queries the association relationship between the public account users and the library users to be tested in the database, determines the user category of the public account users, and the category of the public account users includes library users who follow the public account and non-library users who follow the public account; obtains the historical book borrowing information of the library users who follow the public account in the library to be tested, and extracts the book borrowing features of the library users who follow the public account; A correlation substitution analysis module, wherein the correlation substitution analysis module obtains a library set consisting of the remaining libraries in all libraries in the metropolitan area except the library to be tested, obtains a feature correlation value between the book borrowing characteristics of library users who follow the public account and each library in the library set; combines the location of each library in the library set with the distance between the library to be tested, and generates a comprehensive correlation substitution value of the library to be tested for each element in the library set for the user; A push information management module is provided, wherein the push information management module sorts the libraries in the library set in descending order according to the comprehensive associated replacement values ​​of the tested libraries for the users, generates a replacement library sequence for the corresponding library users who follow the public account during the maintenance period of the tested libraries, binds the sequence with the maintenance announcement and pushes it to the corresponding public account users at the same time. Non-library users who follow the public account do not have a replacement library sequence during the maintenance period of the tested libraries.

8. The smart library management system based on big data according to claim 7 is characterized by: The user borrowing feature analysis module includes a user category classification unit and a borrowing feature extraction unit. The user category classification unit queries the database for the association relationship between the public account users and the library users to be tested, and determines the user category of the public account users, wherein the category of the public account users includes library users who follow the public account and non-library users who follow the public account; The borrowing feature extraction unit obtains historical book borrowing information of library users who follow the official account in the library to be tested, and extracts book borrowing features of the library users who follow the official account.

9. The smart library management system based on big data according to claim 7 is characterized by: The association substitution analysis module includes a feature association value acquisition unit and an association substitution value generation unit. The feature association value acquisition unit acquires a library set consisting of the remaining libraries in all libraries in the metropolitan area except the library to be tested, and obtains the feature association values ​​between the book borrowing features of the library users who follow the public account and the libraries in the library set; The associated substitute value generating unit generates a comprehensive associated substitute value of the library to be tested for each element in the library set for the user, by combining the location of each library in the library set with the distance between the library to be tested and the library to be tested.

Images