CN117076780B - Book mutual borrowing recommendation method and system based on big data - Google Patents
Book mutual borrowing recommendation method and system based on big data Download PDFInfo
- Publication number
- CN117076780B CN117076780B CN202311131241.5A CN202311131241A CN117076780B CN 117076780 B CN117076780 B CN 117076780B CN 202311131241 A CN202311131241 A CN 202311131241A CN 117076780 B CN117076780 B CN 117076780B
- Authority
- CN
- China
- Prior art keywords
- user
- book
- borrowing
- acquiring
- users
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 239000013598 vector Substances 0.000 claims abstract description 87
- 238000013528 artificial neural network Methods 0.000 claims abstract description 20
- 238000007781 pre-processing Methods 0.000 claims abstract description 15
- 239000011159 matrix material Substances 0.000 claims description 66
- 230000003993 interaction Effects 0.000 claims description 43
- 230000006399 behavior Effects 0.000 claims description 40
- 238000010586 diagram Methods 0.000 claims description 21
- 230000007774 longterm Effects 0.000 claims description 20
- 238000012216 screening Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 10
- 230000004931 aggregating effect Effects 0.000 claims description 4
- 230000010354 integration Effects 0.000 claims description 4
- 238000011176 pooling Methods 0.000 claims description 4
- 230000001502 supplementing effect Effects 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 2
- 238000010276 construction Methods 0.000 claims 2
- 230000002776 aggregation Effects 0.000 description 4
- 238000004220 aggregation Methods 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 4
- 238000001914 filtration Methods 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000008034 disappearance Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000003542 behavioural effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
- G06F16/9024—Graphs; Linked lists
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/042—Knowledge-based neural networks; Logical representations of neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a book mutual borrowing recommendation method and system based on big data, comprising the steps of obtaining current idle book resource information of a user to generate book sharing characteristics of the user; integrating book borrowing records of users and historical reading behavior data by utilizing a big data method to perform preprocessing, and extracting data characteristics to locate book demand characteristics of current users; using the graphic representation to represent the user and the book items to generate undirected heterograms, and generating additional features of user nodes through book sharing features and book demand features; based on the graph convolution neural network, a mutual borrowing recommendation model is built, a graph structure is obtained, user feature vector representations with book item preferences are learned, and preference information of users is predicted to recommend mutual borrowing objects for the users. The invention can quickly and accurately find the book resources meeting the demands for users, saves borrowing and book returning time, greatly reduces book purchasing cost and improves the utilization rate of idle book resources.
Description
Technical Field
The invention relates to the technical field of book recommendation, in particular to a book mutual borrowing recommendation method and system based on big data.
Background
With the rapid development of Internet, big data and cloud computing, the indiscriminate recommended content of the existing book management system cannot meet the needs of diversification and individuation of users, and the data value of a large amount of user data accumulated in the book management system all the year round is still to be mined. Therefore, innovation by using big data technology becomes new kinetic energy for library development and transformation. The user portrait is a data analysis tool for intuitively displaying the user overall view by processing data related to the user and extracting user feature vectors so as to obtain a user feature model. The user portraits are applied to the field of libraries, a portraits model is constructed through data analysis, the user preferences, the user interests and the user behaviors are effectively predicted by utilizing the portraits, and accurate recommendation of books can be realized based on the user preferences, so that personalized demands of readers are met.
At present, books in public libraries are more and more abundant, but the library is slow in book update speed, and many readers cannot borrow the latest professional books. Meanwhile, in a traditional book borrowing mode, borrowing time is time-consuming and limited, readers keep possession of books in the validity period, and therefore many hot books are often difficult to borrow. And for books purchased by readers, the books are idle after the readers see or use the books, and the books are not reused for a long time, so that the effective utilization rate of the books is low and the resources are wasted. Therefore, how to construct an intelligent book borrowing platform and a book borrowing recommending system with reasonable design are used for recommending books possibly needed to users, and improving the utilization rate of idle data is one of the problems to be solved.
Disclosure of Invention
In order to solve the technical problems, the invention provides a book mutual borrowing recommendation method and system based on big data.
The invention provides a book mutual borrowing recommendation method based on big data, which comprises the following steps:
acquiring current idle book resource information of a user, extracting key word characteristics of the idle book resource information and generating book sharing characteristics of the user;
acquiring book borrowing records and historical reading behavior data of a user by using a big data method, integrating the book borrowing records and the historical reading behavior data to perform preprocessing, and extracting data characteristics to locate book demand characteristics of the current user;
using the graphic representation to represent the user and the book items to generate undirected heterograms, and generating additional features of user nodes through the book sharing features and the book demand features;
based on the graph convolution neural network, a mutual borrowing recommendation model is built, a graph structure is learned, user feature vector representations with book item preferences are learned according to the graph structure, preference information of a user is predicted, and mutual borrowing objects are recommended to the user according to the preference information.
In the scheme, the current idle book resource information of the user is acquired, and the keyword characteristics of the idle book resource information are extracted to generate book sharing characteristics of the user, specifically:
acquiring current idle book resource information of a user by utilizing code scanning input and book name keyword input, acquiring book borrowing records of the user in a target library, and extracting borrowed time of the book borrowed by the user according to the book borrowing records;
feeding the borrowed books with the borrowed time length being longer than a preset time length threshold back to a user for selection, supplementing current idle book resource information according to the selection of the user, acquiring book profile data according to the current idle book resource information, and performing word segmentation on the book profile data to acquire corresponding word vectors;
carrying out weighted average on the word vectors to construct sentence vector expression, obtaining the occurrence frequency and the distribution breadth of the word vectors in the book profile data, and screening the word vectors meeting preset standards as description keyword vectors;
and extracting key word characteristics according to the book name key words and the description key word vectors, and acquiring book sharing characteristics of the user based on the key word characteristic set sentence vector expression.
In this scheme, integrate books borrow record and history and read behavioral data and carry out the preliminary treatment, draw data characteristic location current user's books demand characteristic, specifically do:
the method comprises the steps of obtaining borrowing records of a user in a target library through big data retrieval, obtaining historical reading behavior data according to interaction behaviors of the user in a reading website, and integrating the borrowing records and the historical reading behavior data with a time stamp;
preprocessing the acquired time sequence after data integration to generate a user sequence, selecting the user sequence with preset time t to acquire a short-term intention item set, embedding the short-term intention item set into a potential space to stack up to generate a matrix, importing the matrix into a convolution layer to perform convolution operation, and acquiring short-term interaction characteristics through maximum pooling;
acquiring a user sequence before a preset time t to acquire a long-term intention project set, acquiring long-term interaction characteristics of a user through a gating circulating unit layer, importing the long-term interaction characteristics and the short-term interaction characteristics into a self-attention layer, and acquiring self-attention weights;
and aggregating the long-term interaction characteristics and the short-term interaction characteristics according to the self-attention weight, and outputting book demand characteristics of the current user.
In the scheme, the user and book items are represented by using the graphic representation to generate undirected heterograms, and additional characteristics of user nodes are generated through the book sharing characteristics and the book demand characteristics, specifically:
taking a user and a book item as nodes, obtaining an undirected heterogram through graph representation, and setting an edge structure according to interaction of the user and the book item, wherein the undirected heterogram is defined as G= (V, E), V is a set of nodes, and E is an edge structure set;
generating additional features of user nodes through the book sharing features and the book demand features, filtering neighbor nodes of users through similarity calculation according to the additional features, and constructing a user adjacency matrix through the neighbor nodes.
In the scheme, a mutual borrowing recommendation model is constructed based on a graph convolution neural network, and specifically comprises the following steps:
constructing a mutual borrowing recommendation model based on a graph convolution neural network, constructing a user subgraph according to a user adjacency matrix, splicing the undirected heterograph and the user subgraph to construct a user-book project graph, and learning the user-book project graph by using the mutual borrowing recommendation model;
setting an encoder to respectively encode and express a user-book project diagram, adding user position description in the user-book project diagram, acquiring a user distance adjacency matrix based on distance information, and re-performing similarity calculation on a user in the user-book project diagram to acquire a user similarity adjacency matrix;
acquiring a book item adjacency matrix through additional features of user nodes, acquiring user embedded representation by combining the distance adjacency matrix and the similarity adjacency matrix with the book item adjacency matrix in each layer of encoder, and carrying out weighted summation and vector splicing to acquire user feature vector representation with book item preference;
and calculating the inner product of the user characteristic vector representation among the users to acquire the borrowing adaptability among the users.
In the scheme, the inner product represented by the user feature vector among the users is calculated to obtain the borrowing adaptability among the users, and the borrowing adaptability is specifically as follows:
acquiring user feature vector representations of target users through a mutual borrowing recommendation model, acquiring book item preferences of the target users based on the user feature vector representations, screening other users meeting preset similarity standards according to the book item preferences, and acquiring other user sets with similar preferences;
calculating inner products between user characteristic vector representations of other users and user characteristic vector representations of target users in the other user sets, and acquiring borrowing adaptability between users according to the inner products;
and generating borrowing priority according to the borrowing suitability, and pushing the borrowing priority to a target user according to a preset method.
The second aspect of the present invention also provides a book mutual borrowing recommendation system based on big data, the system comprising: the book mutual borrowing recommendation method based on big data comprises a memory and a processor, wherein the memory comprises a book mutual borrowing recommendation method program based on big data, and the following steps are realized when the book mutual borrowing recommendation method program based on big data is executed by the processor:
acquiring current idle book resource information of a user, extracting key word characteristics of the idle book resource information and generating book sharing characteristics of the user;
acquiring book borrowing records and historical reading behavior data of a user by using a big data method, integrating the book borrowing records and the historical reading behavior data to perform preprocessing, and extracting data characteristics to locate book demand characteristics of the current user;
using the graphic representation to represent the user and the book items to generate undirected heterograms, and generating additional features of user nodes through the book sharing features and the book demand features;
based on the graph convolution neural network, a mutual borrowing recommendation model is built, a graph structure is learned, user feature vector representations with book item preferences are learned according to the graph structure, preference information of a user is predicted, and mutual borrowing objects are recommended to the user according to the preference information.
The invention discloses a book mutual borrowing recommendation method and system based on big data, comprising the steps of obtaining current idle book resource information of a user to generate book sharing characteristics of the user; integrating book borrowing records of users and historical reading behavior data by utilizing a big data method to perform preprocessing, and extracting data characteristics to locate book demand characteristics of current users; using the graphic representation to represent the user and the book items to generate undirected heterograms, and generating additional features of user nodes through book sharing features and book demand features; based on the graph convolution neural network, a mutual borrowing recommendation model is built, a graph structure is obtained, user feature vector representations with book item preferences are learned, and preference information of users is predicted to recommend mutual borrowing objects for the users. The invention can quickly and accurately find the book resources meeting the demands for users, saves borrowing and book returning time, greatly reduces book purchasing cost and improves the utilization rate of idle book resources.
Drawings
FIG. 1 is a flow chart showing a book mutual borrowing recommendation method based on big data;
FIG. 2 is a flow chart illustrating the present invention locating book demand features of the current user;
FIG. 3 illustrates a flow chart of the present invention for constructing a mutual borrowing recommendation model based on a graph convolution neural network;
FIG. 4 is a block diagram showing a book inter-borrowing recommendation system based on big data according to the present invention.
Detailed Description
In order that the above-recited objects, features and advantages of the present invention will be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description. It should be noted that, in the case of no conflict, the embodiments of the present application and the features in the embodiments may be combined with each other.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those described herein, and therefore the scope of the present invention is not limited to the specific embodiments disclosed below.
FIG. 1 shows a flowchart of a book mutual borrowing recommendation method based on big data.
As shown in fig. 1, the first aspect of the present invention provides a book mutual borrowing recommendation method based on big data, including:
s102, acquiring current idle book resource information of a user, extracting keyword features of the idle book resource information and generating book sharing features of the user;
s104, acquiring book borrowing records and historical reading behavior data of a user by using a big data method, integrating the book borrowing records and the historical reading behavior data for preprocessing, and extracting data characteristics to locate book demand characteristics of the current user;
s106, representing the user and the book items by using the graphic representation to generate an undirected heterogram, and generating additional features of user nodes through the book sharing features and the book demand features;
s108, constructing a mutual borrowing recommendation model based on a graph convolution neural network, learning a graph structure, learning user feature vector representations with book item preferences according to the graph structure, predicting preference information of a user, and recommending a mutual borrowing object for the user according to the preference information.
The method comprises the steps of acquiring current idle book resource information of a user by utilizing code scanning input and book name keyword input, acquiring book borrowing records of the user in a target library, and extracting borrowed time of the book borrowed by the user according to the book borrowing records; acquiring average reading time length of a book borrowed by a user through a big data method, setting a preset time length threshold according to the average reading time length, feeding the borrowed book with the borrowed time length being greater than the preset time length threshold back to the user for selection, reading whether the user has read or is not interested in reading no longer, supplementing current idle book resource information according to the selection of the user, acquiring book profile data according to the current idle book resource information, and segmenting the book profile data to acquire corresponding word vectors; carrying out weighted average on the word vectors to construct sentence vector expression, obtaining the occurrence frequency and the distribution breadth of the word vectors in the book profile data, and screening the word vectors meeting preset standards as description keyword vectors; and extracting key word characteristics according to the book name key words and the description key word vectors, and acquiring book sharing characteristics of the user based on the key word characteristic set sentence vector expression.
FIG. 2 is a flow chart showing the locating book demand features of the current user of the present invention.
According to the embodiment of the invention, the book borrowing record and the historical reading behavior data are integrated for preprocessing, and the book demand characteristics of the current user are positioned by extracting the data characteristics, specifically:
s202, acquiring borrowing records of a user in a target library through big data retrieval, acquiring historical reading behavior data according to interaction behaviors of the user in a reading website, and integrating the borrowing records and the historical reading behavior data with a time stamp;
s204, preprocessing the acquired time sequence after data integration to generate a user sequence, selecting the user sequence with preset time t to acquire a short-term intention item set, embedding the short-term intention item set into a potential space to stack up a generating matrix, importing the matrix into a convolution layer to perform convolution operation, and acquiring short-term interaction characteristics through maximum pooling;
s206, acquiring a user sequence before a preset time t to acquire a long-term intention project set, acquiring long-term interaction characteristics of a user through a gating circulation unit layer, guiding the long-term interaction characteristics and the short-term interaction characteristics into a self-attention layer, and acquiring self-attention weights;
and S208, aggregating the long-term interaction characteristics and the short-term interaction characteristics according to the self-attention weight, and outputting book demand characteristics of the current user.
It should be noted that, the historical reading behavior data includes browsing, searching, borrowing, renewing, collecting, consulting and applying for demands, and the historical reading behavior data not only reflects the searching demands and searching habits of the user, but also maps the research trends and potential interesting contents of the user. Capturing short-term interaction characteristics and long-term interaction characteristics of a user through each continuous item in a user sequence set, acquiring short-term correlation in the user sequence by utilizing a convolution layer, and learning specific sequence characteristics through a convolution filter to output the short-term interaction characteristics; the gating loop layer captures the time dependence of the user sequence through the gating loop cell structure. The gating circulation unit can not clear the previous information along with the time, and can keep the related information and transmit the related information to the next unit, so that the problem of gradient disappearance is avoided. The self-Attention module is used for representing the influence of short-term interaction characteristics and long-term interaction characteristics on demand characteristics, different self-Attention weights are endowed for aggregation, book demand characteristics of a current user are obtained, and the Attention mechanism Attention (Q, K, V) in the self-Attention layer has a calculation formula as follows:
where Q represents a query matrix, K represents a key matrix, V represents a value matrix,representing the scale factor and T representing the matrix transpose.
FIG. 3 illustrates a flow chart of the present invention for constructing a mutual borrowing recommendation model based on a graph roll-up neural network.
According to the embodiment of the invention, a mutual borrowing recommendation model is constructed based on a graph convolution neural network, and the method specifically comprises the following steps:
s302, constructing a mutual borrowing recommendation model based on a graph convolution neural network, constructing a user subgraph according to a user adjacency matrix, splicing the undirected heterograph and the user subgraph to construct a user-book project graph, and learning the user-book project graph by using the mutual borrowing recommendation model;
s304, setting an encoder to respectively encode and represent a user-book project diagram, adding user position description in the user-book project diagram, acquiring a user distance adjacency matrix based on distance information, and re-performing similarity calculation on a user in the user-book project diagram to acquire a user similarity adjacency matrix;
s306, acquiring a book item adjacency matrix through additional features of user nodes, acquiring user embedded representation through the distance adjacency matrix and the similarity adjacency matrix in combination with the book item adjacency matrix in each layer of encoder, and carrying out weighted summation and vector splicing to acquire user feature vector representation with book item preference;
s308, calculating inner products of user feature vector representations among users to obtain borrowing adaptability among users.
It should be noted that, taking the user and the book item as nodes, obtaining an undirected heterogram through graph representation, and setting an edge structure according to interaction of the user and the book item, wherein the undirected heterogram is defined as G= (V, E), V is a set of nodes, and E is an edge structure set; generating additional features of user nodes through the book sharing features and the book demand features, filtering neighbor nodes of users through similarity calculation according to the additional features, and constructing a user adjacency matrix through the neighbor nodes.
Splicing the undirected heterogram and the user subgraph to construct a user-book project diagram, wherein the user-book project diagram M is expressed asWherein G represents an undirected heterogram, and K represents a user subgraph. In a user-book project graph, dividing neighbor nodes of user nodes into user neighbor nodes and book project neighbor nodes according to different edge structures, adding user position description, and constructing a user distance adjacency matrix according to distance between users and a distance threshold according to self-setting acquisition distance threshold of the users for the convenience of book borrowing among the users; and generating a book item adjacency matrix through the book item neighbor nodes, respectively combining the distance adjacency matrix and the similarity adjacency matrix of the user with the book item adjacency matrix by utilizing a neighbor aggregation mechanism to acquire user embedded representations, carrying out weighted summation on the user embedded representations of the L-layer encoder, and acquiring final user feature vector representations by utilizing vector splicing.
Acquiring user feature vector representations of target users through a mutual borrowing recommendation model, acquiring book item preferences of the target users based on the user feature vector representations, screening other users meeting preset similarity standards according to the book item preferences, and acquiring other user sets with similar preferences; calculating an inner product between the user feature vector representation of the other user and the user feature vector representation of the target user in the set of other users, obtaining a borrowing fitness y between the users based on the inner product,wherein f u For target user feature vector representation, f i Representing for other user feature vectors; generating a borrowing priority according to the borrowing suitability, pushing the borrowing priority to a target user according to a preset method, and carrying out mutual borrowing of books by the user through online contact and in-plane exchange or express way;
according to the embodiment of the invention, a book inter-borrowing behavior database is established, specifically:
acquiring the performance of both sides of a user in the process of mutual borrowing of books, establishing a book mutual borrowing performance database, storing the performance into the book mutual borrowing performance database, and evaluating the integrity level of the user according to the performance of the user;
when the borrowing priority of the target user is obtained, marking the borrowing priority according to the honest grade, and recommending the optimal book inter-borrowing user according to the honest grade and the borrowing adaption degree;
when the loyalty grade evaluation of the user is smaller than a preset loyalty grade threshold, suspending the book mutual borrowing behavior of the user, acquiring the borrowing behavior of the user in a preset time period of a target library, and evaluating the user loyalty grade according to the borrowing behavior;
when the book borrowing behavior is greater than or equal to the integrity level threshold, the book borrowing behavior of the user is restored;
the performance behavior includes whether to borrow on time, whether to return to damage, and the like.
It should be noted that, when other used idle book resources borrow books for the library user, the user distance is defined as the same city range by default; when the target user receives the library and determines, the returning period in the library extends for a preset time, the target user is taken as a returning person, and the returning transaction of the library is performed by the target user.
FIG. 4 is a block diagram showing a book inter-borrowing recommendation system based on big data according to the present invention.
The second aspect of the present invention also provides a book mutual borrowing recommendation system 4 based on big data, the system comprising: the memory 41 and the processor 42, wherein the memory comprises a book inter-borrowing recommendation method program based on big data, and the book inter-borrowing recommendation method program based on big data realizes the following steps when being executed by the processor:
acquiring current idle book resource information of a user, extracting key word characteristics of the idle book resource information and generating book sharing characteristics of the user;
acquiring book borrowing records and historical reading behavior data of a user by using a big data method, integrating the book borrowing records and the historical reading behavior data to perform preprocessing, and extracting data characteristics to locate book demand characteristics of the current user;
using the graphic representation to represent the user and the book items to generate undirected heterograms, and generating additional features of user nodes through the book sharing features and the book demand features;
based on the graph convolution neural network, a mutual borrowing recommendation model is built, a graph structure is learned, user feature vector representations with book item preferences are learned according to the graph structure, preference information of a user is predicted, and mutual borrowing objects are recommended to the user according to the preference information.
The method comprises the steps of acquiring current idle book resource information of a user by utilizing code scanning input and book name keyword input, acquiring book borrowing records of the user in a target library, and extracting borrowed time of the book borrowed by the user according to the book borrowing records; acquiring average reading time length of a book borrowed by a user through a big data method, setting a preset time length threshold according to the average reading time length, feeding the borrowed book with the borrowed time length being greater than the preset time length threshold back to the user for selection, reading whether the user has read or is not interested in reading no longer, supplementing current idle book resource information according to the selection of the user, acquiring book profile data according to the current idle book resource information, and segmenting the book profile data to acquire corresponding word vectors; carrying out weighted average on the word vectors to construct sentence vector expression, obtaining the occurrence frequency and the distribution breadth of the word vectors in the book profile data, and screening the word vectors meeting preset standards as description keyword vectors; and extracting key word characteristics according to the book name key words and the description key word vectors, and acquiring book sharing characteristics of the user based on the key word characteristic set sentence vector expression.
According to the embodiment of the invention, the book borrowing record and the historical reading behavior data are integrated for preprocessing, and the book demand characteristics of the current user are positioned by extracting the data characteristics, specifically:
the method comprises the steps of obtaining borrowing records of a user in a target library through big data retrieval, obtaining historical reading behavior data according to interaction behaviors of the user in a reading website, and integrating the borrowing records and the historical reading behavior data with a time stamp;
preprocessing the acquired time sequence after data integration to generate a user sequence, selecting the user sequence with preset time t to acquire a short-term intention item set, embedding the short-term intention item set into a potential space to stack up to generate a matrix, importing the matrix into a convolution layer to perform convolution operation, and acquiring short-term interaction characteristics through maximum pooling;
acquiring a user sequence before a preset time t to acquire a long-term intention project set, acquiring long-term interaction characteristics of a user through a gating circulating unit layer, importing the long-term interaction characteristics and the short-term interaction characteristics into a self-attention layer, and acquiring self-attention weights;
and aggregating the long-term interaction characteristics and the short-term interaction characteristics according to the self-attention weight, and outputting book demand characteristics of the current user.
It should be noted that, the historical reading behavior data includes browsing, searching, borrowing, renewing, collecting, consulting and applying for demands, and the historical reading behavior data not only reflects the searching demands and searching habits of the user, but also maps the research trends and potential interesting contents of the user. Capturing short-term interaction characteristics and long-term interaction characteristics of a user through each continuous item in a user sequence set, acquiring short-term correlation in the user sequence by utilizing a convolution layer, and learning specific sequence characteristics through a convolution filter to output the short-term interaction characteristics; the gating loop layer captures the time dependence of the user sequence through the gating loop cell structure. The gating circulation unit can not clear the previous information along with the time, and can keep the related information and transmit the related information to the next unit, so that the problem of gradient disappearance is avoided. The self-Attention module is used for representing the influence of short-term interaction characteristics and long-term interaction characteristics on demand characteristics, different self-Attention weights are endowed for aggregation, book demand characteristics of a current user are obtained, and the Attention mechanism Attention (Q, K, V) in the self-Attention layer has a calculation formula as follows:
where Q represents a query matrix, K represents a key matrix, V represents a value matrix,representing the scale factor and T representing the matrix transpose.
According to the embodiment of the invention, a mutual borrowing recommendation model is constructed based on a graph convolution neural network, and the method specifically comprises the following steps:
constructing a mutual borrowing recommendation model based on a graph convolution neural network, constructing a user subgraph according to a user adjacency matrix, splicing the undirected heterograph and the user subgraph to construct a user-book project graph, and learning the user-book project graph by using the mutual borrowing recommendation model;
setting an encoder to respectively encode and express a user-book project diagram, adding user position description in the user-book project diagram, acquiring a user distance adjacency matrix based on distance information, and re-performing similarity calculation on a user in the user-book project diagram to acquire a user similarity adjacency matrix;
acquiring a book item adjacency matrix through additional features of user nodes, acquiring user embedded representation by combining the distance adjacency matrix and the similarity adjacency matrix with the book item adjacency matrix in each layer of encoder, and carrying out weighted summation and vector splicing to acquire user feature vector representation with book item preference;
and calculating the inner product of the user characteristic vector representation among the users to acquire the borrowing adaptability among the users.
It should be noted that, taking the user and the book item as nodes, obtaining an undirected heterogram through graph representation, and setting an edge structure according to interaction of the user and the book item, wherein the undirected heterogram is defined as G= (V, E), V is a set of nodes, and E is an edge structure set; generating additional features of user nodes through the book sharing features and the book demand features, filtering neighbor nodes of users through similarity calculation according to the additional features, and constructing a user adjacency matrix through the neighbor nodes.
Splicing the undirected heterogram and the user subgraph to construct a user-book project diagram, wherein the user-book project diagram M is expressed asWherein G represents an undirected heterogram, and K represents a user subgraph. In a user-book project graph, dividing neighbor nodes of user nodes into user neighbor nodes and book project neighbor nodes according to different edge structures, adding user position description, and constructing a user distance adjacency matrix according to distance between users and a distance threshold according to self-setting acquisition distance threshold of the users for the convenience of book borrowing among the users; and generating a book item adjacency matrix through the book item neighbor nodes, respectively combining the distance adjacency matrix and the similarity adjacency matrix of the user with the book item adjacency matrix by utilizing a neighbor aggregation mechanism to acquire user embedded representations, carrying out weighted summation on the user embedded representations of the L-layer encoder, and acquiring final user feature vector representations by utilizing vector splicing.
Acquiring user feature vector representations of target users through a mutual borrowing recommendation model, acquiring book item preferences of the target users based on the user feature vector representations, screening other users meeting preset similarity standards according to the book item preferences, and acquiring other user sets with similar preferences; calculating an inner product between the user feature vector representation of the other user and the user feature vector representation of the target user in the set of other users, obtaining a borrowing fitness y between the users based on the inner product,wherein f u For target user feature vector representation, f i Representing for other user feature vectors; and generating borrowing priority according to the borrowing suitability, and pushing the borrowing priority to a target user according to a preset method.
The third aspect of the present invention also provides a computer readable storage medium, wherein the computer readable storage medium includes a book inter-borrowing recommendation method program based on big data, and when the book inter-borrowing recommendation method program based on big data is executed by a processor, the steps of the book inter-borrowing recommendation method based on big data as described in any one of the above are implemented.
In the several embodiments provided in this application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above described device embodiments are only illustrative, e.g. the division of the units is only one logical function division, and there may be other divisions in practice, such as: multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. In addition, the various components shown or discussed may be coupled or directly coupled or communicatively coupled to each other via some interface, whether indirectly coupled or communicatively coupled to devices or units, whether electrically, mechanically, or otherwise.
The units described above as separate components may or may not be physically separate, and components shown as units may or may not be physical units; can be located in one place or distributed to a plurality of network units; some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present invention may be integrated in one processing unit, or each unit may be separately used as one unit, or two or more units may be integrated in one unit; the integrated units may be implemented in hardware or in hardware plus software functional units.
Those of ordinary skill in the art will appreciate that: all or part of the steps for implementing the above method embodiments may be implemented by hardware related to program instructions, and the foregoing program may be stored in a computer readable storage medium, where the program, when executed, performs steps including the above method embodiments; and the aforementioned storage medium includes: a mobile storage device, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk or an optical disk, or the like, which can store program codes.
Alternatively, the above-described integrated units of the present invention may be stored in a computer-readable storage medium if implemented in the form of software functional modules and sold or used as separate products. Based on such understanding, the technical solutions of the embodiments of the present invention may be embodied in essence or a part contributing to the prior art in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a removable storage device, ROM, RAM, magnetic or optical disk, or other medium capable of storing program code.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (4)
1. The book mutual borrowing recommendation method based on big data is characterized by comprising the following steps of:
acquiring current idle book resource information of a user, extracting key word characteristics of the idle book resource information and generating book sharing characteristics of the user;
acquiring book borrowing records and historical reading behavior data of a user by using a big data method, integrating the book borrowing records and the historical reading behavior data to perform preprocessing, and extracting data characteristics to locate book demand characteristics of the current user;
using the graphic representation to represent the user and the book items to generate undirected heterograms, and generating additional features of user nodes through the book sharing features and the book demand features;
constructing a mutual borrowing recommendation model based on a graph convolution neural network, learning a graph structure, learning user feature vector representations with book item preferences according to the graph structure, predicting preference information of a user, and recommending a mutual borrowing object for the user according to the preference information;
and using the graphic representation to represent the user and the book items to generate undirected heterograms, and generating additional characteristics of user nodes through the book sharing characteristics and the book demand characteristics, wherein the additional characteristics are as follows:
taking a user and a book item as nodes, obtaining an undirected heterogram through graph representation, and setting an edge structure according to interaction of the user and the book item, wherein the undirected heterogram is defined as G= (V, E), V is a set of nodes, and E is an edge structure set;
generating additional features of user nodes through the book sharing features and the book demand features, calculating and screening neighbor nodes of users through similarity according to the additional features, and constructing a user adjacency matrix through the neighbor nodes;
the construction of the mutual borrowing recommendation model based on the graph convolution neural network comprises the following specific steps:
constructing a mutual borrowing recommendation model based on a graph convolution neural network, constructing a user subgraph according to a user adjacency matrix, splicing the undirected heterograph and the user subgraph to construct a user-book project graph, and learning the user-book project graph by using the mutual borrowing recommendation model;
setting an encoder to respectively encode and express a user-book project diagram, adding user position description in the user-book project diagram, acquiring a user distance adjacency matrix based on distance information, and re-performing similarity calculation on a user in the user-book project diagram to acquire a user similarity adjacency matrix;
acquiring a book item adjacency matrix through additional features of user nodes, acquiring user embedded representation by combining the distance adjacency matrix and the similarity adjacency matrix with the book item adjacency matrix in each layer of encoder, and carrying out weighted summation and vector splicing to acquire user feature vector representation with book item preference;
calculating inner products represented by user feature vectors among users to obtain borrowing fitness among the users;
the inner product represented by the user feature vector among the users is calculated to obtain the borrowing adaptability among the users, and the borrowing adaptability is specifically as follows:
acquiring user feature vector representations of target users through a mutual borrowing recommendation model, acquiring book item preferences of the target users based on the user feature vector representations, screening other users meeting preset similarity standards according to the book item preferences, and acquiring other user sets with similar preferences;
calculating inner products between user characteristic vector representations of other users and user characteristic vector representations of target users in the other user sets, and acquiring borrowing adaptability between users according to the inner products;
and generating borrowing priority according to the borrowing suitability, and pushing the borrowing priority to a target user according to a preset method.
2. The book mutual borrowing recommendation method based on big data according to claim 1, wherein the method is characterized in that current idle book resource information of a user is obtained, keyword features of the idle book resource information are extracted to generate book sharing features of the user, and specifically comprises the following steps:
acquiring current idle book resource information of a user by utilizing code scanning input and book name keyword input, acquiring book borrowing records of the user in a target library, and extracting borrowed time of the book borrowed by the user according to the book borrowing records;
feeding the borrowed books with the borrowed time length being longer than a preset time length threshold back to a user for selection, supplementing current idle book resource information according to the selection of the user, acquiring book profile data according to the current idle book resource information, and performing word segmentation on the book profile data to acquire corresponding word vectors;
carrying out weighted average on the word vectors to construct sentence vector expression, obtaining the occurrence frequency and the distribution breadth of the word vectors in the book profile data, and screening the word vectors meeting preset standards as description keyword vectors;
and extracting key word characteristics according to the book name key words and the description key word vectors, and acquiring book sharing characteristics of the user based on the key word characteristic set sentence vector expression.
3. The book mutual borrowing recommendation method based on big data according to claim 1, wherein the book borrowing record and the historical reading behavior data are integrated for preprocessing, and the book demand features of the current user are located by extracting data features, specifically:
the method comprises the steps of obtaining borrowing records of a user in a target library through big data retrieval, obtaining historical reading behavior data according to interaction behaviors of the user in a reading website, and integrating the borrowing records and the historical reading behavior data with a time stamp;
preprocessing the acquired time sequence after data integration to generate a user sequence, selecting the user sequence with preset time t to acquire a short-term intention item set, embedding the short-term intention item set into a potential space to stack up to generate a matrix, importing the matrix into a convolution layer to perform convolution operation, and acquiring short-term interaction characteristics through maximum pooling;
acquiring a user sequence before a preset time t to acquire a long-term intention project set, acquiring long-term interaction characteristics of a user through a gating circulating unit layer, importing the long-term interaction characteristics and the short-term interaction characteristics into a self-attention layer, and acquiring self-attention weights;
and aggregating the long-term interaction characteristics and the short-term interaction characteristics according to the self-attention weight, and outputting book demand characteristics of the current user.
4. A book mutual borrowing recommendation system based on big data is characterized in that the system comprises: the book mutual borrowing recommendation method based on big data comprises a memory and a processor, wherein the memory comprises a book mutual borrowing recommendation method program based on big data, and the following steps are realized when the book mutual borrowing recommendation method program based on big data is executed by the processor:
acquiring current idle book resource information of a user, extracting key word characteristics of the idle book resource information and generating book sharing characteristics of the user;
acquiring book borrowing records and historical reading behavior data of a user by using a big data method, integrating the book borrowing records and the historical reading behavior data to perform preprocessing, and extracting data characteristics to locate book demand characteristics of the current user;
using the graphic representation to represent the user and the book items to generate undirected heterograms, and generating additional features of user nodes through the book sharing features and the book demand features;
constructing a mutual borrowing recommendation model based on a graph convolution neural network, learning a graph structure, learning user feature vector representations with book item preferences according to the graph structure, predicting preference information of a user, and recommending a mutual borrowing object for the user according to the preference information;
and using the graphic representation to represent the user and the book items to generate undirected heterograms, and generating additional characteristics of user nodes through the book sharing characteristics and the book demand characteristics, wherein the additional characteristics are as follows:
taking a user and a book item as nodes, obtaining an undirected heterogram through graph representation, and setting an edge structure according to interaction of the user and the book item, wherein the undirected heterogram is defined as G= (V, E), V is a set of nodes, and E is an edge structure set;
generating additional features of user nodes through the book sharing features and the book demand features, calculating and screening neighbor nodes of users through similarity according to the additional features, and constructing a user adjacency matrix through the neighbor nodes;
the construction of the mutual borrowing recommendation model based on the graph convolution neural network comprises the following specific steps:
constructing a mutual borrowing recommendation model based on a graph convolution neural network, constructing a user subgraph according to a user adjacency matrix, splicing the undirected heterograph and the user subgraph to construct a user-book project graph, and learning the user-book project graph by using the mutual borrowing recommendation model;
setting an encoder to respectively encode and express a user-book project diagram, adding user position description in the user-book project diagram, acquiring a user distance adjacency matrix based on distance information, and re-performing similarity calculation on a user in the user-book project diagram to acquire a user similarity adjacency matrix;
acquiring a book item adjacency matrix through additional features of user nodes, acquiring user embedded representation by combining the distance adjacency matrix and the similarity adjacency matrix with the book item adjacency matrix in each layer of encoder, and carrying out weighted summation and vector splicing to acquire user feature vector representation with book item preference;
calculating inner products represented by user feature vectors among users to obtain borrowing fitness among the users;
the inner product represented by the user feature vector among the users is calculated to obtain the borrowing adaptability among the users, and the borrowing adaptability is specifically as follows:
acquiring user feature vector representations of target users through a mutual borrowing recommendation model, acquiring book item preferences of the target users based on the user feature vector representations, screening other users meeting preset similarity standards according to the book item preferences, and acquiring other user sets with similar preferences;
calculating inner products between user characteristic vector representations of other users and user characteristic vector representations of target users in the other user sets, and acquiring borrowing adaptability between users according to the inner products;
and generating borrowing priority according to the borrowing suitability, and pushing the borrowing priority to a target user according to a preset method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311131241.5A CN117076780B (en) | 2023-09-04 | 2023-09-04 | Book mutual borrowing recommendation method and system based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311131241.5A CN117076780B (en) | 2023-09-04 | 2023-09-04 | Book mutual borrowing recommendation method and system based on big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117076780A CN117076780A (en) | 2023-11-17 |
| CN117076780B true CN117076780B (en) | 2024-03-26 |
Family
ID=88704137
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311131241.5A Active CN117076780B (en) | 2023-09-04 | 2023-09-04 | Book mutual borrowing recommendation method and system based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117076780B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108133032A (en) * | 2018-01-05 | 2018-06-08 | 北京思享家网络科技有限公司 | A kind of shared books method, system and medium based on geographical location |
| CN113159352A (en) * | 2021-04-21 | 2021-07-23 | 广州宏途教育网络科技有限公司 | Book recommendation interchange borrowing method for campus |
| CN114612184A (en) * | 2022-03-10 | 2022-06-10 | 广州博高信息科技有限公司 | Book drift sharing method and system, computer equipment and storage medium |
| CN115062732A (en) * | 2022-08-18 | 2022-09-16 | 睿驰(深圳)智能有限公司 | Resource sharing cooperation recommendation method and system based on big data user tag information |
| KR20230018677A (en) * | 2021-07-30 | 2023-02-07 | 김건욱 | System for recommending implementation of unborrowed book |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106934959A (en) * | 2017-03-30 | 2017-07-07 | 徐亦隽 | A kind of book self-help borrows system and method |
-
2023
- 2023-09-04 CN CN202311131241.5A patent/CN117076780B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108133032A (en) * | 2018-01-05 | 2018-06-08 | 北京思享家网络科技有限公司 | A kind of shared books method, system and medium based on geographical location |
| CN113159352A (en) * | 2021-04-21 | 2021-07-23 | 广州宏途教育网络科技有限公司 | Book recommendation interchange borrowing method for campus |
| KR20230018677A (en) * | 2021-07-30 | 2023-02-07 | 김건욱 | System for recommending implementation of unborrowed book |
| CN114612184A (en) * | 2022-03-10 | 2022-06-10 | 广州博高信息科技有限公司 | Book drift sharing method and system, computer equipment and storage medium |
| CN115062732A (en) * | 2022-08-18 | 2022-09-16 | 睿驰(深圳)智能有限公司 | Resource sharing cooperation recommendation method and system based on big data user tag information |
Non-Patent Citations (3)
| Title |
|---|
| "Design of personalized recommendation algorithm for mobile intelligent book management system based on cloud computing technology";Xiaolei Zhong;《2022 IEEE 2nd International Conference on Data Science and Computer Application (ICDSCA)》;20221030;748-756 * |
| "共享经济背景下家庭图书共享模式构建研究";康思本等;《图书馆研究》;20210131;18-25 * |
| 基于大数据和支持向量机分类法的图书馆中转站构建研究;杨志腾;孙萍;朱天怡;苏冠文;马俊隆;;价值工程;20200218(第05期);224-226 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117076780A (en) | 2023-11-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110866181B (en) | Resource recommendation method, device and storage medium | |
| CN103793465B (en) | Mass users behavior real-time analysis method and system based on cloud computing | |
| CN112052387B (en) | Content recommendation method, device and computer readable storage medium | |
| CN113538070B (en) | User life value cycle detection method and device and computer equipment | |
| CN117235586B (en) | Hotel customer portrait construction method, system, electronic equipment and storage medium | |
| CN112685635A (en) | Item recommendation method, device, server and storage medium based on classification label | |
| CN115718826A (en) | Method, system, device and medium for classifying target nodes in graph structure data | |
| CN115964568A (en) | Personalized recommendation method based on edge cache | |
| CN110795613A (en) | Commodity searching method, device and system and electronic equipment | |
| CN114223012A (en) | Push object determination method and device, terminal equipment and storage medium | |
| CN108647295B (en) | Image labeling method based on depth collaborative hash | |
| CN108090074A (en) | Book recommendation system and method | |
| CN115730125A (en) | Object identification method and device, computer equipment and storage medium | |
| CN117076780B (en) | Book mutual borrowing recommendation method and system based on big data | |
| CN113327154B (en) | E-commerce user message pushing method and system based on big data | |
| CN115186188A (en) | Product recommendation method, device and equipment based on behavior analysis and storage medium | |
| CN114528491A (en) | Information processing method, information processing device, computer equipment and storage medium | |
| CN113256024A (en) | User behavior prediction method fusing group behaviors | |
| CN111460300A (en) | Network content pushing method and device and storage medium | |
| CN112989020A (en) | Information processing method, apparatus and computer readable storage medium | |
| CN113065911A (en) | Recommendation information generation method and device, storage medium and electronic equipment | |
| CN115062230B (en) | Data processing method, apparatus, device, medium, and program product | |
| CN114298118B (en) | Data processing method based on deep learning, related equipment and storage medium | |
| CN117938951B (en) | Information pushing method, device, computer equipment and storage medium | |
| CN117786234B (en) | Multimode resource recommendation method based on two-stage comparison learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |