20 results about "Library collection" patented technology

The embodiment of the invention relates to the technical field of data processing, and particularly provides a smart library information retrieval method and system based on AI session interaction. According to the embodiment of the invention, an information retrieval session request initiated by a user through a smart library interaction interface is received; collecting session interaction data containing natural language query content and context association information; performing semantic structuring processing on the session interaction data to generate a session semantic feature set containing an entity relationship network and an intention classification vector; carrying out deep mining on the session semantic feature set by utilizing a debugged retrieval intention understanding model, identifying potential retrieval demands and demand priority ranking, and generating retrieval demand feature vectors; and on the basis of library collection information in the vector matching intelligent library resource database, generating an information retrieval result set containing resource association degree sorting, and feeding back the information retrieval result set to the interactive interface for visual display. According to the embodiment of the invention, the accuracy and comprehensiveness of information retrieval are improved, and more intelligent and efficient retrieval experience is provided for users.

The invention discloses a personalized recommendation method and system for library books, and relates to the field. Comprising the following steps: multi-source data acquisition and preprocessing: periodically acquiring multi-source heterogeneous data of a user u, performing deep processing on library collection book data, and constructing a high-dimensional feature vector bi for each book i; carrying out multi-level user interest modeling; predicting user-book preference; a diversity recommendation list is generated, and a personalized recommendation list is claimed according to the predicted preference score ru, i; and result recommendation: presenting the final recommendation list to the user through a library doorway website, a mobile application or a retrieval system interface. According to the method, the static interest portrait is combined with the dynamic interest sequence, and context perception fusion is carried out by utilizing an attention mechanism, so that the real interest of the user in the current environment can be more comprehensively and accurately described, and the accuracy of preference prediction is greatly improved.

The invention provides a museum collection cultural relic preventive protection risk assessment method and system based on a knowledge graph, and the method comprises the steps: collecting to-be-assessed museum collection data which covers all aspects of museum collection cultural relics; extracting various entities and attributes thereof from library collection data, and constructing a knowledge graph by taking the entities as nodes and taking a relation mode between the entities as an association edge; extracting dynamic time sequence features and static spatial data features based on the constructed knowledge graph, and performing splicing to form a comprehensive feature set; based on the trained risk assessment model, the corrosion degradation rate of the cultural relics is output, and historical relic risk assessment is carried out; the risk assessment model is pre-established and trained, takes at least one part of the comprehensive feature set as input, and takes the corrosion degradation rate of the cultural relic as output. According to the method, by constructing the comprehensive knowledge graph, multi-source heterogeneous data are effectively integrated, and the accuracy and efficiency of library collection historical relic risk assessment are greatly improved.

The invention provides a library management system based on big data, and relates to the technical field of library management, and the system comprises a data collection and preprocessing module which is used for obtaining multi-dimensional information, and the multi-dimensional information comprises library collection data, user borrowing behavior data and book resource data; and the data acquisition and preprocessing module performs standardization processing on the multi-dimensional information to generate a structured data set. According to the library management system based on the big data, accurate prediction and efficient management of library resources are realized. The system obtains multi-dimensional information through a data acquisition and preprocessing module, and adopts a cross-modal deep fusion technology of deep learning to fuse text features, user behavior features and book resource features to generate unified high-dimensional semantic representation. Through the method, the borrowing peak time, the hot book type and the resource access fluctuation can be accurately predicted, so that the accuracy and the service efficiency of library resource configuration are effectively improved.

The invention provides a library table parallel retrieval method and system for large language model SQL generation and a medium, and the method comprises the steps: obtaining SQL generation demand information provided by a user, and extracting a core keyword in the SQL generation demand information; inputting the core keyword into a text embedding model to generate a query vector; similarity calculation is conducted in a library knowledge vector table and a table knowledge vector table according to the query vector to obtain an effective library set and an effective table set, and the query vector is the same as vectors in the library knowledge vector table and vectors in the table knowledge vector table in dimension; constructing candidate pairs according to the effective library set and the effective table set, calculating a weight score of each candidate pair, sorting the candidate pairs according to the weight scores, and outputting a retrieval result; according to the method, the accuracy of generating the SQL statement through large language model retrieval is effectively improved.

The invention discloses an RAG-based information retrieval method and device, a storage medium and equipment, and belongs to the technical field of deep learning. Obtaining a retrieval statement input by a user; the retrieval statement is processed through a pre-trained decision model, the token proportion of each database in a database set is obtained, and the database set is a set of all databases connected with the question and answer model when information retrieval is conducted based on RAG; for each database, generating a token result of the database according to the retrieval statement, the token proportion corresponding to the database and the total number of tokens capable of being input by the question and answer model; splicing token results of all the databases into cue words of a question and answer model; and processing the cue word by utilizing the question and answer model to obtain a retrieval result of the retrieval statement. According to the method and the device, token results provided by different databases can be reasonably allocated, and irrelevant tokens are prevented from being introduced, so that the RAG can obtain more accurate answers.

According to the intelligent library collection conveying system based on the FPGA, the FPGA serves as a main control core, a mechanical arm executing mechanism is integrated on the basis of a moving trolley integrating intelligent tracking and intelligent recognition, books are intelligently recognized in combination with image recognition, and book grabbing, taking and conveying are achieved through omnidirectional movement of the moving trolley and cooperative control of a trolley arm; the method is suitable for automatic book distribution in libraries, archives and other scenes. The main control FPGA module is electrically connected with the multi-path infrared tracking module, the mechanical arm executing mechanism, the image recognition module and a driving module of the moving trolley, drives the multi-path infrared tracking module, the mechanical arm executing mechanism, the image recognition module and the driving module of the moving trolley to work and is used for controlling operation of the intelligent book conveying system, and when the image recognition module recognizes books, the moving trolley and the mechanical arm executing mechanism are controlled to move in linkage. The linkage of the movement control of the moving trolley and the grabbing action of the mechanical arm executing mechanism is realized.

The invention discloses a library collection book supplement demand analysis method based on big data, and belongs to the technical field of big data, and the method comprises the steps: obtaining book names retrieved by all readers in a library; obtaining a support relationship network among existing library collection books in the library; on the basis of the book names retrieved by all readers, pre-supplemented book names are obtained; constructing a high-priority to-be-supplemented support book library based on a support relationship network between the names of the to-be-supplemented books and the library collection books; acquiring the demand priority of the library reader for the books in the high-priority to-be-supplemented support book library, so as to obtain the names of the to-be-supplemented support books; and supplementing the library collection books based on the names of the books to be supplemented and the names of the books to be supplemented and supported. The problems that in the prior art, the demand analysis precision is insufficient in the library collection book supplement demand, and the auxiliary books cannot be predicted are solved, so that the analysis precision of the library collection book supplement demand is fully improved, and the high-precision prediction of the auxiliary books can be improved.

The invention relates to the technical field of data processing, in particular to a cultural data classification method and device based on digital processing and electronic equipment, and the method comprises the steps: extracting the center point coordinates of a target image music score symbol, and extracting the topological features of the music score symbol, the topological features including an adjacent feature and a spacing feature; determining candidate images according to the topological characteristics of the music score symbols; constructing a carbonization influence factor according to the number of target storage years, and updating a determination process of the candidate image based on the carbonization influence factor; constructing a five-degree phase generation index of the candidate image, and extracting a jump index of the candidate image according to a construction result of the five-degree phase generation index; and constructing feature parameters of the candidate images according to the adjacency features of the music score symbols of the candidate images and the jump indexes of the candidate images, and classifying the candidate images according to the feature parameters. According to the invention, the digitalized classification efficiency and classification accuracy of the collected ancient books are effectively improved.

The invention discloses a non-destructive comprehensive intelligent identification system for ancient ceramics. Comprising a library collection standard porcelain reference system and an age imitation and modern porcelain reference system, wherein the library collection standard porcelain reference system comprises a library collection standard database which is established according to a result by adopting a physical parameter measurement tool and a component analysis tool to carry out parameter measurement and component analysis on genuine library collection; according to the age imitation and modern porcelain reference system, parameter measurement and component analysis are carried out on age imitation and modern porcelain by adopting a physical parameter measurement tool and a component analysis tool, an imitation and modern porcelain database is established according to a result, and a counterfeit feature data model is established through the imitation and modern porcelain database; the invention relates to the technical field of ceramic identification, a physical parameter measurement tool and a component analysis tool can help people to obtain detailed information of the appearance and even the internal structure of porcelain and obtain detection data of various counterfeit products, and during identification, real and fake modes need to be identified from the front and back aspects at the same time.

According to the intelligent library collection conveying system based on the FPGA, the FPGA serves as a main control core, a mechanical arm executing mechanism is integrated on the basis of a moving trolley integrating intelligent tracking and intelligent recognition, books are intelligently recognized in combination with image recognition, and book grabbing, taking and conveying are achieved through omnidirectional movement of the moving trolley and cooperative control of a trolley arm; the method is suitable for automatic book distribution in libraries, archives and other scenes. The main control FPGA module is electrically connected with the multi-path infrared tracking module, the mechanical arm executing mechanism, the image recognition module and a driving module of the moving trolley, drives the multi-path infrared tracking module, the mechanical arm executing mechanism, the image recognition module and the driving module of the moving trolley to work and is used for controlling operation of the intelligent book conveying system, and when the image recognition module recognizes books, the moving trolley and the mechanical arm executing mechanism are controlled to move in linkage. The linkage of the movement control of the moving trolley and the grabbing action of the mechanical arm executing mechanism is realized.

The present invention belongs to the technical field of document scanning and discloses an X-ray-based three-dimensional scanning device for ancient books and documents, comprising: an X-ray source module, a detection module, a main control module, a document positioning and support module, an ink component identification module, an imaging and three-dimensional reconstruction module, and a data storage and processing module; the present invention avoids damage to document materials caused by physical contact, greatly reducing the wear and tear of ancient books during the scanning process; through multi-level scanning, the internal text structure of the ancient book is reconstructed to achieve page-by-page display, and high-definition text images can be obtained even when the document is in a closed state; the absorption difference of X-rays on different component materials is utilized to achieve effective distinction between ink and paper, making the content of the ancient book clearly visible and adapting to different historical ink compositions; non-contact batch scanning of a large number of documents is achieved, meeting the digital protection needs of library collections of ancient books and providing researchers with a more convenient access method.

This invention discloses a distributed book borrowing and returning system and method based on cloud-edge collaboration, comprising: receiving a reader's borrowing request; providing the reader with the location of the target book when the library detects that the target book is in the library; the reader borrowing the book based on its location; returning the book in response to the reader's return request; periodically inputting information on fluctuations in book demand into a book demand prediction model to determine the book demand; scheduling books based on the library's collection and demand; and providing immediate fulfillment and proactive guidance services when the target book is not available in the library. This invention can improve the overall efficiency of the library collection and the reader service experience, and optimize resource allocation.

This invention belongs to the technical field of library management systems, specifically relating to a large-scale model-driven intelligent library management system. It includes: a semantic analysis module for inputting book resource data into a large language model for analysis and extraction of key semantic knowledge; a clustering module for clustering semantic vectors transformed from the large language model using a quadtree segmentation algorithm; an RDF data generation module for generating triple RDF data and then creating URI links between books using a similarity matching method; and an RDF editor module for providing a user interface. The descriptive labels of the quadtree-segmented vector clusters are mapped to the attributes of the triple RDF data; URI links are generated based on the similarity matching of vector clusters of different books using the BIBFRAME system; this improves the knowledge relevance of the entire library collection.

The invention provides a dynamic library management method and system based on artificial intelligence, and relates to the technical field of library management, and the method comprises the steps: obtaining the borrowing record of library books, extracting the borrowing time and frequency, comparing the number of inventory books, calculating the inventory dynamic change rate, analyzing the change of library collection resources, and generating the library collection circulation state. According to the invention, through collecting and analyzing the time data of book borrowing and returning, real-time monitoring and accurate management of dynamic changes of library resources are realized, through a data-driven management mode, the dynamic response capability of library services is improved, and through optimizing the book storage position and adjusting the borrowing deadline, the book borrowing and returning efficiency is improved. According to the method, resource utilization maximization is achieved, borrowing requirements are predicted and adjusted through the artificial intelligence technology, adaptability to reader behaviors and library resource allocation efficiency are remarkably improved, library collection layout can be accurately adjusted through tracking analysis of book circulation paths, and it is ensured that books in a high-requirement area are sufficiently supplied.

The application discloses a dependency relationship establishment method and device and a storage medium. The method comprises the following steps: in response to a compilation instruction for at least two service plug-ins, extracting rule information for plug-in preprocessing from a preset detection rule; when a dependency library in a dependency library set has a version conflict, taking a dependency library with the highest version as a target library based on the rule information, and compiling resource source codes of the at least two service plug-ins by using the target library; when resource source code compilation of a remaining service plug-in fails, copying common resources of a dependency library corresponding to the remaining service plug-in and the target library, and establishing a dependency relationship between a storage path of the common resources and non-common resources in the dependency library corresponding to the remaining service plug-in. The application automatically establishes the dependency relationship by using the rule information, thereby improving the efficiency of analyzing plug-in dependencies. With the established dependency relationship, the convenience of locating problems when performing abnormality troubleshooting is improved, and the development efficiency is improved.

The application discloses a kind of based on information fusion's parking stall detection method and system, detection method includes the position information of longitudinal and transverse distance of parking stall when seeking library collection;Collect and detect the image information of wheel hub, vehicle body and license plate of parking stall obstacle vehicle;Image information is detected using deep learning algorithm, the characteristic information of wheel hub, vehicle body and license plate of obstacle vehicle is determined;According to position information and characteristic information, the space position information of obstacle vehicle's vehicle body posture parameter and the parking stall to be parked is calculated;Parking stall space model is constructed and is planned out parking path, then signal is output to ESP chassis control system and EPS system, control vehicle tracks planned path and enters parking stall;The application improves prediction accuracy, and the generalization of model is good and robustness is stronger;While prediction speed is greatly accelerated, solves the problem that CenterNet original detector overflows in memory on commonly used embedded device, cannot train, enhances practicality.

Embodiments of the present disclosure a method for dynamically generating a generative operating system based on a hardware and software environment feature, the method is configured to generate the generative operating system, and an architecture of the generative operating system includes: a hardware architecture adaptation layer and a meta kernel layer and an external library of the operating system. The method comprises: systematically and comprehensively perceiving an environment state and application requirements by an environment sensor and generating a configuration file; parsing the corresponding micro-libraries through the configuration files and obtaining a corresponding micro-libraries collection through a micro-library marketplace; assembling configured micro-libraries into the generative operating system by a system organizer; and dynamically adjusting and optimizing the generative operating system in real time by a run optimizer based on changes in a runtime environment and operational requirements.

Techniques for intelligently prompting an LLM to fix code are disclosed. A corpus of release notes for a set of libraries is accessed. The release notes include information describing deprecated or removed APIs associated with the libraries. The corpus is stored in a vector database. A code snippet is accessed. This snippet is identified as potentially using a deprecated API. The code snippet is used to identify a set of release notes from the vector database. These release notes are determined to satisfy a threshold level of similarity with the code snippet. An LLM prompt is built and is fed to the LLM. The LLM prompt instructs the LLM to update the code snippet based on the identified set of release notes. Output of the LLM is displayed. This output includes a proposed rewritten version of the code snippet.

The application provides a library table parallel retrieval method, system and medium for large language model SQL generation, the method comprising: obtaining SQL generation requirement information proposed by a user, and extracting core keywords in the SQL generation requirement information; inputting the core keywords into a text embedding model to generate a query vector; performing similarity calculation in a library knowledge vector table and a table knowledge vector table according to the query vector to obtain an effective library set and an effective table set, wherein the dimensions of the query vector, the vectors in the library knowledge vector table and the vectors in the table knowledge vector table are all the same; constructing candidate pairs according to the effective library set and the effective table set, calculating the weight score of each candidate pair, and then outputting a retrieval result after sorting the candidate pairs according to the weight score; and the application effectively improves the accuracy of large language model retrieval and SQL statement generation.