43results about How to "Easy retrieval" patented technology

The invention discloses an operation site safety management and control method and system based on a multi-modal large model, and the method comprises the steps: collecting site violation data, and carrying out the preprocessing, and obtaining multi-modal data; constructing a multi-modal large model; carrying out training and optimization on the multi-modal large model by utilizing the preprocessed multi-modal data; and using the optimized multi-modal large model to realize operation site safety management and control. According to the invention, a visible light image, a safety violation knowledge base and other modal data are fused, and a comprehensive safety monitoring and management platform is constructed by using a large model and a big data processing technology, so that real-time monitoring, early warning and intelligent response of potential safety hazards of a complex operation site are realized, and the safety management level of the operation site is improved.

The invention discloses a multi-source information fusion query method based on RAG Fusion, which comprises the following steps: receiving original query data input by a user, and carrying out preprocessing and expansion processing on the original query data to obtain an expanded query set; retrieving each extended query in the extended query set in a preset vector library by using a plurality of retrieval strategies to obtain a document set corresponding to each retrieval strategy, and reordering all documents in the document set to obtain a document list; dynamic context construction is carried out on documents in the document list, and a context document set with weights is generated; performing sentence-level segmentation on each document in the context document set, extracting fragments related to the original query data from segmented sentences by utilizing a pre-trained question and answer model, and performing duplicate removal and combination on the extracted fragments to generate an information unit set; and inputting the original query data, the context document set and the information unit set into a pre-trained large-scale language model to generate a target answer.

The invention discloses a large-scale embedded index data desensitization method suitable for a forestry database, and belongs to the technical field of data desensitization, and the method comprises the steps: obtaining a forestry data description document D1, splitting the document D1 into sentences, and processing the sentences into embedded vectors; a user customizes a query sentence containing the desensitization word, and a query vector is generated; performing hash coding on the embedded vector and the query vector to obtain a corresponding hash function cluster; searching data to be desensitized through cosine similarity calculation; and inputting to-be-desensitized data and the selected desensitization method into the large language model to generate desensitized data, and replacing the corresponding sentences in the D1 to obtain a desensitized forestry data description document D2. According to the method, the input data is converted into the high-dimensional vector through the embedding layer, so that the semantic features of the data can be effectively expressed in the vector space, the retrieval speed and accuracy of large-scale data are improved by means of the locality sensitive hashing technology, the calculation overhead is reduced, and a technical guarantee is provided for real-time desensitization processing of a large-scale data set.

The invention belongs to the technical field of natural language processing, and discloses a natural language processing platform and method based on deep learning, and the platform comprises a data collection module which is used for collecting font data, text data, image data and environment data; the data processing module is used for preprocessing the collected font data, text data, image data and environment data and obtaining a comprehensive feature data set; the character recognition module is used for predicting character recognition results through a character recognition model and judging whether the character recognition results are consistent or not; the semantic recognition module is used for generating corresponding character paraphrases through a semantic recognition model; collecting character recognition results and corresponding character paraphrases through the software terminal; the voice output module is used for converting the character recognition result and the paraphrases of the corresponding characters into natural voices and outputting the natural voices by adopting a TTS technology; through the implementation of the technology, students can be helped to better learn and understand.

The invention relates to an insurance industry-oriented customer full-life-cycle event data modeling method and system. The method comprises the steps of collecting customer full-life-cycle event data through a customer contact channel and a business processing link of an insurance industry; preprocessing the collected data, and converting the data into structured data; key elements are extracted from the preprocessed structured detail data; performing hierarchical and classified storage on the structured data based on a hierarchical and classified event data model, and respectively establishing a trunk index model and an expansion detail model in combination with the key elements; generating a unique event ID (Identity) for each event from the key elements; and performing data query optimization and providing data service based on the trunk index model, the expansion detail model and the event ID. Compared with the prior art, the method has the advantages that massive customer event data can be efficiently integrated and analyzed, the data processing and query capability is improved, more efficient and flexible data services are provided, and decision support is provided for the insurance industry.

The human cost data storage method for the power grid engineering system comprises the steps that human cost data related to cost data in the power grid engineering system are acquired, and the human cost data comprise work type data, man-hour data and salary data of constructors; determining a DBSCAN clustering radius range based on the average man-hour salary of each type of work; based on a DBSCAN clustering radius range, performing clustering analysis in a man-hour-salary two-dimensional plane formed by the human cost data to obtain a preliminary classification result; when it is determined that the isolated data points exist based on the preliminary classification result, secondary classification is carried out on the isolated data points based on the distance between the isolated data points and the known clustering cluster, and a final classification result is obtained; and realizing partition storage of the cost data based on a final classification result. By implementing the method, optimized storage of the cost data of the power grid engineering system can be realized.

The invention provides a multi-source software-defined science and education resource recommendation method and system based on a knowledge graph, and the method comprises the steps: constructing an artificial intelligence knowledge graph based on the relation between artificial intelligence knowledge entities; obtaining science and education resource data of a science and education institution, extracting knowledge entities in the science and education resource data, and connecting the knowledge entities to the artificial intelligence knowledge graph to obtain a science and education resource knowledge graph; extracting local resource features / user features based on a graph neural network to obtain local resource features and user features of each client; constructing a local knowledge graph based on the science and education resource knowledge graph, and performing model training by taking the local resource features of each client as initial features until the model converges; and inputting the users and the resources into the graph neural network for prediction, and recommending the resources with high scores to each user. According to the scheme, the relationship between knowledge entities in the field of artificial intelligence is effectively extracted, a structured knowledge graph is formed, and the recommendation precision is improved.

The present application discloses a method for constructing a knowledge graph based on a large language model and a vector library, which relates to the field of knowledge graph technology, including: using a pre-trained RoBERTa model to perform word segmentation, part-of-speech tagging and semantic analysis on the collected natural language text to obtain a semantic representation vector of the natural language text; obtaining the association between the semantic representation vector and the corresponding named entity through the Seq2Seq framework and the attention mechanism, and obtaining the named entity through the obtained association; constructing a knowledge graph; fusing the constructed knowledge graph with a relational database of a preset domain to generate a fused knowledge graph; using dependency syntactic analysis to obtain the named entity and the corresponding attribute, obtaining the entity relationship of the named entity, and constructing a structured triple of the named entity, attribute and entity relationship of the fused knowledge graph. In view of the difficulty of entity linking in the prior art, the present application converts the knowledge graph into a low-dimensional dense vector representation, and constructs an inverted index library, etc., to reduce the difficulty.

The invention discloses a data storage method and system for a storage chip, and the method comprises the steps: selecting reference data and sample data from target storage data, and carrying out the analysis, and obtaining a data similarity value; the data similarity value is compared with a data similarity threshold value, and if the data similarity value is larger than or equal to the data similarity threshold value, the reference data and the sample data are classified into the same classification data set; obtaining a residual storage space, judging whether the residual storage space is smaller than a residual storage space threshold value or not, and if yes, analyzing data in the classified data set to obtain a data set key value; identifying the non-key data group to generate a compression optimization signal; based on the compression optimization signal, judging whether the residual storage space of the system is smaller than a residual storage space threshold after compression processing, and if yes, analyzing the non-high access frequency data to obtain a compression priority value; and performing optimization compression on the non-high access frequency data according to the compression priority value.

The invention relates to the technical field of information and data processing, and discloses an intelligent recommendation method and system for a bone fracture rehabilitation training plan based on traditional Chinese medicine, and the method comprises the steps: dividing image information according to labels, and obtaining skeleton data sets with different functions; taking the first position information and the second position information as a special information set; deleting the special data set from the skeleton data set to obtain a target data set; obtaining position information of the bones to be subjected to bone fracture rehabilitation training from the target data set, and selecting matched contents such as shadowboxing, eight-section brocade or five-poultry play from a pre-stored traditional Chinese medicine rehabilitation training database; deleting related actions of the second position information from an action data set corresponding to the content; guiding the trainee to perform bone fracture rehabilitation training through the visual terminal; the system comprises a first data set construction module, a second data set construction module and a target action output module. According to the bone fracture rehabilitation training system and method, biological feature recognition and personalized, precise and systematic bone fracture rehabilitation training are achieved.

The invention relates to the technical field of archive management, and discloses an archive management method and system.The method comprises the steps that archive basic data are obtained, and the archive basic data comprise archive forming time, archive content and archive confidentiality deadline; performing classification score calculation according to the archive basic data to obtain archive classification scores; based on the archive classification score and a preset classification scoring standard, obtaining an archive archiving category; calculating an importance rating factor according to the archive basic data and the archive archiving category; performing importance score calculation based on the importance rating factor to obtain an archive importance score; and according to the archive importance score and a preset storage period standard, selecting a storage period for archiving. According to the method, the file storage efficiency can be improved, and the retrieval speed is increased.

The invention relates to the technical field of image retrieval, and discloses a Hash image retrieval method and system based on improved triple loss, and the method comprises the following steps: inputting a query set image into a trained Hash network model based on improved triple loss, and obtaining a Hash code of the query set image; comparing the difference degree between the Hash code of the query set image and each Hash code in the Hash code library; sorting the difference degrees, and outputting a retrieval result according to the sorting; and calculating the average accuracy of the image query set according to the retrieval result. According to the invention, the retrieval precision can be improved.

The invention relates to the technical field of data processing, in particular to an order data processing method and system based on a block chain. The method comprises the following steps: acquiring a multi-source order data set; performing transaction address analysis on the multi-source order data set to generate order transaction address data; order processing server construction is carried out on the multi-source order data through the order transaction address data, and order processing distribution server data is generated; order processing access frequency analysis is carried out on the multi-source order data set through order processing distribution server data, and order processing access frequency data is generated; and performing order screening on the multi-source order data set based on the order processing access frequency data to generate high-frequency order access data and low-frequency order access data. According to the invention, through optimization of multi-source data integration, order access frequency analysis, transaction synchronization and channel optimization, the reliability and efficiency of order transaction are improved.

The utility model relates to the technical field of goods taking platforms, and discloses an internet-based goods taking platform device which comprises a goods taking platform, a controller is fixedly installed on the side wall of the goods taking platform, a scanning platform is fixedly installed on the top of one side of the controller, and a rapid goods taking mechanism is installed in the goods taking platform. A large number of goods needing to be stored are labeled, the labels are aligned and input into the scanner, the content of the labels is input into the controller, the goods with the scanned labels are placed in the corresponding placing boxes, and the content of the labels can be uploaded into the cloud storage. According to the intelligent goods taking device, a label is arranged on a mobile phone of a person who needs to take goods, a signal of the label is sent to the mobile phone of the person who needs to take goods in cooperation with a communication module, meanwhile, after the person who takes goods arrives at a goods taking table, the person holds the mobile phone to align the received label with a scanning table, and the controllers receiving the signal turn on the corresponding illuminating lamps, so that the person who takes goods can take away the goods in the placing box rapidly; and the Internet goods taking efficiency is greatly improved.

The invention provides a steel mill production data management method and system, and relates to the technical field of data management. The system comprises a data receiving module, a data preprocessing module, a data analysis module, a data storage module, a data transmission management module and a model training module. According to the method, possible problems and trends in the production process are predicted, so that a decision maker can make a more accurate decision based on a data analysis result, guarantee is provided for the safety and integrity of data, the requirements of steel mill production management and decision are met, the transmission requirements of steel mill production data in different scenes are met, and the production efficiency is improved. And the reliability of data transmission is improved.

The invention provides a photovoltaic measurement and control system, and relates to the technical field of photovoltaic measurement and control, and the system comprises a signal collection unit which is used for collecting sensing data generated by a sensor; the main control unit is used for controlling the whole test unit and summarizing the acquired sensing data into a programmable controller for display and storage; and the upper computer software unit is used for analyzing and processing the acquired data and outputting an operation control command to the lower computer. Sensing data stored in the main control unit are analyzed and processed through the upper computer, monitoring of the state of each module in the photovoltaic system is achieved, and the operation efficiency of the photovoltaic system is improved.

The present invention discloses a keyword search method and system for text-containing pictures. Through the three stages of text recognition, text search, and target word positioning, the keywords required by users and the specific page numbers required by users can be easily retrieved on non-editable PDF documents and pictures, solving the problem that most users cannot use keywords to search and retrieve specific pages and specific contents in a large number of pictures and non-editable text PDF documents, and can easily meet the needs of outdoor learning process to locate the specific pages required by users through keywords.

The utility model discloses an elevator car flat layer gap covering structure, including the box body (1) of front side opening, be equipped with the moving plate (2) in the box body (1), the rear side of moving plate (2) is equipped with a plurality of mutual parallel swing plate (3), and one end of swing plate (3) is hinged with moving plate (2), and the other end of swing plate (3) is hinged with box body (1), and the rear side of moving plate (2) is equipped with the drive plate (4) parallel with swing plate (3), and one end of drive plate (4) is hinged with moving plate (2), and the other end of drive plate (4) is equipped with the drive shaft (5) that projects box body (1) downward, and the bottom of box body (1) is equipped with the motor (6) of connecting drive shaft (5). The utility model has can avoid the advantage that passenger falls off article and enters the hoistway.

The application discloses an automatic workflow execution method for a dynamic model set and a real-time inference service, comprising: acquiring natural language task instructions and multi-modal file inputs, and generating a bound pending directed acyclic graph workflow; for each task node to be executed, performing node-level model selection under the current environment state, generating a node binding plan containing a primary binding and a backup binding through multi-dimensional comprehensive scoring; triggering real-time inference service calls to execute the workflow according to data dependency relationships; during the execution process, in response to detecting that a task node fails to execute, preferentially performing node-level recovery, performing graph-level repair when the node-level recovery fails, and recovering the whole execution of the workflow by locally re-planning and re-executing the affected subgraph. Through decoupling of planning and binding, execution perception model selection and a hierarchical recovery mechanism, the application improves the end-to-end task execution success rate and process stability in a dynamic environment.

The invention relates to the technical field of information, in particular to a log storage system construction method and system based on multidimensional logic relation interaction.The method comprises the steps that a log storage instruction is received, an original log set is collected based on the log storage instruction, analysis operation is executed on the original log set, and a multidimensional data set is obtained; calculating a user weight according to the multi-dimensional data and the priority user sequence, calculating a time weight according to the transaction time and the priority time sequence in the multi-dimensional data, obtaining a headquarters position, calculating a position weight according to an initiation position and the headquarters position in the multi-dimensional data, and obtaining a headquarters position; and calculating the log priority according to the user weight, the time weight, the position weight and the transaction amount value, performing classification compression operation on the backlog set to obtain a log compression package, and constructing a log graph database by using the common log set to complete the construction of the log storage system. According to the method, the log storage space can be saved, and the log retrieval efficiency is improved.

The invention relates to a method and a system for automatically generating table printing images and texts of a sports skateboard, and relates to the technical field of skateboard table printing, and the system comprises an image management module which is used for establishing an image database and storing a plurality of template images; the image-text generation module is used for generating an image to be transferred based on the user image-text demand and the image database; the parameter determining module is used for determining thermal transfer printing parameters, at least including thermal transfer printing temperature, thermal transfer printing pressure, thermal transfer printing time and thermal transfer printing speed, based on the related information of the sports skateboard and the image features of the image to be transferred; the image-text transfer printing module is used for controlling thermal transfer printing equipment to print the image to be transferred on the surface of the sliding plate based on the thermal transfer printing parameters of the image to be transferred; and the quality detection module is used for carrying out quality evaluation based on the printed image on the surface of the skateboard and the image features of the image to be transferred, and has the advantage of realizing personalized and automatic design and printing of the surface printing image-text of the sports skateboard.

The utility model discloses a book cover tying opening structure, belongs to the technical field of book cover tying openings, and aims to solve the problems that multiple pages cannot be recorded through book cover tying openings, and the book cover structure is not high in strength and easy to damage. The book cover tying opening structure comprises a book body and a book shell, the book shell is arranged on the outer side of the book body, a book spine is arranged on one side of the book body, and a tying opening body is arranged on one side of the book shell; through the arrangement of the tying opening body, the front cover and the back cover can be prevented from being curled, the book body can be decorated, the aesthetic feeling of the book body is further improved, the book has the information supplementing and displaying function, author information, book information and the like can be pasted, a reader can be helped to read better, and the reading efficiency is improved. And the reading position of the user in the book can be quickly found.

The invention relates to the field of hearing aids, in particular to a hearing aid positioning and communication system and method.The hearing aid positioning and communication system comprises a processing module, a connecting module, a positioning module, a voice receiving module, a voice playing module and a navigation module, and the processing module comprises a processor module, a key module, a power module and a transmission module; the processor module is used for collecting and processing data in the system and controlling the operation of the hearing aid, and the hearing aid is bound and connected with a smart phone through the Bluetooth connection module, so that the hearing aid can play music, voice and telephone calls in the mobile phone; when the hearing aid is separated from the connection range of the Bluetooth connection module of the mobile phone, it is judged that the hearing aid is lost, at the moment, mutual searching between a user and a picker is facilitated through mutual positioning and navigation between the mobile phone and the hearing aid, and when positioning cannot be achieved, the hearing aid can play user information to facilitate offline searching; therefore, the hearing aid can be found conveniently.

The invention relates to a surround-view video processing method and device, computer equipment, a computer readable storage medium and a computer program product, and relates to the technical field of computer vision. The method comprises the following steps: acquiring an all-round video file obtained by splicing a plurality of original video files; the plurality of original video files are acquired in real time by a plurality of cameras at preset angles and positions; according to at least one preset target event identification condition, identifying the all-round video file in parallel, and according to an identification result corresponding to each target event identification condition, performing video extraction on the all-round video file to obtain a target video clip corresponding to each target event identification condition; classifying the target video clip by adopting a pre-trained classification model to obtain a classification result; and according to the abnormal event included in the classification result and the target video clip corresponding to the abnormal event, constructing an association table and storing the association table. By adopting the method, the video backtracking efficiency can be improved.

The present application relates to a circuit and method, chip and electronic device for storing point clouds, wherein the circuit for storing point clouds includes: a cache module, configured to cache point cloud data, the point cloud data including multiple point cloud frames; a recognition result acquisition module, coupled with the cache module, configured to obtain recognition results for multiple point cloud frames, and obtain recognition results for target objects; an assembly module, coupled with the recognition result acquisition module, configured to determine the association relationship between the target object and multiple point cloud frames in response to the recognition result characterizing the existence of the target object; an output module, coupled with the recognition result acquisition module and the assembly module, respectively, configured to output the recognition result and the multiple point cloud frames corresponding to the recognition result in association based on the association relationship, so as to write into a non-volatile memory. The present application can realize the targeted storage of valid data and improve the convenience of data retrieval.

The application provides an automatic data governance method and system based on a multi-modal large model, comprising: collecting multi-source heterogeneous industrial data and performing standardization processing to form standardized multivariate time series data; constructing a process knowledge base, performing semantic embedding coding on process knowledge text, and storing; constructing and fine-tuning a KTSF multi-modal large model, fusing process knowledge semantics and multivariate time series data through a cross-modal attention mechanism to generate joint semantic representation; based on the prediction of the KTSF multi-modal large model, outputting the residual error between the actual data, dynamically identifying abnormal data; and performing attribution analysis; based on the attribution result, calling the KTSF multi-modal large model to generate a repair value, and intelligently correcting the abnormal data; designing a quality evaluation and feedback learning module for calculating data quality scores and driving model incremental updating; designing a rule self-learning module for automatically refining governance rules through cluster analysis and updating the knowledge base.

The invention provides an intelligent retrieval system and method based on an ontology model and data fusion, relates to the field of information retrieval and data processing, and solves the problems of low efficiency and poor user experience in the aspects of data cleaning, ontology design and result presentation in the traditional technology. The system comprises an administrator login and configuration module used for an administrator to log in and configure a data optimization processing tool; the domain expert ontology design module is used for designing and managing the ontology model by domain experts and storing ontology data; the user document uploading and analyzing module is used for receiving a document and analyzing the document into structured data; the graph analysis module is used for generating a knowledge graph; the common retrieval module is used for matching subject fields corresponding to user questions and obtaining retrieval results; the advanced retrieval and graph query module is used for matching corresponding entities from the knowledge graph on the basis of common retrieval and fusing to form an advanced retrieval result; the field specialty and accuracy of the intelligent retrieval system are improved.

The invention discloses a punching machine for metal manufacturing, and relates to the technical field of punching machine tables. Comprising a machining table, a stamping mechanism is fixedly installed at the top of the machining table, a buffer damper is fixedly installed at the top of the machining table, a lifting plate is fixedly installed at the top of the buffer damper, a hollow cylinder is fixedly installed on the surface of the telescopic end of the buffer damper, and a Z-shaped piece is fixedly installed at the bottom of the hollow cylinder. The annular plate is fixedly mounted at the bottom of the Z-shaped piece, the mounting frame is fixedly mounted on the surface of the hollow cylinder, the rotating plate is rotatably mounted on the surface of the mounting frame, the hollow cylinder moves upwards to generate negative pressure between the hollow cylinder and the buffer damper, and the rotating plate is provided with air holes. By increasing the resistance borne by the telescopic end of the buffer damper when the telescopic end of the buffer damper resets upwards, the resilience force of the buffer damper is reduced, and therefore the buffer damping effect is improved.

Provided are a user-friendly single-crystal X-ray structure analysis device and method whereby single-crystal X-ray structure analysis based on a fine-pore complex crystal (crystal sponge) can be quickly performed and management of related information is facilitated. A single crystal X-ray structure analyzer is provided with: a sample holder containing a porous complex crystal capable of occluding a sample in a plurality of micropores formed therein; the goniometer is used for installing and rotating the sample holder; an X-ray irradiation unit that irradiates X-rays from the X-ray source to a sample held in a sample holder attached to a goniometer; an X-ray detection and measurement unit that detects and measures X-rays diffracted or scattered by the sample; and a structure analysis unit that performs structure analysis of the sample on the basis of the diffracted or scattered X-rays measured by the X-ray detection and measurement unit, the structure analysis unit performing structure analysis of the sample on the basis of the occlusion results in the step of occluding the sample with the porous complex crystal.

The utility model discloses a general explorator storage device, which belongs to the technical field of steel processing, solves the problem that the existing explorator cannot be quickly and accurately taken and stored, and comprises a base, four groups of first angle steels and side baffles, the first angle steels are symmetrically arranged on the upper surface of the base, and the side baffles are arranged on the upper surface of the base. The first angle steel is installed at the top corner of the base, a side baffle is fixedly installed between the first angle steel, a collecting mechanism is installed between the first angle steel and located on the side face of the side baffle, and the collecting mechanism comprises a discharging plate and a storage hole. According to the template storage device, templates of different specifications can be stably stored, the templates are reasonably placed according to the sizes of the templates, follow-up operators can conveniently and accurately retrieve and take the templates, and the working efficiency during steel machining is guaranteed.