133 results about "Information mining" patented technology

The invention discloses a multi-modal data retrieval method and device, a storage medium and computer equipment. The method comprises the following steps: collecting multi-modal original data; on the basis of description information of metadata of original data, all metadata belonging to the same associated items and logic relations among all the metadata are obtained, a metadata chain is constructed, a distributed graph database is constructed on the basis of the metadata chain, the metadata chain is expressed in the distributed graph database in the form of a graph, the graph comprises nodes and edges, the nodes represent the metadata, and the edges represent the metadata. The edge represents a logical relationship between the metadata; and when a data retrieval instruction is received, traversing each node in the metadata chain along the logical relationship of the metadata chain in the distributed graph database, obtaining a target node matched with a data retrieval requirement corresponding to the data retrieval instruction, and returning original data corresponding to metadata represented by the target node. Multi-modal data dynamic association retrieval can be realized, and cross-modal information mining efficiency and accuracy are improved.

The invention discloses a multi-source data resource key information mining method and system, and the method comprises the steps: obtaining multi-source data of various forms and types, and carrying out the preprocessing; performing division according to data types; word frequency statistics and word correlation analysis are carried out on the character data, and numerical value meanings of the digital data are extracted through context semantic correlation; constructing a cross-modal key information data set based on the key information data set of each data type; designing a multi-branch long-short-term memory neural network model, inputting a cross-modal key information data set into the model for training, and generating key information mining models for different fields by dynamically adjusting network parameters; and selecting to carry out key information real-time extraction and updating on the optimal model, generating a knowledge graph, and carrying out visual operation. The method has the advantages that through multi-source data fusion and processing, the long-short term memory neural network model is combined, the knowledge graph is generated, and domain-specific key information mining and visualization operation are efficiently carried out.

The invention relates to the technical field of data processing, and discloses a gait rehabilitation evaluation method and system based on spatial-temporal characteristics of multi-modal data, a terminal and a storage medium, and the method comprises the steps: synchronously collecting myoelectricity, electroencephalogram and exoskeleton robot IMU signals of a dyskinesia crowd in a certain motion normal form in a specific period interval; a motion function scoring result of a doctor on a patient according to a scale is synchronously recorded, data preprocessing, channel selection and feature extraction are performed on myoelectricity, electroencephalogram and exoskeleton robot IMU signals, a model is established based on a neural network, motion evaluation is performed according to a fused feature array, and a motion function quantitative score is obtained; and the motion function state of the evaluated person is graded according to the score. According to the method, the multi-modal synchronous data rehabilitation evaluation model of the movement function of the dyskinesia crowd is constructed according to multiple information mining, and the movement function state of the evaluated person is rapidly evaluated, predicted and graded.

The invention relates to a GIS (Gas Insulated Switchgear) equipment mechanical defect diagnosis method based on a Gramb angle field and a dual-channel PCNN-Attention neural network, and belongs to the technical field of gas insulated switchgear mechanical vibration defect diagnosis. The method solves the problems that traditional diagnosis depends on artificial feature extraction, so that subjectivity is high, information mining is insufficient, and defect severity evaluation is missing. According to the technical scheme, the method comprises the steps that a one-dimensional vibration signal is converted into a GASF two-dimensional image and a GADF two-dimensional image through a GASF field so as to completely reserve time sequence topological features; carrying out data enhancement by adopting an image geometric transformation technology so as to improve the generalization ability of the model; and a dual-channel PCNN-Attention model is constructed, and synchronous intelligent identification of defect types and severity is realized through parallel feature extraction and dynamic weight optimization of an attention mechanism. According to the method, the diagnosis accuracy, reliability and adaptive capacity are improved, and support is provided for equipment state operation and maintenance.

The invention discloses a transformer substation fault detection method and related equipment, and the method comprises the steps: obtaining the real-time data of a transformer substation secondary system, forming a scene prediction model through a deep learning model and a dynamic Bayesian network, and determining a fault detection result in combination with a digital twin system. The deep learning model can deeply mine multi-source real-time data features, accurately recognize fault information in a complex scene, make up for the problem of insufficient information mining of a traditional data driving method, and improve the accuracy of fault detection and positioning. The dynamic Bayesian network carries out probability modeling on the dynamic change of the system state, and the problems that a traditional method is difficult to adapt to complex working conditions and cannot accurately simulate system evolution are solved. Meanwhile, the digital twin system realizes deep fusion analysis of multi-source data, constructs an intelligent processing closed loop from data acquisition to maintenance decision, remarkably improves the timeliness, accuracy and intelligent level of fault detection, and provides reliable guarantee for stable operation of a transformer substation.

The invention discloses an innovative resource collaborative matching method and system oriented to distributed operation, and relates to the field of data processing recommendation, and the method comprises the steps: carrying out the standardization processing of a heterogeneous resource portrait description text through a large language model, constructing and forming a resource attribute knowledge graph, generating extension description information of the resources in the target industrial chain through the large language model, and constructing and forming an industrial chain knowledge graph; performing semantic alignment and hidden associated information mining on the fused graph by using a pre-trained SACN model, and constructing a resource collaborative knowledge graph representing resource attributes and collaborative logic in an industrial chain; and performing semantic fusion on the original description text representation of the resource and the resource collaborative knowledge graph representation based on a large language model, generating comprehensive vector representation through multi-modal fusion, and outputting a resource matching recommendation result by adopting a collaborative filtering recommendation model. According to the method, the cross-organization and cross-industry-chain innovative resource collaborative knowledge graph is constructed, instant and clear demands of users are accurately responded, and the resource allocation efficiency is improved.

The invention provides a hotspot technology mining method and system based on scientific and technological literature data, and relates to the technical field of scientific and technological information mining, and the method comprises the steps: firstly obtaining a scientific and technological literature data set containing papers and patent literatures, and constructing a current technical element network through technical phrase extraction and semantic normalization processing; dividing the technical element network segment into a time-serialized technical element network segment set according to time slices, calculating nodes and edge time sequence importance indexes to judge an emerging technical cluster, carrying out dynamic graph representation learning and future associated edge prediction on the set, generating a potential technical combination prediction set, and evaluating the potential technical combination prediction set; and finally, generating an emerging technology theme list and a technology fusion direction list in combination with an emerging technology cluster node growth trend and a prediction association edge evaluation result, thereby improving the accuracy, comprehensiveness and foresight of hotspot technology mining.

The invention provides a multi-view space transcriptomics clustering method based on sharing-specific information mining. The technical problems that in an existing space domain recognition method, robustness of a single view is insufficient, multi-view information fusion is insufficient, and description of data statistical characteristics is not accurate are solved. According to the technical scheme, the method comprises the following steps: S1, acquiring and preprocessing original data of a spatial transcriptome; s2, constructing a sharing-specific decomposition coding network; s3, using a ZINB expression decoder, a structure decoder and a Student's t distribution clustering module to construct a multi-task loss function; and S4, outputting a spatial domain division result based on the trained network. According to the method, the spatial domain recognition precision and robustness are remarkably superior to those of an existing method, the data statistical characteristics and biological significance can be accurately described, stable support is provided for tissue function analysis and tumor microenvironment research, and the analysis quality and application value of spatial transcriptomics data are effectively improved.

The invention provides an information deep mining method and system based on a multi-agent cooperation framework, and the method comprises the steps: firstly receiving a to-be-mined information demand, carrying out the task decomposition and cooperative scheduling through a plan expert agent, generating a cooperative task scheduling scheme, and then carrying out the multi-source information cooperative search through a search expert agent according to the scheme, the method comprises the following steps: acquiring multi-source original information, reading an expert agent to carry out deep analysis and association extraction on the multi-source original information, generating structured association information, analyzing the expert agent to carry out multi-level deep reasoning on the structured association information to obtain a deep mining intermediate result, and finally, integrating multi-dimensional verification and fusion on the intermediate result by the expert agent to obtain a deep mining result. And generating a final information deep mining result meeting the requirement, thereby realizing automation, intelligence and high efficiency of information mining, and mining the potential value of the information.

The embodiment of the invention provides a multi-modal data processing method and system based on a knowledge graph, and aims to solve the problems of semantic association deficiency and potential association mining difficulty in multi-source information processing. The method comprises the following steps: firstly, acquiring a multi-source information set containing different expression form information units and source identifiers, and performing entity mapping on the multi-source information set and a preset semantic association network to generate an information unit set with semantic association tags; performing cross-source feature association processing to obtain a comprehensive feature set; then, a semantic reasoning rule is called for association extension, and an extension feature set is generated; and finally, based on the extended feature set, generating an information processing result conforming to an application scene and feeding back the information processing result to a corresponding interface, thereby effectively integrating multi-source information and mining potential association.

The application discloses a kind of multi-source data resource key information mining method and system, comprising: obtaining each form type multi-source data, and pre-processing;According to data type, it is divided;Word frequency statistics and word association analysis are carried out to literal data, and the numerical value meaning of digital data is extracted by context semantic association;Based on each data type key information dataset, cross-modal key information dataset is constructed;Design multi-branch long short-term memory neural network model, input model is carried out training to cross-modal key information dataset, and key information mining model for different fields is generated by dynamically adjusting network parameters;Optimal model is selected to carry out key information real-time extraction and update, generate knowledge graph, and carry out visual operation.The application has the advantages that: by multi-source data fusion and processing, long short-term memory neural network model is combined and knowledge graph is generated, and field-specific key information mining and visual operation are efficiently carried out.

The present invention provides a TIN-DDM terrain complexity measurement method based on rolling ball transformation, belonging to the technical fields of seabed terrain information mining and analysis and TIN-DDM multi-scale expression technology. The present invention introduces the change in sampling point depth during the rolling ball transformation process into the construction process of the TIN-DDM sampling point TCI calculation model. The present invention clarifies the correlation between the depth change of the TIN-DDM sampling point and its TCI during the rolling ball transformation process, analyzes the selection principle of the optimal rolling ball radius, constructs and optimizes the sampling point TCI calculation model, and finally achieves an effective measurement of TIN-DDM terrain complexity by interpolating the complex field with the TCI. This solves the problem that the TCI calculation models established by the current TIN-DDM terrain complexity measurement methods are unable to effectively reflect the differences in morphological change characteristics between different terrain units by comparing the TCI values of the sampling points.

This invention discloses a method for mining alumni information based on a large-scale model, comprising the following steps: 1) acquiring raw alumni-related data; 2) preprocessing the acquired data; 3) performing natural language processing on the acquired data; 4) semantic parsing and information extraction using a large-scale model; 5) constructing a knowledge graph; storing the extracted alumni information and relationships in a graph database; and constructing a visualized alumni knowledge graph; 6) updating the knowledge graph data; and 7) providing intelligent retrieval and question-answering services for alumni information based on the knowledge graph data. This invention uses a large-scale model to perform deep semantic understanding and information extraction from unstructured text. Compared to traditional keyword-based or small-scale model-based methods, it can more accurately identify alumni identities, professional experiences, and social relationships, while reducing the false matching rate.

The application provides an atoll vegetation index prediction method, which can be applied to the technical fields of geographic information and deep learning. The method comprises the following steps: obtaining monthly normalized vegetation index data of a target atoll, corresponding marine environment data and meteorological data, and constructing a vegetation growth dataset; calculating the partial correlation of the marine environment data, the meteorological data and the normalized vegetation index change, so as to determine the key environmental data affecting the normalized vegetation index change; determining the lag response period of the normalized vegetation index to each key marine environment data; based on the lag response period, combining attention mechanism and double-time-scale information mining, constructing a double-time-phase convolution long short-term memory neural network model considering the climate time lag effect of the atoll, wherein the neural network model cell structure comprises convolution operation; training the model by using the vegetation growth dataset to obtain a trained model; and predicting the future short-term atoll vegetation normalized vegetation index change by using the trained model.

The invention discloses an unsupervised domain adaptation method for a remote sensing time sequence with phonological feature alignment, which mainly solves the problem of poor generalization performance caused by difficulty in coping with time migration and distribution migration in a cross-domain task due to neglect of phonological information in time sequence data in the prior art. According to the scheme, the method comprises the following steps: constructing a self-adaptive phenological segmentation module based on time constraint, extracting a representative phenological stage from a satellite image time sequence, and accurately capturing key time sequence characteristics of a crop growth cycle; a comparison learning module based on phenological difference is introduced, stage-level positive and negative sample pairs are established, double loss optimization is carried out, and fine alignment of cross-regional phenological features is realized. Through phenological information mining and stage-level comparison and alignment, the precision and robustness of cross-regional crop semantic segmentation under an unsupervised condition are significantly improved, and the generalization ability of a remote sensing deep learning model in an unknown target region is effectively improved.

The present invention relates to a binocular narrow butt weld detection method based on deep learning, comprising the following steps: projecting stripe structured light and passive light by a projector, collecting weld images, and obtaining the point cloud of the welded part; performing two-dimensional data annotation based on image binarization processing and binocular consistency correction; constructing a two-dimensional weld extraction model based on spatial information mining to extract two-dimensional welds; mapping two-dimensional pixels into three-dimensional space coordinates based on a binocular vision model. Pose estimation is performed based on the point cloud of the welded part and the local neighborhood feature information of the weld points. The binocular narrow butt weld detection system of the present invention realizes accurate and efficient weld extraction; specifically, it is an accurate and reliable automatic data annotation method; specifically, it is a weld extraction network that realizes the extraction of two-dimensional coordinates of the weld to assist three-dimensional weld positioning. The present invention provides a necessary strategy for realizing more accurate and efficient narrow butt weld detection.

The present invention discloses a method, device and storage medium for mining key nodes in a directed network, belonging to the field of network information mining technology. The mining method comprises: decomposing strong connectivity to obtain a strongly connected subgraph with zero in-degree, selecting a node with the smallest in-degree difference and the smallest in-degree from all strongly connected subgraphs with zero in-degree to meet the reachability condition; then using a greedy algorithm to select nodes that are farther away from the key node group and have a smaller in-degree minus out-degree and in-degree from the strongly connected subgraph with the smallest eigenvalue in each cycle and add them to the key node group; that is, using the graph theory of measurement indicators, the shortest path concept and the greedy algorithm to achieve the mining of key nodes in a large-scale directed network. Since key nodes can directly or indirectly affect all other nodes in the network, the stability of the existing complex network node importance evaluation method in the control synchronization and diffusion capabilities of the directed network is improved.

The present invention proposes a remote sensing image segmentation method and system based on an improved SwinTransformer. The encoder is composed of a Swin transformer backbone network and a global information enhancement module. The Swin transformer backbone network fully mines the global context information of the remote sensing image to obtain global feature maps of different scales and levels. The global information enhancement module is used to perform global pooling operations on the feature maps in the horizontal and vertical directions. The global information enhancement module is fused with the global feature maps output by the corresponding Swin transformer backbone network at different stages, thereby making up for the global information mining capability of the SwinTransformer backbone network limited by the window mechanism. The decoder fuses the global enhanced feature maps of different scales and levels output by the encoder through an attention mechanism, thereby retaining more local information while obtaining more extensive context information.

The invention discloses a consumer complaint information mining method, device and equipment based on a large language model, and relates to the field of product defect risk assessment and recall management. Constructing a standardized product security data set according to the multi-modal data, and performing semantic recognition and labeling on multi-modal complaint information in the data set to obtain a structured product security event information flow; in response to expiration of a preset update period, analyzing the product security event information stream and the current latest product security dynamic knowledge graph to obtain a product security event; the current latest product security dynamic knowledge graph is generated by fusing newly added product security event information flow on the basis of a historical knowledge graph; and carrying out situation awareness, risk early warning and defect correlation analysis on the product security event to finally obtain a product defect assessment result. The result is fed back to the current latest product safety dynamic knowledge graph to be optimized and updated; according to the method, semantic recognition and understanding capability can be realized, and dynamic association analysis capability is improved.

The application discloses a kind of digital archives intelligent information mining method, it is related to file information processing technical field, the present application is evaluated data quality by a series of quantitative indicators including consistency index, integrity index, importance index and comprehensive importance index, which effectively reduces subjective judgment when manual screening, make data screening process more objective, reliable and scientific, simultaneously, by combining quantitative indicators with comprehensive weight, to automatically generate weighted threshold based on data-driven, the application of this weighted threshold not only can quickly process large-scale data set, to reduce the demand for a large number of manual annotation of manpower, and can effectively filter out high-quality data set, and be applied to subsequent deep learning model training, to significantly improve the output accuracy and stability of model, ensure that more accurate prediction results can be obtained, and based on this more accurate prediction results, more scientific and reasonable business decisions can be made.

The invention belongs to the technical field of data mining, is applied to the field of medical health and the field of automobile insurance, and discloses a data mining method, device and equipment and a storage medium. The method comprises the steps of obtaining original data in a data source, and performing feature extraction on the original data; performing data dimension reduction on the extracted features, and projecting high-dimensional data to a low-dimensional space; performing data mining on the data after dimension reduction, wherein the data mining method comprises clustering analysis, association rule mining, classification and regression analysis; the mined data is evaluated; and displaying an evaluated result by adopting a visual tool. According to the method, the efficiency and accuracy of data mining are improved, and information mining from mass data is facilitated.

The application discloses a market supervision internet information monitoring and analysis system, which comprises a CNN model construction module, an internet information collection module, an information preprocessing module and an information mining module; the application uses a CNN model algorithm to model sample data marked by artificial marking, and uses a model to determine illegal behaviors; through continuously accumulating sample data, repeatedly training and achieving the purpose of machine learning, the determination accuracy is gradually improved.

The present invention provides a fire detection method and system based on power data analysis. After obtaining a set of power equipment to be analyzed and a historical abnormal monitoring data set of each power equipment to be analyzed in the set of power equipment to be analyzed, one or more abnormal operation events are determined in the historical abnormal monitoring data set. According to the historical abnormal monitoring data set, the power equipment to be analyzed and the abnormal operation events are used as graph vertices to generate a power analysis graph. The power analysis graph is mined for characterization information to obtain a power equipment characterization vector of the power equipment to be analyzed. Based on the power equipment characterization vector, the power equipment to be analyzed is grouped to obtain one or more pending power equipment groups, and a target power equipment group is determined in the pending power equipment groups. The present application can improve the accuracy of the determined power equipment groups, help focus on monitoring the equipment in the group during fire monitoring, and make the fire monitoring more targeted.

The invention relates to the technical field of intelligent monitoring, and discloses a system operation state monitoring method based on multi-source information mining and organic fusion, which comprises the following steps: acquiring internal operation state data and external operation state data, calibrating the internal operation state data and the external operation state data by using a sensor calibration algorithm based on artificial intelligence; the method comprises the following steps: extracting internal operation state data and external operation state data, performing organic fusion to form a comprehensive data set, performing correlation analysis on the comprehensive data set, mining a correlation relationship between the internal operation state data and the external operation state data, and deriving key variables influencing the operation state of a monitoring target system; and comparing the real-time data of the key variables with a preset normal operation range to obtain a comparison result, evaluating the current operation state of the monitoring target system based on the comparison result, obtaining an evaluation result, and when the evaluation result exceeds a preset standard value, triggering an early warning mechanism to prompt a user to take corresponding measures. According to the invention, the monitoring precision of the monitoring target can be improved.

The invention belongs to the technical field of alloy fatigue life prediction, and discloses an alloy fatigue life prediction method fused with an information mining technology, which comprises the following steps: introducing key physical quantities into a feature set by combining a fatigue theory; mining inherent information in the data by adopting a symbolic regression algorithm driven by genetic programming, and adding the inherent information into a feature set as derivative features so as to construct a fatigue life prediction model of the alloy; a prediction model without derivative features is constructed for performance comparison, so that the improvement effect of fusion information mining technology aided modeling on prediction precision is evaluated; derivative feature expansion is applied to fatigue life prediction of a multi-component alloy system, and the universality of mined information is verified. According to the method, the information mining technology is adopted, hidden information in the data is extracted to assist modeling, and compared with a traditional machine learning method, more universal and high-precision fatigue life prediction is achieved, so that more reliable prediction support is provided for risk assessment of structural components.

The present disclosure provides a method and device for information mining, electronic equipment and storage medium, relates to the technical field of data processing, and particularly relates to the technical field of big data. The specific implementation scheme is as follows: obtaining user basic information and behavior related information of each user from log information; generating a graph database based on the user basic information and the behavior related information, the graph database comprising a plurality of nodes and connection relationships between the nodes, the plurality of nodes comprising a to-be-reviewed node and an abnormal node; dividing the nodes included in the graph database into a plurality of communities based on a community discovery algorithm; and for each to-be-reviewed node in the graph database, if the closeness of the to-be-reviewed node to abnormal nodes in the community to which the to-be-reviewed node belongs is greater than a preset threshold, determining that the to-be-reviewed node is an abnormal node. In this way, potential abnormal users can be identified.

The present disclosure provides an advertisement business information mining method and device, an electronic device and a storage medium, and relates to the technical field of artificial intelligence, specifically to the technical field of data mining, and can be applied to scenarios such as advertisement business information mining. The specific implementation scheme is: at least one first candidate business point matched with the semantics of the target description content is selected from a business point set, and the business points in the business point set represent the business scope of the advertisement; attribute comparison results of each first candidate business point and the attribute information of the corresponding target description content are obtained; and according to the attribute comparison results corresponding to each first candidate business point, a target business point set is determined from the at least one first candidate business point. The above method can automatically determine the actual business point of the advertisement from the business point set based on the description content of the advertisement, and can reduce the labor cost and time cost consumed for business mining of the advertisement.

The invention provides a passive sonar target space-time frequency feature intelligent mining and fusion identification method, which comprises the following steps: firstly, designing a data preprocessing method, and generating a space domain sample, a time domain sample and a frequency domain sample through space-time frequency feature analysis; secondly, constructing a spatial domain feature extraction module, a frequency domain feature extraction module and a time domain feature extraction module which are parallel, constructing a serial cross-domain dynamic attention fusion module and a feature secondary fusion and classifier module on the basis, and combining to form a complete space-time-frequency feature mining and fusion recognition network model; designing a loss function and configuring an optimizer, and completing space-time frequency feature mining and fusion identification network model training based on the labeled sample; and finally, processing the unknown target space-time frequency data based on the designed preprocessing method and the well-trained space-time frequency feature mining and fusion recognition network model to obtain a target classification recognition result. Compared with an existing passive sonar target classification and recognition method, the method provided by the invention is richer in target useful feature information mining, deeper in fusion and utilization, and higher in classification and recognition confidence.

This invention discloses a method and device for optimizing prostate MRI images for clinical surgery, belonging to the field of prostate MRI image processing technology. The optimization method includes: retrieving an initial MRI image; retrieving a trained data fitting network; and adjusting the initial MRI image using the data fitting network to generate an optimized, improved MRI image. The data fitting network includes an information mining module and an enhancement residual module. The information mining module is used to mine image features from the initial MRI image, and the enhancement residual module processes the feature map preceding the information mining module to generate a residual feature map. Then, the residual feature map is used to register the feature map following the information mining module. This invention uses the enhancement residual module to perform dual-domain constrained enhancement adjustment on the preceding feature data to generate a residual feature map, which can greatly enhance the network's targeted optimization effect on the target region and avoid the distortion of image detail information.

The invention discloses a video self-supervision deblurring method based on supervision information mining, belongs to the field of video deblurring, and aims to solve the problem that an existing self-supervision video deblurring method is poor in effect on a real scene. The method comprises a clear frame supervision video deblurring step and a self-enhancement video deblurring step. A clear frame supervision video deblurring step: selecting clear frames existing in the input fuzzy video, aligning the clear frames with the output of the deblurring model and then taking the clear frames and the deblurring model as supervision, and training the deblurring model; a self-enhancement video deblurring step: constructing high-quality pairing training data for further training a deblurring model; when the deblurring model is trained, space consistency keeping limitation is carried out, the result of the deblurring model earlier in the training stage is used as a constraint, and it is guaranteed that output of the deblurring model is kept consistent with input in space.