45 results about "Confidence set" patented technology

The invention provides a two-process error correction method and device for real-time speech transcription, and relates to the technical field of speech processing, and the method comprises the steps: extracting the Mel spectrum features of each segment, inputting each Mel spectrum feature into a lightweight end-to-end model, and obtaining a preliminary transcription text; splicing the segments according to a preset number to obtain a plurality of long segments, and inputting each long segment into a speech recognition model to obtain a high-precision transcription text; performing text comparison on the preliminary transcription text and the high-precision transcription text according to the confidence degree set of the preliminary transcription text to obtain all error vocabularies in the preliminary transcription text; and performing corresponding error correction processing on each error vocabulary in the preliminary transcription text according to the type of the error vocabulary and the high-precision transcription text to obtain a final transcription text. According to the method, through a two-process transcription error correction mechanism of the preliminary transcription text and the high-precision transcription text, transcription error accumulation is reduced on the premise that the real-time performance is not affected, and the transcription accuracy in a complex scene is improved.

The invention provides a multi-voltage adapter compatible method and system based on Feiteng D3000 and a medium, and the method comprises the steps: firstly, generating a first switch control signal based on a preset voltage threshold comparison logic according to first voltage information inputted by an adapter; then, according to the first switch control signal, a corresponding power supply path is switched on, second voltage information is obtained, and for the second power supply path, a first voltage regulation parameter is configured based on preset voltage regulation logic, and second voltage information is obtained; and finally, on the basis of a preset confidence degree setting logic, first confidence degree information is obtained according to the second voltage information and is used for judging and triggering a redundancy protection mechanism. According to the invention, the compatibility problem of the product is solved through voltage detection and logic decision, multiple protection is formed through dynamic voltage regulation and redundancy protection, and the reliability and safety of the product are improved.

The invention discloses a multi-time-scale park integrated energy system distribution robust optimization scheduling method, which comprises the following steps of: constructing an electric heating collaborative system model taking a combined heat and power generation unit as a core, and introducing a carbon transaction mechanism; constructing a confidence set in combination with a 1-norm and an infinity-norm, and respectively making a robust start-stop plan and a flexible operation strategy in day-ahead and intra-day two-stage scheduling; a column and constraint generation algorithm is adopted to decompose the constructed day-ahead and intra-day two-stage model into a main problem and a sub-problem for repeated iterative solution, and an optimal scheduling scheme with both economical efficiency and robustness is obtained. According to the method, the uncertainty of new energy prediction is fully considered, and the coping of the park to the randomness of the new energy can be better played through processing of different time scales.

The invention relates to the field of computer information processing, in particular to a multi-source heterogeneous knowledge graph construction method and system. The method comprises the steps that ancient books, clinical literatures and traditional Chinese medicine, medicine and disease database data are accessed, and entity candidate sets are formed after field mapping extraction; calculating a name similarity, an attribute similarity and a semantic vector similarity for the entity candidate set, and inputting multiple classes of similarities into a fusion network to generate an entity alignment mapping table; constructing a remote supervision annotation set and training a relationship classification model, driving expert annotation increment update through model uncertainty, and outputting a relationship triple set and a corresponding relationship confidence set; and finally, forming a three-layer heterogeneous graph structure communicated with the syndrome type, the disease, the prescription and the medicinal components, and writing the three-layer heterogeneous graph structure into a graph database to generate a knowledge graph database. Unified modeling of entity alignment, relation extraction and relation strength quantification in the multi-source heterogeneous data is achieved, and the structural consistency in the knowledge graph construction process is improved.

The invention relates to the technical field of hydrology and water resource monitoring and emergency scheduling, in particular to a water source multi-modal monitoring fusion method and system, and the method comprises the steps: firstly obtaining multi-source water environment observation, and forming evidence graph initialization and assimilation initialization; secondly, carrying out cross-modal evidence fusion, generating topological boundary constraints, flow direction constraints and assimilation time windows, and compiling the topological boundary constraints, the flow direction constraints and the assimilation time windows into a constraint set; then executing data assimilation and source item joint inversion under the constraint of a partial differential equation to obtain a source item confidence set, the arrival time of a pollution front outer envelope and a key position, and a quantile interval of peak concentration and standard exceeding duration; and finally, constructing a risk functional and an opportunity constraint based on a quantile interval, and generating a scheduling instruction and partition early warning by adopting rolling predictive control, and performing closed-loop recharge. The method gives consideration to accuracy, robustness and performability.

The invention provides an industrial large model construction method and a storage medium. The method comprises the steps of obtaining multi-modal fusion data of an industrial field; performing scene classification on the industrial field to obtain industrial subdivided scenes; according to the multi-modal fusion data, determining a multi-modal data set of each type of industrial subdivision scene; according to the multi-modal data set, constructing a knowledge organization structure of each type of industrial subdivision scene; according to a text mode in the multi-mode data set, finely adjusting the small language model to obtain a routing center command body; the routing center command body is used for identifying a received user instruction and determining a target industrial subdivision scene confidence degree set; the target industrial subdivision scene confidence coefficient set is used for screening a target model from a combat model library by adopting an adaptive routing decision algorithm according to the target industrial subdivision scene confidence coefficient set, and generating a target answer by adopting the target model; wherein the combat model library comprises a scene model library, a standby reasoning body and a scene expert agent.

The invention discloses an equipment health state intelligent identification-oriented low-quality data robust learning method. The method comprises the following steps of: obtaining multi-source sensor time sequence data of an electromechanical composite transmission system and preprocessing the multi-source sensor time sequence data; constructing a multi-sensor fusion identification network composed of a feature encoder, a projection head and a classification head; generating neighbor consensus tags and consistency measurement for samples according to the neighbor consistency of the embedded space, dynamically screening high-confidence samples and similar sample pairs, performing characterization pre-training by using a joint target of supervised contrast learning, classified cross entropy and sample pair similarity, and introducing a hybrid enhanced smooth discrimination boundary to obtain a robust encoder; freezing a robust encoder, training a new classification head only by using a high-confidence set, and obtaining a state recognition model; a state recognition model is used for directly reasoning a complete channel sample, a feature channel adaptability calibration module is introduced into a sensor channel data missing scene to dynamically compensate feature distribution offset, and probability distribution of each state category is output.

The present invention discloses a multi-path thinking chain reasoning generation method based on fine-grained knowledge retrieval, including: using a large language model to perform deep semantic analysis on user questions, extracting entities contained in the question text and the relationship between entities to generate a triple set; dynamically dividing the triples into different confidence sets by setting high and low double confidence thresholds; constructing a thinking chain framework using pre-defined paths to obtain an initial reasoning path set; optimizing the initial reasoning path, sampling the optimized reasoning path set using Top‑k scoring, and obtaining a reasoning path set with high semantic relevance; dividing the reasoning path set into multiple reasoning path groups with complementary internal path information to obtain multiple candidate answers; using a majority voting mechanism to select multiple candidate answers, and finally obtaining a consistent answer. The present invention can significantly improve the accuracy and interpretability of reasoning, and greatly improve the reliability and robustness of the final conclusion.

The invention discloses a general scientific research entity corpus construction method and device based on scientific and technical literature and a medium, and relates to the technical field of natural language processing and knowledge engineering. Performing automatic pre-labeling of general scientific research entities on the scientific and technical literature abstract set to obtain an extracted general scientific research entity set and a confidence coefficient set; filtering and screening the confidence coefficient set to obtain a difficult sample set; extracting the difficult sample set to obtain a large language model pre-labeling result set; and obtaining a large language model correction labeling result set and a large language model correction degree set. According to the method and the device, the technical problems of low scientific research entity extraction efficiency and insufficient accuracy caused by incomplete general scientific research corpus construction process and lack of an effective feedback optimization mechanism of a model in the prior art are solved, and the technical effects of realizing a high-quality general scientific research entity corpus construction process and improving the entity extraction efficiency and accuracy are achieved.

The invention relates to the technical field of power systems, in particular to a salt-cavern integrated energy system distribution robust optimization scheduling method, which comprises the following steps: establishing an operation model of a salt-cavern integrated energy system; establishing a scheduling model of the salt cavern integrated energy system based on the operation model; obtaining a confidence set of the salt cavern integrated energy system, inputting data in the confidence set into the scheduling model, and outputting scheduling data of the salt cavern integrated energy system by the scheduling model; and optimizing operation data of the salt cavern integrated energy system based on the scheduling data. According to the method, the operation optimality of the salt-cavern comprehensive energy system can be ensured when the salt-cavern comprehensive energy system is scheduled.

The application relates to a target tracking method and system based on classification information assistance, which comprises the following steps: firstly, calculating target motion state prediction and target existence state prediction at the current time; then, combining spatial information of sensor measurement at the current time to calculate associated spatial probability and associated prior probability of the target and the measurement; then, calculating an associated initial confidence set of the target and the measurement according to the two probabilities; then, extracting features from radar range-Doppler spectrum information by using a radar range-Doppler spectrum feature extraction model; then, calculating an associated confidence set of the target and the measurement after fusing classification confidence obtained by using a target classification model and the associated initial confidence of the target and the measurement; then, calculating an update message of the measurement according to the associated confidence set and the spatial information of the measurement; and finally, calculating motion state estimation and existence state estimation of the target at the current time according to the update message and the target motion state prediction and the target existence state prediction. The application can significantly improve the performance and robustness of target tracking.

The present application relates to the field of artificial intelligence tracking technology, and discloses a pedestrian trajectory tracking method based on artificial intelligence.The method comprises collecting multi-source sensor pedestrian movement feature data stream, extracting initial trajectory key points as tracking reference points;dividing associated historical position data into a high-confidence trajectory segment set and a to-be-verified trajectory segment set, and discarding the to-be-verified set;arranging the high-confidence set in descending order according to the number of trajectory points, selecting the first several segments as optimal trajectory segments, and calculating the proportion;comparing the proportion with a dynamic screening threshold, outputting the high-confidence set as an initial tracking result, or outputting a trajectory checking instruction generated by comparing trajectory data;in response to the checking instruction, dividing the space grid of the optimal trajectory segment, and outputting an optimization instruction in combination with historical trajectory features;updating the reference points based on the optimization instruction, and screening and outputting secondary divided trajectory data;finally, obtaining parameters to generate a tracking efficiency evaluation coefficient, and starting a parameter correction strategy when the coefficient is lower than an optimization threshold.

The embodiment of the invention provides a construction method and device of an energy storage cluster regulation and control model, equipment and a medium, and relates to the technical field of energy management. The method comprises the following steps: acquiring a plurality of source load powers and a plurality of energy storage equipment data; performing set construction according to the plurality of source load powers to obtain a confidence set; clustering is carried out according to the multiple pieces of energy storage equipment data to obtain multiple energy storage clusters and an adjustable load capacity curve of each energy storage cluster; and performing model construction according to the confidence set, the adjustable load curve and a plurality of preset power grid parameters to obtain an energy storage cluster regulation and control model. A probability distribution deviation boundary of power fluctuation is quantified by constructing a confidence set, so that a regulation and control plan has robustness for coping with source-load random fluctuation; a plurality of energy storage clusters and adjustable load capacity curves thereof are formed through clustering analysis, so that the energy storage cluster regulation and control model realizes full-level refined regulation and control from clusters to equipment, and the problem of low regulation and control efficiency of the energy storage cluster regulation and control model in the prior art is solved.

The invention discloses a multi-frame superposition point cloud super-resolution system for three-dimensional reconstruction, particularly relates to the field of structured light three-dimensional scanning, and is used for solving the problem that a thick shell and false details are formed due to the fact that abnormal points and geometric conflicts are mixed and fused during repeated scanning and superposition. The method comprises the following steps: collecting multi-frame structured light point clouds of the same target to generate a multi-frame point cloud set, synchronously generating a point-level confidence set, registering the multi-frame point cloud set to obtain a pose transformation result, generating a unified coordinate point cloud set, and generating a point-level registration residual set; generating a complementary sampling label on the unified coordinate point cloud set according to a fringe phase reproduction degree parameter and a line-of-sight interlayer separation parameter, outputting a complementary point set, an abnormal point set and a to-be-corrected point set, and executing attachment merging and limited fusion according to the complementary sampling label to obtain a super-resolution point cloud result. And the point-level confidence set and the complementary sampling label are updated through re-input consistency verification, and a super-resolution point cloud result subjected to verification constraint is output.

The application provides a compatible method, system and medium of a multi-voltage adapter based on Feiteng D3000, first, according to first voltage information input by the adapter, a first switch control signal is generated based on preset voltage threshold comparison logic, then, according to the first switch control signal, a corresponding power supply path is turned on to obtain second voltage information, wherein, for the second power supply path, a first voltage regulating parameter is configured based on preset voltage regulating logic to obtain the second voltage information, and finally, according to the second voltage information, first confidence information is obtained based on preset confidence setting logic to judge whether a redundant protection mechanism is triggered. Through voltage detection and logic decision, the application solves the compatibility problem of the product, and through dynamic voltage regulation and redundant protection, multiple protections are formed to improve the reliability and safety of the product.

The application provides an improved same-name disambiguation method and device based on cross-source cross error correction, comprising: obtaining a cross-data-source data set; constructing an initial matching model, comparing the similarity of authors associated with a paper in an internal academic graph and authors associated with the paper in an external academic graph, dividing the cross-data-source data set to obtain a confidence set, an unconfidence set and a fuzzy set; obtaining first data in the fuzzy set, converting the hard label of the first data into a soft label, and generating a first target loss function according to the soft label; obtaining second data in the confidence set and third data in the unconfidence set, and performing cross error correction on the second and third data by using a preset cross error correction method to generate a second target loss function; training the initial matching model according to the first and second target loss functions to obtain an enhanced target matching model, and correcting a same-name disambiguation result by using the enhanced target matching model.

The invention discloses an account classification method and device, a storage medium and electronic equipment. Relates to the field of financial science and technology, and the method comprises the steps: in each first classification model of an account classification model, predicting the type of a target account based on service data generated when the target account uses a resource replacement service, and obtaining a confidence coefficient set, the first classification model is trained based on sample data obtained after oversampling processing is carried out on original sample data; in a second classification model of the account number classification model, fusion processing is carried out on the confidence coefficient set to obtain a fusion confidence coefficient, and the second classification model is trained based on a training output result obtained by processing the sample data through the first classification model; and when the fusion confidence is greater than a second threshold, determining that the target account belongs to the low-frequency account. Through the account classification method and device, the problem that the accuracy of account classification is low in the prior art is solved.

The invention discloses an ultrasonic video ovary lesion analysis method fusing anatomical prior and spatial-temporal characteristics. Extracting features of three scales from each frame of image in the ultrasonic video sequence through the backbone network, and stacking the features of the same scale along the time dimension to obtain multi-scale frame-level features; the features of different scales are optimized and fused through a feature pyramid module, and scale enhancement features are obtained; through an anatomical prior enhancement module, generating a prediction segmentation map of the ovary and the focus according to the scale enhancement features; through a gating mechanism, obtaining anatomical prior enhancement features according to the prediction segmentation map and the scale enhancement features; performing space-time fusion on the anatomical prior enhancement features through a pixel-level time sequence attention mechanism through a space-time feature fusion module to obtain fusion features; and through the target detection head, generating a focus bounding box set and a confidence set of each frame of image based on the fusion features. The problems that a traditional single-frame detection method lacks time sequence association and background interference in a dynamic video is large can be solved.

The invention relates to the technical field of enterprise credit risk assessment, and discloses an enterprise credit risk assessment method and system based on information cross validation, and the method comprises the steps: constructing an iterative risk index combination according to a risk assessment index and a forward risk index combination, calculating the default rate confidence of the risk assessment index according to a default rate approximation value set, and calculating the default rate confidence of the risk assessment index; constructing a credit risk regression assessment model set according to the default rate confidence set, identifying the current index default rate of the current risk assessment index to obtain a current index default rate set, and performing information cross validation calculation according to the current index default rate set and a target risk regression assessment model. And obtaining an enterprise credit risk assessment value. According to the invention, the evaluation timeliness and the evaluation precision in the enterprise credit risk evaluation process can be improved.

This invention discloses an intelligent warehouse automated inventory management method and system, relating to the field of warehouse management technology. The method includes: deploying automatic identification tags at each storage location in the warehouse and constructing a digital map of the warehouse; a mobile inventory device collecting tag signals from the automatically identified tags at the current location and the real-time positioning coordinates of the inventory device itself; filtering all storage locations within a preset radius centered on the current location of the inventory device; determining the spatial association and attribution of the candidate storage location set based on the tag signals and real-time positioning coordinates to obtain a storage location attribution confidence set; performing optimal matching based on the storage location attribution confidence set to obtain the tag identification result; and updating the inventory data based on the tag identification result. This invention effectively suppresses interference caused by multipath reflection of radio frequency signals and coupling of signals from adjacent storage locations in dense metal environments through multi-constraint verification in the spatial dimension, significantly improving the accuracy of storage location identification and the reliability of inventory data while ensuring real-time performance.

The embodiment of the application provides a device safety risk identification method and system based on continuous interaction of mine agents, after the mine agent receives and analyzes device abnormal natural language description, calls analysis multi-source time sequence signal flow, outputs an initial fault hypothesis set and an initial fault confidence set, according to the confidence deviation of the initial fault confidence set relative to a preset confidence threshold, constructs a fault hypothesis space convergence strategy of the initial fault hypothesis set to call multi-source compensation signal flow and iteratively update fault root cause, until the updated fault confidence set meets the confidence threshold, and outputs a structured device fault diagnosis report. The problems that the existing mine device fault diagnosis exists cross-monitoring system data island and relies on artificial experience, which leads to deviation of fault diagnosis and delay of diagnosis result feedback response are solved. The mine device fault diagnosis achieves a second-level response, completely eliminates the artificial experience blind area and cross-system data island, and greatly reduces the operation and maintenance cost.

The application relates to the technical field of intelligent sorting, in particular to a car headrest sorting method and system based on confidence matching, which comprises the following steps: collecting a target headrest image and preprocessing the target headrest image; performing multi-dimensional physical feature extraction on the preprocessed target headrest image to generate a physical feature vector of the target headrest; calculating the confidence of the physical feature vector of the target headrest and each template feature vector in a preset headrest type feature library; judging the type to which the target headrest belongs based on the calculated confidence set of all headrest types and a judgment threshold; and performing corresponding operations according to the judgment result of the target headrest. The application can improve the recognition accuracy and stability, and ensure that the headrest type recognition is label-free, automated and scalable in a flexible production environment.

The application discloses an ultrasonic nodule matching method and system, electronic equipment and medium. It relates to the technical field of digital image processing. The method comprises judging whether the ROI image of the frame is processed for the first time; if yes, adding the ROI image of the frame to two image libraries; if no, processing the ROI image of the frame and the two image libraries by using a triple co-occurrence neural network to obtain two confidence sets; judging whether there are confidence values less than a first set threshold in the two confidence sets; if yes, classifying the nodule in the frame image into a target nodule and updating a nodule best cross-section image library and a nodule latest frame image library according to the ROI image of the frame; if no, adding the ROI image of the frame to the nodule best cross-section image library and the nodule latest frame image library. The application can improve the accuracy and robustness of the nodule matching result.

The invention provides an unmanned area road identification method and system based on a convolutional neural network, and relates to the technical field of data identification, and the method comprises the steps: obtaining a plurality of road scene images; road semantic segmentation and semantic uncertainty analysis are carried out on the plurality of road scene images based on a convolutional neural network, and road space structure optimization and geometric uncertainty analysis are respectively carried out on a plurality of road scene image segmentation region sets by using a geometric consistency mechanism; and based on the plurality of pixel-level road probability-semantic confidence degree sets, the plurality of road space structure sets and the plurality of road geometric consistency coefficient sets, carrying out dual-uncertainty cooperative credible region screening, and carrying out network graph structure topology on the initial road space structure set to obtain a road recognition result. The technical problem of insufficient recognition accuracy and reliability of a depopulated area road recognition method in the prior art is solved. The technical effect of improving the accuracy and reliability of road recognition in the unmanned area is achieved.

The present application relates to the field of computer information processing, and particularly relates to a multi-source heterogeneous knowledge graph construction method and system. The method comprises the following steps: accessing ancient books, clinical literature, and data of traditional Chinese medicine, drugs and disease databases, and forming an entity candidate set after field mapping extraction; calculating the name similarity, attribute similarity and semantic vector similarity of the entity candidate set, and inputting the multiple types of similarity into a fusion network to generate an entity alignment mapping table; constructing a remote supervision annotation set and training a relationship classification model, driving expert annotation incremental update by model uncertainty, and outputting a relationship triple set and a corresponding relationship confidence set; finally forming a three-layer heterogeneous graph structure connected by syndrome type, disease, prescription and drug components, and writing into a graph database to generate a knowledge graph database. The present application realizes unified modeling of entity alignment, relationship extraction and relationship strength quantification in multi-source heterogeneous data, and improves the structural consistency of the knowledge graph construction process.

This invention provides a method and apparatus for automatically segmenting racket-swinging sports videos. The method includes: decomposing the racket-swinging sports video stream into video frames and inputting them into a motion recognition model to obtain a confidence set of the athlete's actions on each first-type video frame; inputting the confidence sets of each first-type video frame into a keyframe discrimination model to obtain the action video frame type to which each first-type video frame belongs; segmenting the sports video stream according to the positions of each serve start video frame and rally end video frame in the sports video stream; and combining them to obtain a set of valid rally videos for racket-swinging sports. This invention can automatically segment racket-swinging sports videos, improving the segmentation efficiency of racket-swinging sports videos.

The invention provides a source load coordinated planning method considering multi-energy complementation and direct current delivery, and relates to the technical field of power system planning. The method comprises the following steps: constructing a source-load coordination expansion planning model by taking the minimum total cost of a power system as a target; introducing the confidence set into the source-load coordination expansion planning model, and determining the constraint of the confidence set; establishing a two-stage source load coordination distribution robust planning model based on the source load coordination expansion planning model introducing the confidence set; and solving the source-load coordinated distribution robust planning model by adopting a column constraint uploading algorithm, and generating a planning scheme. The method can give full play to the dynamic regulation and control capability of the hydropower station and the pumped storage power station, and improves the multi-energy complementary benefit; and the power transmission flexibility can be improved by optimizing the power of a direct-current transmission line.

The present disclosure relates to a method, device, medium and equipment for determining battery capacity confidence, capacity and SOH, and relates to the field of vehicle battery technology. The method includes: determining corresponding target operating parameters based on a triggered target algorithm, and then determining a candidate confidence set based on the target algorithm and its corresponding target operating parameters. The candidate confidence set includes the candidate capacity confidence corresponding to the candidate battery capacity calculated by each preset algorithm in the target algorithm, and then determining the target capacity confidence of the target battery based on the original capacity confidence corresponding to the original battery capacity and the candidate capacity confidence corresponding to each candidate battery capacity. The target battery capacity can be determined based on the target capacity confidence, thereby determining a target battery capacity that better matches the current situation based on different battery conditions, so that the original battery capacity can be accurately updated by the target battery capacity to obtain a more accurate battery capacity health status.

This invention discloses a lane line tracking method, apparatus, terminal device, and storage medium. The method includes: acquiring a current set of lane lines and a set of historical lane lines corresponding to each current lane line; performing a lane line confidence subset acquisition action on each historical lane line set to obtain a lane line confidence set; and obtaining a lane line tracking result by combining the current set of lane lines with the historical set of lane lines based on the lane line confidence set. The lane line confidence subset acquisition action includes: for any current lane line, acquiring a set of distances between matching sampling points and a set of angles between matching sampling points for each historical lane line, based on the current lane line and each historical lane line; obtaining a lane line distance confidence based on the set of distances between matching sampling points; obtaining a lane line angle confidence based on the set of angles between matching sampling points; and obtaining a lane line confidence subset based on the lane line distance confidence and the lane line angle confidence to improve the accuracy of lane line tracking.

A computer-implemented method (100) for calibrating that an amount of uncertainty in an industrial process (1a) and / or an electrical network (1b) will exceed a given threshold, the method comprising the steps of: providing (110) an initial desired value; providing (120) a predictive model (2) configured to predict future values of the quantity from the values of the quantity; a history of past values of measurements of the quantity is provided (130). Determining (140), for this history, a difference between a value predicted by the prediction model and a measured past value as a residual (3); determining (150) an estimated probability density function (4) of the residuals (3); determining (160), based at least in part on the history and / or residual (3), a confidence set to which the estimated probability density function (4) belongs, said confidence set comprising a probability density function that is equivalently reasonable given the history and / or residual (3); and determining (170) a new value based at least in part on the dimensions of the confidence set such that if less than is exceeded, it is ensured that the initial desired value is not exceeded even if the residual should be represented according to any probability density function in the confidence set instead of the estimated probability density function (4).