8 results about "Time clustering" patented technology

The application discloses a curtain wall system connecting node stress state recognition method based on deep learning, and the method comprises the following steps: collecting original stress time series data and synchronous environment temperature data of a curtain wall connecting node to form a sample set; based on dynamic time clustering, the original stress time series data of each sample is segmented and divided, and an adaptive feature mapping function combining segmented information and a wavelet base function is used for feature mapping, and then dimension reduction is performed through principal component analysis to obtain a dimension reduction feature vector; a deep learning recognition network is constructed, and a double supervision loss function containing a weighted time series focal loss and an attention consistency regularization term is used to train the deep learning recognition network; and the obtained enhanced feature sequence and external physical features are input into the trained deep learning recognition network to output a stress state category of a node to be recognized. The application realizes automatic and engineering deployable transformation from original multi-source monitoring data to a clear state grade.

This invention discloses a speech signal recognition and separation method based on spatiotemporal clustering of robot voiceprints. It relates to the fields of robot voice interaction and digital signal processing technology. The method includes: a robot voice pickup module acquiring mixed speech signals in complex scenarios and performing preprocessing; extracting voiceprint feature vectors from the preprocessed mixed speech signals and marking human voice feature anchor points; constructing a feature matrix based on the voiceprint feature vectors and human voice feature anchor points, and selecting target human voice signal clusters through weight optimization and automatic clustering; using LSTM adaptive spectrum compensation to perform spectrum correction on the target human voice signal clusters to obtain a clean speech signal; inputting the clean speech signal into a robot speech recognition model and outputting the final recognition result. This invention can solve the problems of sound source number dependence, poor robustness to non-stationary noise, speech feature distortion, and long processing delay in existing technologies, improving the accuracy of robot speech recognition and real-time interaction capabilities in complex scenarios.

The application discloses a kind of multi-contract interactive transaction network modeling method and system based on hypergraph, belong to information security technical field, the application first abstracts four kinds of heterogeneous nodes of user, contract, token and application program, generates feature vector using the layered encoder of fusion static attribute and dynamic timing;Second, based on atomic operation, price influence and other semantic rules, combined with space-time clustering algorithm dynamically generates hyperedge, and the weight is calculated by fusing the multi-factor function of time attenuation, fund size and attack mode;Finally, sliding window is used to construct timing sub-hypergraph, and the global network is updated and matrix representation is generated by union operation.The application accurately describes cross-contract collaborative behavior such as lightning loan by hyperedge, and improves the real-time and accuracy of transaction modeling by space-time feature fusion, providing high-dimensional and accurate data basis for attack detection and vulnerability positioning in the field of decentralized finance.

The application discloses a time-aware adaptive interest point recommendation method based on K-means clustering, which comprises the following steps: first, collecting and sorting check-in data sets, and converting to generate a user-time-location three-dimensional score matrix; second, extracting a two-dimensional check-in score matrix in each time slot, and generating a one-dimensional score vector of each time slot by using a data compression technology; based on the one-dimensional score vector, the K-means method is used to cluster the time slot; third, calculating the dynamic similarity of users in each time slot; based on the time clustering, the score method of the traditional user-based collaborative filtering algorithm is improved, so that the interest point prediction score can be adaptively generated according to the current recommendation time; a plurality of unvisited addresses ranking at the front at the current time are recommended to the user; fourth, the recommendation quality is evaluated by using a recommendation precision index, and the accuracy and effectiveness of the proposed technology are evaluated by comparing the recommendation precision of the technology proposed by the application with that of other classical recommendation systems.

The application discloses a dense single-label multi-target stable tracking method and system based on grid coordinates and space-time clustering, and is applied to the technical field of data processing. The method comprises the following steps: acquiring a video stream, LED detection basic data and equipment inspection parameters, and generating an enhanced image data set by preprocessing the video stream; then, based on an improved YOLO architecture, combining a CBAM attention mechanism and an HSV dynamic threshold, an LED light-off state and an initial coordinate data set are constructed; a grid relative coordinate data set is generated through coordinate system conversion, and an LED physical target clustering data set is obtained through DBSCAN clustering; then, a complete flickering time sequence data set is generated through time sequence completion and binaryzation, a flickering frequency is interpreted by using a zero-crossing detection algorithm, and tracking reliability is evaluated; finally, a light flow method is introduced for anti-jitter compensation, and a dense single-label multi-target stable tracking result and real-time flickering frequency interpretation result are generated through space-time clustering fusion.

The application discloses a kind of based on microseismic space-time clustering's fracturing effect evaluation method and system, belong to hydraulic fracturing technical field, the present application is by introducing the connectivity between fracture cluster, comprehensively consider the spatial distance between fracture, time sequence evolution and energy characteristics, construct the quantization index that can reflect the connection strength of fracture;It also abstracts the relationship between fracture cluster as weighted fracture network diagram, so as to identify the potential through path from the crack region cluster set to target region cluster set, the overall fracture connectivity is obtained by calculating the path connectivity of each connected path, finally based on overall fracture connectivity, the accuracy, reliability and engineering practicability of evaluation result are significantly improved by objectively, quantitatively grading evaluation of fracturing effect.Solved in the evaluation of fracturing effect, unable to consider the time sequence of microseismic event, unable to identify fracture geometry and unable to realize the quantitative evaluation of the whole fracture network fracturing effect problem.

This application provides a multi-task learning model and training method, electronic device, and computer storage medium, relating to the field of multi-task learning technology. The method includes: dividing data according to timeliness to obtain processed data; wherein the processed data includes real-time layer data, short-term layer data, and long-term layer data; inputting the processed real-time layer data into a real-time network for training to obtain real-time feature vectors and corresponding real-time cluster centers; inputting the processed short-term layer data into a short-term network for training to obtain short-term feature vectors and corresponding short-term cluster centers; inputting the processed long-term layer data into a long-term network for training to obtain long-term feature vectors and corresponding long-term cluster centers; and optimizing the real-time network, short-term network, and long-term network using a loss function. This improves the generalization ability and prediction accuracy of the multi-task learning model for data with varying timeliness.

The application discloses a track stream clustering method based on Flink, which converts a track stream clustering problem in a global space into independent subspace real-time clustering problems, maintains dynamic clustering in different subspaces in parallel, and then efficiently combines them. The application also provides a track stream clustering device and an electronic equipment. The method provided by the application can compress stream track points while keeping the amount of space-time information unchanged, reduces unnecessary calculation, improves the throughput of the whole system, and reduces the delay. Meanwhile, the spatial partition clustering of subtracks fully utilizes the parallel computing capacity of a computer, and realizes linear performance scalability on the basis of the Flink stream engine.