11 results about "Linear svm" patented technology

The invention discloses an isoline correction method based on machine learning, and relates to the field of geological data processing and analysis, and the method comprises the steps: data preprocessing: carrying out the preprocessing of known scatter data, dividing the data into a training set and a test set, and converting the training set and the test set into a vector form; feature extraction: carrying out feature extraction on all to-be-corrected isoline points; model training: performing model training on the training set by using a nonlinear support vector machine algorithm, dividing data into a correct class and a wrong class, and finding an optimal hyperplane to separate the data; isoline correction: inputting the feature vectors of the isoline points to be corrected in the test set into the trained model, and judging whether the isoline points are correct or not; correcting the isoline points which are judged to be wrong through an interpolation algorithm; error detection and correction suggestion: detecting errors and inconsistency in the isoline data by using the trained model, providing correction suggestions, and sorting the correction suggestions according to the confidence coefficient calculated by the model.

The application belongs to the field of mechanical fault diagnosis of disconnectors, and provides a disconnector fault diagnosis method and system. The method comprises obtaining real-time vibration signals of the disconnector, and sequentially performing singular value filtering noise reduction and hybrid modal decomposition processing on the vibration signals to extract feature vectors; and performing fault diagnosis on the disconnector according to the feature vectors and a deep weighted fusion model based on an SVM classifier; wherein the construction process of the deep weighted fusion model based on the SVM classifier is as follows: obtaining an initial weak SVM classifier according to the feature vectors of the disconnector vibration signals and a linear support vector machine with a Gaussian kernel as an initial kernel function; and iteratively optimizing the initial weak SVM classifier by means of a deep fusion weighting algorithm and optimal allocation of the weights of the disconnector vibration signal samples to obtain an SVM classifier satisfying a preset condition and serving as the deep weighted fusion model based on the SVM classifier.

Technology is disclosed for a method for screening genetic mutations that can be used to predict vaccine composition, the method may include selecting a plurality of genome samples, partitioning the plurality of genome samples into N groups, where N is an integer larger than 1, identifying genomic isolates with phenotypic statuses from each of the N groups of genome samples by training at least one linear support vector machine with the genome samples, the identification of the isolates between each of the N groups of the genomic isolates performed in parallel, and assessing the identified genomic isolates using a performance metric.

The application discloses a kind of electroencephalogram emotion signal recognition method based on graph regular non-negative matrix decomposition, comprising the following steps: step one, using nearest neighbor method for electroencephalogram emotion signal constructs an adjacent matrix;Step two, establish a graph regular non-negative matrix decomposition model, and the non-negative constraint of matrix after decomposition is carried out;Step three, introduce a projection matrix, further to matrix implement three decomposition;Step four, the model is optimized;Step five, using the new representation PX obtained after the multiplication of trained model parameter P and sample X replaces the representation matrix V of original model, joins corresponding label matrix and trains a classifier in linear SVM classifier;Step six, the class of test sample is predicted by inputting into classifier.The application can effectively identify electroencephalogram emotion signal, compared with other classic electroencephalogram emotion recognition method, the application effectively improves recognition rate.

The invention provides an intelligent cardiovascular disease recognition method based on a multi-scale integrated network and a stacked attention mechanism. The method comprises the steps that electrocardiosignals are preprocessed; establishing five sub-networks, wherein each sub-network comprises a multi-scale scanning module, a multi-angle convolution layer, a secondary convolution layer, a stacking attention mechanism module and a full connection layer; processing is carried out through the five sub-networks respectively; performing decision value evaluation on the features of each self-network through a linear support vector machine; by establishing a multi-power weighted average module, decision values based on the linear support vector machine under all sub-networks are fused to obtain a final classification result. According to the method, the recognition accuracy rates of five types of noiseless electrocardiograms in patients and among patients under normal forms respectively reach 99.6% and 96.62%, the recognition accuracy rate of electrocardiograms containing Gaussian noise exceeds 89%, and the recognition accuracy rate of electrocardiograms containing power frequency interference exceeds 96%.

The invention discloses a log association judgment method and system based on TF-IDF and data balance. According to the method, a unified corpus is constructed through preprocessing, then the preprocessed BUG overview and description are spliced to serve as a query text, log fragments serve as documents, and a feature matrix is generated through TF-IDF vectorization. And further calculating the similarity between the query text and the document, and generating a final feature vector in combination with length weighting and position weighting. The problem of data imbalance is solved through undersampling or oversampling processing, and model training is carried out by using logistic regression or a linear SVM (Support Vector Machine). And finally, performing storage-friendly persistence processing on the trained model, and performing log association judgment on the processed model. According to the method, the accuracy of correlation judgment between the Chinese BUG report and the DMSG log fragment is remarkably improved, the misjudgment rate is reduced, meanwhile, the efficiency of model training and deployment is improved, and the generalization ability and engineering adaptability of the model are enhanced.

The invention relates to the field of aerial remote sensing image target detection and recognition, and discloses an aerial remote sensing image target detection and recognition method based on multi-scale feature fusion, which comprises the following steps: S1, SAR image denoising: based on a denoising auto-encoder, constructing a deep denoising auto-encoder capable of performing feature fusion through training; step S2, a dense scale invariant local feature (SIFT) and a beamlet feature are extracted; s3, performing sparse representation on the features; step S4, feature fusion: based on the SIFT feature points and the beamlet feature points extracted in the step S3, feature fusion is carried out through a feature fusion network, and the feature fusion belongs to feature vectors after fusion of the multi-scale SIFT feature points and the beamlet feature points; and S5, linear support vector machine classification: after the final image of the training image in the step S4 is subjected to vector description, a classifier is trained to carry out SAR image target recognition, SAR image features are extracted through a multi-scale analysis method, and the difficulty of target recognition caused by specific speckle noise of the SAR image is overcome.

The application relates to a recognition method based on voiceprint extraction and multi-index feature screening, and belongs to the technical field of speech processing and machine learning. The method comprises the following steps: pre-processing collected voice signals to obtain effective signals; performing feature extraction on the effective signals to obtain statistical features; calculating the difference degree and the correlation degree of the statistical features; performing multi-index feature screening according to the correlation degree and the difference degree of the statistical features to obtain effective features; and feeding the effective features into a linear support vector machine for classification. The method realizes high classification accuracy.

This invention provides a method for inverting the surrounding rock mechanical parameters of hydraulic cavern groups based on the PO-SVM collaborative optimization algorithm, relating to the field of water conservancy and hydropower engineering technology. This method combines the efficient search capability of the Parrot Optimization (PO) algorithm with the Linear Support Vector Machine (SVM), which can fully utilize the advantages of the PO algorithm in its strong global optimization capability and the efficient data learning capability of the SVM surrogate model, significantly reducing the number of global optimizations and the number of fitness evaluations of the objective function, thereby making the inversion of the surrounding rock mechanical parameters of the refined multi-grid model of hydraulic cavern groups more efficient and reliable.

The application discloses a fire warning method based on a YOLOv7 network and a linear support vector machine SVM, which comprises the following steps: collecting video images about fire; performing target detection on the images by using the YOLOv7 network, and outputting the position of a target bounding box; screening the target bounding box to construct a multi-dimensional feature vector; labeling the images; arranging the labels of all the images and the constructed multi-dimensional feature vectors into a data set, and training and verifying the linear support vector machine SVM to obtain an anomaly detection model; and performing fire detection and warning on the real-time collected video or image by using the anomaly detection model. The application solves the problems of large amount of calculation, difficulty in obtaining samples, false alarm or missed alarm in the prior art.

The present application relates to a kind of based on spatial form feature's towed jamming countermeasure method, comprising: obtaining and channel, azimuth difference channel and the echo data of pitch difference channel, and utilize single pulse three-dimensional imaging technology to the three-dimensional imaging of radar forward-looking area to echo data, obtain three-dimensional point cloud;Utilize the spatial filtering algorithm based on density clustering to cluster clustering and noise rejection of three-dimensional point cloud, obtain several point cloud clusters;Based on the spatial covariance eigenvalue of point cloud cluster, the spatial form feature descriptor of each point cloud cluster is calculated;Based on the difference of target and jamming in spatial form, utilize soft interval linear SVM target discriminator to each point cloud cluster in point cloud cluster sample set Target identification and jamming suppression, obtain the target imaging result after anti-jamming.This method makes full use of the difference of jamming and target in spatial form feature realizes towed jamming countermeasure, jamming discrimination accuracy is higher, and the quality of target imaging result after anti-jamming is higher.