8results about How to "Improve clustering quality" patented technology

The invention provides a user intention mining processing method and device and related equipment, and aims to solve the problem of low user intention mining efficiency when self-service voice service intelligent equipment constructs a dialogue system in the traditional technology. The user intention mining processing method comprises the following steps: obtaining an initial to-be-mined text of a to-be-mined user intention, and according to a TF-IDF value, obtaining an initial to-be-mined text of the to-be-mined user intention; the method comprises the steps of obtaining word weights of corresponding lexical units, filtering lexical unit sentence texts corresponding to the lexical units according to the word weights to obtain target lexical unit sentence texts, vectorizing the target lexical unit sentence texts, determining target lexical unit sentence text vectors, clustering all the target lexical unit sentence text vectors, and determining target user intentions according to clusters. According to the user intention mining method and device, repeated corpora and corpora low in importance degree can be removed by filtering the lexical text corresponding to the lexical, the data scale is reduced, the user intention mining efficiency and quality are improved, the user intention mining effect is improved, and the user intention mining method and device can be applied to the field of medical health.

The embodiment of the invention provides a clustering method and device of multiple view data, equipment and a storage medium, and relates to the technical field of data processing. The method comprises the following steps: acquiring view data corresponding to a plurality of views, projecting the corresponding view data to a low-dimensional space based on a projection matrix of the view for each view to obtain a subspace representation matrix and an error matrix, and calculating to obtain local total information according to the subspace representation matrix and a weight matrix, and constructing an optimization target item according to the nuclear norm of the subspace representation matrix, the preset norm of the error matrix and the local total information, solving an optimization target function to obtain a representation optimization matrix corresponding to each view, and performing data clustering according to all the representation optimization matrixes to obtain at least one clustering result. In the dimension reduction process, subspace representation is synchronously carried out, in the process of optimizing the target function, mutual promotion and co-evolution of the projection matrix and the subspace representation matrix are achieved, the accuracy of subspace representation is improved, and the clustering quality is improved.

The application relates to the field of intelligent electric meter detection, in particular to an intelligent electric meter power consumption anomaly detection method and system based on mode classification, which comprises the following steps: obtaining historical current sequences and to-be-detected current sequences of each user, performing multi-scale equal-length adaptive segmentation and frequency band division on the historical current sequences, combining user intra and inter power consumption mode differences to calculate power consumption tolerance of each frequency band, performing EMD decomposition and feature extraction on the historical current sequences, calculating mode classification distances based on power consumption tolerance and clustering to obtain power consumption mode clusters, and finally comparing distances between the to-be-detected sequences and the mode clusters, calculating anomaly scores and determining power consumption anomaly states. Through multi-scale equal-length adaptive segmentation, frequency band differential analysis on the historical current sequences, and combination of user intra and inter power consumption mode differences, tolerance and anomaly scores are constructed, precise and personalized anomaly detection of user power consumption is realized, and detection accuracy and efficiency are improved.

The present invention relates to the technical field of blood pressure monitoring, and in particular to a method and system for monitoring blood pressure of patients for clinical care. The method obtains the controllability of blood pressure by the distribution fluctuation degree of blood pressure data in the neighborhood range at each moment; obtains the blood pressure disorder by the distribution of blood pressure data corresponding to the sampling moment at the same position in different period segments and the blood pressure controllability; obtains the blood pressure disorder at each sampling moment by the blood pressure disorder in the period segment at the current moment and the blood pressure change in the neighborhood range, obtains the blood pressure disorder based on the distribution of blood pressure disorder, and performs DBSCAN clustering on the blood pressure data in the period segment at the current moment to obtain the monitoring clustering result for blood pressure status monitoring. The present invention combines the blood pressure fluctuation situation and analyzes the abnormality of blood pressure data through multiple periods to adaptively adjust the neighborhood radius of clustering, improve the sensitivity of clustering to abnormal points, obtain more accurate clustering results, and make the monitoring results more reliable.

The invention discloses a charging pile network attack behavior determination method and apparatus, and an electronic device. The method comprises the following steps: acquiring charging pile data in a charging pile system; extracting charging features corresponding to the charging pile data; the corresponding distances between the charging features and multiple target clustering center features are determined, multiple target distances are obtained, the multiple target clustering center features are obtained according to multiple initial clustering center features, and the multiple initial clustering center features are obtained according to subsample features included in the corresponding updating subspace; the plurality of updated subspaces are subspaces which are determined from the plurality of initial subspaces and have corresponding sample densities greater than a first threshold value; and determining an attack behavior category corresponding to the charging pile system according to the plurality of target distances. According to the method and the device, the technical problem of low attack detection accuracy when a traditional clustering detection method is used for detecting whether the charging pile network is attacked or not in the prior art is solved.

The application discloses a power system load curve clustering analysis method, determines the target and requirement of power system load curve clustering analysis, carries out data preprocessing, uses a self-organizing feature mapping neural network to construct a feature extraction model, uses an MAML algorithm to carry out multi-task learning, carries out rapid adjustment through meta learning, applies a Gaussian mixture model to cluster the extracted features, uses indexes to evaluate the clustering effect, carries out clustering result analysis, determines whether the predetermined performance standard is met, and further carries out iterative optimization, so that the self-organizing feature mapping neural network, meta learning and the Gaussian mixture model are combined, and greater value is brought to power system operation and energy management.

The application provides a power user power consumption behavior portrait method considering power consumption sensitivity, and belongs to the technical field of industrial data processing. The method solves the problem that in most power user portrait methods, sensitive loads and basic loads influenced by other factors are mixed for analysis, which makes the portrait result not fine enough and unable to accurately grasp the power consumption behavior characteristics of users. The method comprises the following steps: obtaining a power load data set, drawing a load curve, and performing preliminary clustering; labeling the preliminary clustering cluster; inputting the preliminary clustered load data set into an STL model for decomposition and division into basic loads and sensitive loads; performing clustering analysis on the basic loads and the sensitive loads respectively, defining a basic load label library and an external factor sensitive load label library according to the clustering results; and labeling each user with two types of labels according to the basic load label library and the external factor sensitive load label library, combining the labels of the initial clustering results, and generating a precise portrait of the power user.

The application provides a brain function connection map construction method based on non-negative matrix three-factor orthogonal decomposition (ONMTF). The method first establishes a group level function similarity matrix between two brain region voxels of each subject in a resting state functional magnetic resonance imaging data set according to a function connection measurement between the two brain regions, wherein the rows and columns of the matrix represent the voxels of the two brain regions. Then, the group level function similarity matrix X is decomposed into the product of three non-negative matrices, namely ASY, through ONMTF. According to the two posterior probability label matrices A and Y obtained through the decomposition, a refined function connection network between the two brain regions is obtained. The function of the unknown subregion network extracted by the method can be inferred through the function of the subregion network which has been studied thoroughly. Moreover, the brain function connection network extracted by the method has better clustering quality, namely better function consistency.