17 results about "Biclustering" patented technology

A single cell sequencing data interpolation method based on a biclustering and graph convolutional neural network comprises the following steps: 1) respectively downloading matrix-form gene expression data of human and mouse from a gene database, and screening out cells of which the total number of gene expressions is less than 3 and genes of which the total number of cell expressions is less than 200; 2) performing feature selection on the gene expression matrix, finding out cells with consistent growth trend, forming a new two-dimensional matrix, and constructing a cell connection diagram; 3) constructing a BiGCN single-cell interpolation model by using a graph convolutional neural network, learning a nonlinear relationship between cells, dividing a data set into a training set and a test set, training the model by using the training set, and reserving the trained BiGCN single-cell interpolation model; 4) inputting the test set into the trained BiGCN single cell interpolation model to obtain an interpolation result, and evaluating the interpolation result by using an evaluation index; the method has the characteristics that the feature extraction capability is improved, interference of redundant zero expression is avoided, data sparsity is effectively relieved, and the interpolation accuracy is improved.

The invention discloses a load clustering method based on load adjustable potential, and particularly relates to the technical field of power regulation and control, and the method comprises the steps: collecting the behavior and resource information of a load user, constructing feature vectors, carrying out the bi-clustering analysis, and extracting a user intersection with the behavior and resource dual matching; high-regulation potential users with high response reliability are screened out after identification; and combining the load pressure, resource saturation and historical scheduling feedback of the power grid, dynamically adjusting clustering parameters by using the trained machine learning model, optimizing the classification boundary and precision, and realizing accurate identification of a load adjustment user and stable support of a power grid scheduling strategy. According to the method, synchronous recognition of user response behaviors and resource controllability is achieved through two-dimensional clustering and intersection screening, the neural network is introduced to achieve dynamic self-adaptive adjustment of clustering parameters, the user screening reliability is improved in combination with a response confidence verification mechanism, and therefore the accuracy, practicability and stability of load adjustment resources are remarkably enhanced.

The invention provides a training method of a classification neural network and a target classification method, and belongs to the field of artificial intelligence and computer science. According to the method, a proper reference solution in a current population is selected by designing a bi-clustering method, the diversity of the population in the optimization process is kept, and the search direction of weight optimization is guided; secondly, a power value conversion function is designed in a problem conversion link, the mapping mode of a decision space is changed, and the original decision space is effectively expanded; and finally, a mechanism for adaptively distributing evaluation times is provided, and the evaluation times of standard optimization and weight optimization can be adaptively adjusted in different stages of population optimization, so that the population convergence can be accelerated in the development stage of population evolution, and the diversity of the population can be balanced in the exploration stage. When the method is applied to neural network model training, the performance of neural network training is effectively improved.

The invention discloses a method for monitoring vegetative propagation growth of roots and stems of bluegrass based on clustering processing. The method comprises the steps of multi-source monitoring data acquisition, multi-source data preprocessing, growth state monitoring clustering model construction, clustering structure optimization and layered clustering growth state monitoring. The invention relates to the technical field of plant propagation data processing, in particular to a method for monitoring vegetative propagation growth of roots and stems of bluegrass based on clustering processing, which innovatively proposes a bi-clustering mechanism based on a main-sub layered structure, and significantly improves the accuracy, stability and agricultural availability of growth state monitoring. A misclassification point recognition mechanism and a misclassification triggering mechanism are designed, and a cluster adjustment factor is introduced, so that the adaptability and the expression ability of a complex plant growth monitoring behavior mode are enhanced; an optimization algorithm and an optimal parameter combination of a growth state monitoring clustering model are improved by adopting a mirror surface reverse mechanism and a sine and cosine function method, and a clustering structure is optimized, so that the accuracy of a growth state monitoring result is improved.

The invention relates to a DUET photovoltaic power generation power prediction method and system based on space-time bi-clustering and dynamic gating fusion, and belongs to the technical field of new energy power system prediction. For the problems of low prediction precision and poor robustness caused by meteorological dynamic disturbance of photovoltaic power, a time clustering module TCM is adopted to extract potential distribution and a linear mode of historical power data, and a time drift error is solved; key control variables such as scattered radiation and temperature are screened through a channel clustering module (CCM) in combination with causal sparse constraint and a frequency domain attention mechanism, and periodic features are enhanced; a bidirectional space-time cross attention and adaptive gating technology of a fusion module FM is utilized to dynamically integrate a time sequence self-correlation characteristic and a cross-channel joint influence. According to the method, the prediction precision in sudden weather change is remarkably improved, the generalization ability and the real-time performance of the model are enhanced, and reliable decision support is provided for power grid dispatching.

According to the anti-noise bi-clustering method and system for the gene expression data, the thought that a seed list is obtained through a longest common subsequence searching method is utilized, the longest approximate common subsequence is searched through an element exchange permitting strategy, interference of noise in a data set on sequence matching can be effectively reduced, and the anti-noise bi-clustering method and system for the gene expression data are provided. The algorithm can still accurately identify the similarity in the gene expression mode in a complex noise environment, so that the accuracy of a bi-clustering result is improved; the binary search algorithm is adopted to determine the insertion interval, the method can fully consider the characteristics of noise data, and compared with a traditional expansion mode, the method can more reasonably process the influence of noise and avoid error expansion caused by the noise, so that the bi-clustering structure is more accurately constructed.

The invention discloses a robust convex biclustering method based on metric learning in the technical field of data mining and machine learning. The robust convex biclustering method comprises the following steps: step 1) initializing variables, data representation and symbol paraphrasing required in a model; 2) constructing a robust convex biclustering model based on metric learning, introducing the metric learning into the model, and reducing the number of non-zero items in a regularization sum by using k-nearest neighbor; the method comprises the following steps of (1) calculating a target function, (2) calculating the target function, (3) optimizing the model by using an alternating direction multiplier method (ADMM), (4) carrying out continuous iteration until the target function is converged, and (5) calculating a final clustering result according to variables and outputting the result.The robustness and optimization stability are improved, the applicability is expanded, and a more effective and reliable solution is provided for the field of clustering analysis.

The invention discloses a convex biclustering method based on robust statistics in the technical field of pattern recognition. The convex biclustering method comprises the following steps: S1, inputting data information into a model; s2, the ADMM model is optimized; s3, performing iterative optimization and convergence judgment; s4, generating a difference matrix; s5, calculating and outputting a clustering result according to label distribution; according to the method, robust statistics are introduced into a convex biclustering objective function, so that the influence of abnormal data on a clustering result is effectively weakened; the problem of distortion of clustering results caused by abnormal values is fundamentally relieved, the practical application range of the convex biclustering method in high-noise, large-scale and strong-interference environments is expanded, and the convex biclustering method has important theoretical value and practical significance; through reasonable decomposition and an ADMM optimization framework, a common local optimal trap problem in a traditional non-convex biclustering method is avoided, and an optimal solution can be stably converged; and the accuracy and reliability of the clustering result are ensured.

The invention provides an abnormal transaction customer detection method and apparatus, and a computer program product. The method comprises the steps of obtaining effective data transaction matrixes corresponding to all target customers; inputting the effective data transaction matrix into a bi-clustering proximity model, and performing anomaly detection on the effective data transaction matrix by using the bi-clustering proximity model to obtain all abnormal transaction clients; and executing corresponding exception handling strategies for all the exceptional transaction clients. According to the method and the device, the problem that the accuracy of an abnormal detection result is relatively low due to a misjudged scene caused by the fact that frequent suspicious transaction clients with similar behaviors are mined without considering regular behaviors of part of normal clients is solved.

The present invention discloses a method for monitoring the asexual reproduction and growth of rhizomes of bluegrass based on clustering processing, and the method includes multi-source monitoring data acquisition, multi-source data preprocessing, construction of a growth status monitoring clustering model, clustering structure optimization and hierarchical clustering growth status monitoring. The present invention relates to the technical field of plant reproduction data processing, and specifically refers to a method for monitoring the asexual reproduction and growth of rhizomes of bluegrass based on clustering processing. This solution innovatively proposes a dual clustering mechanism based on a master-sub hierarchical structure, which significantly improves the accuracy, stability and agricultural usability of growth status monitoring; designs a misclassification point recognition mechanism, a misclassification trigger mechanism and introduces a cluster adjustment factor to enhance the adaptability and expression ability of complex plant growth monitoring behavior patterns; adopts a mirror inversion mechanism and a sine-cosine function method to improve the optimization algorithm, the optimal parameter combination of the growth status monitoring clustering model, optimizes the clustering structure, and improves the accuracy of the growth status monitoring results.

The application discloses a high-frequency oscillation rhythm detection method and system based on an adaptive fuzzy double clustering algorithm, and relates to the medical treatment technical field. The particle swarm optimization algorithm is a kind of random optimization algorithm based on group, has high requirements for the initial position of the solution, has strong global search capability, and is not easy to fall into local optimal value, and has strong processing capability for complex samples with large quantity, so that the particle swarm optimization algorithm is used to optimize the fuzzy double clustering algorithm, and the optimized clustering center can be obtained, and the high-frequency oscillation rhythm of the epilepsy electroencephalogram signal can be detected based on the optimized clustering center. The application can improve the detection precision of the high-frequency oscillation rhythm of the epilepsy electroencephalogram signal, and help doctors to diagnose epilepsy and remove the epileptogenic focus.

The invention provides a gene expression profile bi-clustering method based on a swarm intelligence algorithm, and belongs to the technical field of gene expression, and the method comprises the following steps: constructing an index matrix for a gene expression data matrix, obtaining a bi-clustering seed, initializing a population, taking the bi-clustering seed as a first population individual, and constructing a second population individual; adding the first population individuals into a second population for population expansion to obtain two population individuals, evaluating the first population individuals by using a fitness function, and evaluating a plurality of individuals which contribute to the second population to the greatest extent by the individuals in the first population until a preset maximum number of evolution times is reached; and selecting an optimal bi-cluster from the second population. The index matrix is constructed for the gene expression data matrix, so that later data processing is simpler and higher in efficiency, meanwhile, data values of each gene under each condition can be searched by using the swarm intelligence algorithm, and therefore, the search range is wide, and important biological information is not easy to miss.

The invention discloses a microscopic image core semantic text generation method in a bi-clustering process, and belongs to the technical field of medical image processing. The method comprises the following steps: acquiring a general semantic set through a large language model, and processing microscopic images to obtain a microscopic target image set of each category; preprocessing the text and the image, inputting the preprocessed text and image into a trained image-text model, generating text and image representation vectors, and constructing a category-semantic similarity matrix through cosine similarity calculation; extracting different similar sets from the matrix, and respectively performing intra-class and inter-class clustering to obtain generalization and discrimination semantic text sets of all classes; and taking an intersection of the two types of sets to obtain a core semantic text. According to the method, accurate screening of microscopic image semantics is realized through a biclustering strategy, the generated core semantic text has generalization and discrimination, the semantic alignment quality is improved, and high-quality semantic data support can be provided for fine tuning of a multi-modal model.

This application discloses a method and device for biclustering popular recommendations based on user behavior sequences. The method includes: constructing a network graph of items based on user behavior sequence data; obtaining a vector representation of each item node using a graph embedding method; clustering the vectors of the item nodes into K clusters; dividing the items on which the user has performed actions into M sets according to the clusters, and obtaining vector representations of each user's M clusters; calculating the user's preference score for each cluster; taking a weighted average of the user's vectors under each cluster based on the user's cluster preference scores to obtain a unique interest vector representation of the user; further clustering the user's interest vectors to obtain a cluster for each user; statistically analyzing user behaviors under each cluster to form a cluster hot list; and pushing the hot list of each cluster to each user. This application uses a biclustering algorithm, resulting in high recommendation accuracy and improved recommendation effectiveness.

This invention discloses a method for generating core semantic text from microscopic images using a dual-clustering process, belonging to the field of medical image processing technology. The invention obtains a general semantic set through a large language model, processes the microscopic images to obtain sets of microscopic target images for each category, and then inputs these preprocessed sets into a trained image-text model to generate text and image representation vectors. A category-semantic similarity matrix is ​​constructed using cosine similarity calculation. Different similarity sets are extracted from the matrix, and intra- and inter-class clustering is performed to obtain generalized and discriminative semantic text sets for each category. The intersection of the two sets yields the core semantic text. This invention achieves accurate semantic filtering of microscopic images through a dual-clustering strategy, and the generated core semantic text possesses both generalization and discriminative properties, improving semantic alignment quality and providing high-quality semantic data support for fine-tuning multimodal models.

The invention discloses an underwater target tracking classification method and system. The tracking classification method comprises the following steps: S3, carrying out ROI region identification, and identifying a potential target region by using an FCM and K-means biclustering algorithm; s4, generating a high-overlapping-degree ROI mask, and calculating the intersection-to-union ratio of an FCM result and a K-means result; s5, target segmentation is carried out, and target segmentation is realized in combination with FCM pre-segmentation and a PCNN neural network; s6, performing target detection, analyzing and detecting the segmented target based on a connected region, and extracting static features of the target; s7, multi-target tracking is carried out, and target tracking is realized based on an improved SORT algorithm; and S8, target intelligent classification is carried out, static features and dynamic features of the target are fused, and classification and identification of the underwater target are realized. According to the tracking classification method and system, accurate detection, stable tracking and automatic classification of underwater biological targets can be realized.

The invention discloses a robust convex biclustering method in the technical field, and the method comprises the following steps: 1) data representation and symbol paraphrasing: dividing pollution data into the sum of clean data and sparse abnormal values; 2) constructing a robust convex biclustering model, and adding a Lasso type regularization item in an objective function of the robust convex biclustering model; 3) calculating a weight, ensuring that all data points contribute in calculation by using the weight, and reducing the number of non-zero items in a regularization sum by using k-nearest neighbor to simplify calculation; the method comprises the following steps of (1) optimizing a target function according to a variable V1 and a variable V2, (2) optimizing the target function according to the variable V1 and the variable V2, (3) using an alternating direction multiplier method ADMM to optimize the model, and carrying out variable updating according to an ADMM rule, (5) continuously iterating until the target function is converged, and (6) calculating a final clustering result according to the variable V1 and the variable V2 after convergence, and the method can better adapt to a complex and changeable data environment, and provides a firmer and more reliable support for data analysis and decision making.