117 results about "Minority class" patented technology

The invention discloses a federated learning method based on partial label mask weighted distillation, which comprises the following steps of: constructing a federated learning system comprising a central server and a plurality of clients, and broadcasting a global model to each client by the central server; the client performs calculation by using the received global model to obtain a teacher model, regards a local model as a student model, performs distillation training on the local model by using local data, and uploads trained model parameters to the central server; and the central server performs federal aggregation on the updated local model parameters, and issues the aggregated and updated local model to each client for next round of local training. According to the method, the client side learns knowledge of the global data set in the teacher model through partial label mask distillation, learning of minority class samples in the local data set is enhanced, learning of the local model on the knowledge of the global data set is enhanced by constructing the teacher model with more complete knowledge, and the learning efficiency of the local model is improved. And the risk of disastrous forgetting caused by data isomerism is reduced.

Class imbalance in a training dataset may negatively impact the accuracy of a machine-learning model in classifying rare events that are underrepresented in the training dataset. Training datasets comprising time-series data present a unique challenge. Accordingly, resampling techniques for up-sampling and / or down-sampling a training dataset of time series are disclosed. The up-sampling may respect the temporal correlation of time samples in the time series, while generating synthetic time series that mimic the feature values of time series belonging to the minority class. Down-sampling may be used to fine-tune the ratio of time series belonging to the minority class to the time series belonging to the majority class.

The invention discloses an anti-fact generation method for processing class imbalance based on a real sample. The method comprises the following steps: preprocessing input data; performing causal feature selection by adopting a causal discovery algorithm; calculating causal feature tendency scores, and performing matching; carrying out anti-fact generation, forming a synthesized minority class set, and integrating the synthesized minority class set with original data to obtain an enhanced data set; and performing data cleaning on the enhanced data set to obtain a balanced data set. According to the method, a data set is effectively balanced by generating a high-quality and close-to-reality anti-fact sample, so that the performance of a downstream classifier on key indexes is remarkably improved; the feature values of the real instances are combined to ensure that the generated samples are located in a reasonable area of data distribution, so that the credibility and availability of the enhanced data are improved; the generated anti-fact sample is located in a boundary region between the majority class and the minority class, the decision region of the minority class is effectively expanded, and the unique post-cleaning avoids the influence of noise accumulation on model training.

The invention relates to the technical field of machine learning, and provides a sample generation method and device, computer equipment and a readable storage medium, and the method comprises the steps: obtaining a to-be-generated original data set, and separating the original data set into a minority class sample set and a majority class sample set; for the minority class samples, by calculating a contour coefficient under each clustering number, determining an optimal clustering number enabling the contour coefficient to be maximum, and clustering the minority class samples based on the optimal clustering number to obtain a plurality of minority class sample clusters so as to generate a plurality of candidate samples; and respectively acquiring a first average nearest neighbor distance from each candidate sample to the minority class sample and a second average nearest neighbor distance from each candidate sample to the majority class sample, so as to determine the target sample as a newly generated minority class sample. According to the method, samples conforming to a real fraud mode can be accurately generated in financial fraud detection, rare disease feature distribution can be effectively simulated in medical diagnosis, and the early screening accuracy is improved.

The invention relates to the technical field of unbalanced data processing, in particular to an oversampling method, system and device for avionics equipment test data and a medium, and the method comprises the steps: dividing original avionics equipment test data into minority class samples and majority class samples; respectively clustering the two types of samples by using a density peak value clustering algorithm, redistributing a new label according to a cluster to which each sample belongs, removing noise samples, and combining into a new unbalanced training set; constructing a sample generation model, and training the sample generation model with the new unbalanced training set until the sample generation model is converged to an equilibrium state; and determining the number of new and minority class samples required to be generated, and inputting the minority class samples in the new unbalanced training set into the trained sample generation model to obtain the required number of new and minority class samples. The method can highly fit and approach the real distribution of minority samples, enhance the comprehensiveness, diversity and fidelity of the generated samples, and effectively improve the performance of the sample generation model.

The invention provides an intrusion detection method based on boundary sensitive federated expert multi-modal detection, which comprises the following steps of: splicing, synthesizing and fusing seven types of discriminative characteristics based on original traffic characteristics, designing a multi-modal collaborative attention model MultiModalFusion, dividing the characteristics into four modals, namely a protocol state, a traffic behavior, statistical distribution and a connection relationship, and realizing cross-modal information interaction by utilizing dynamic weight learning. In order to solve the problem of data imbalance, a boundary sensitive condition generator BSGenemator is developed to guide generation of minority class samples through a dynamic boundary strategy in combination with a composite loss function method. And finally, constructing a federal element strategy expert committee, dynamically fusing decisions of four experts by adopting a learnable strategy network, and verifying the characteristic contribution degree through an SHAP interpretable module. And finally, the efficiency of the scheme is verified by using a data set UNSW-NB15, through comparison of multiple schemes, the scheme has significant accuracy, the weighted average F1 score is improved, and a new normal form is provided for a real-time intrusion detection scheme.

The invention relates to a dynamic density guided generative adversarial network unbalanced data processing method (DAG-WGAN). The DAG-WGAN realizes unbalanced data processing through data preprocessing, dynamic density estimation and weight distribution, potential structure learning based on a variational auto-encoder (VAE), and density guide generation and dynamic feedback optimization based on WGAN-GP. The DAG-WGAN adaptively evaluates the sample density by using kernel density estimation (KDE) and a Gaussian kernel function, and allocates a weight for a generation process, thereby emphatically enhancing the low density and discriminating the sample generation of a difficult region. The VAE learns a potential manifold structure of a minority class of samples, realizes density-guided generation of a potential space under a WGAN-GP framework, and ensures diversity and manifold consistency of generated samples. In addition, a dynamic feedback mechanism is introduced, the weight and the gradient penalty coefficient are adaptively adjusted and generated, and the training stability and the sample generation robustness are improved.

The invention discloses a data classification method and system based on hierarchical clustering and resampling, and a storage medium. The method comprises the following steps: dividing samples in a data set into a minority class and a majority class; one part of samples in the minority class are distributed to a training set, and the rest part is distributed to a test set; dividing samples in the training set into a plurality of clusters to form a corresponding number of sub-training sets; training a test set classifier by using the cluster labels of the sub-training sets, and classifying the test set to form sub-test sets; the samples in the clusters are resampled, so that the proportion of minority class samples to majority class samples in the training sub-sets is 1: 1, and an enhanced training sub-set is formed; and training a data classifier by using the enhanced sub-training set, and testing by using the sub-test set. According to the method, the inter-class overlapping degree is reduced through the clustering algorithm, the sample proportion is balanced through the resampling technology, a classifier is made to pay more attention to minority class sample features, the recognition capacity of minority classes such as financial abnormal conditions is enhanced, and the misjudgment risk is reduced.

The invention relates to the technical field of machine learning and data mining, in particular to an unbalanced data-oriented multi-agent reinforcement learning classification method, which comprises the following steps of: firstly, introducing a Markov decision process to construct a classification model based on multi-agent reinforcement learning, the environment provides feedback to the intelligent agent according to the classification result, so that the intelligent agent can learn the unbalanced sample; secondly, different rewards are set for the majority class samples and the minority class samples respectively, a dual deep neural network is used for calculating accumulated rewards, the training process is played back and optimized through priority experience, and learning of the minority class is improved; and finally, intelligent agent centralized training and decentralized execution are combined with distributed calculation, and intelligent agent interaction and learning efficiency is optimized. Compared with an existing unbalanced data classification method, the method has the advantages that the classification accuracy of minority classes can be effectively improved, and the multi-agent cooperative classification time is shortened.

The invention discloses an unbalanced node classification method based on graph contrast learning, and belongs to the technical field of artificial intelligence. According to the method, a graph comparison learning framework of adaptive balance data is provided, minority classes can be automatically identified, the minority class performance is improved, and then the overall performance of the model is improved. Firstly, an Encoder-Decoder architecture is used for pre-training, and compared with a traditional pseudo tag generation method, an unbalance rate self-adaptive sampling strategy is designed, the unbalance rate of data is calculated according to pseudo tags, and the sampling strategy is selected in a self-adaptive mode. For a data set with a low unbalance rate, a simple downsampling method is adopted, and the proportion of minority class information is increased; for a data set with a relatively high unbalance rate, a mixed sampling strategy is adopted, and over-sampling and down-sampling are combined, so that the information loss of majority of nodes is reduced while the information proportion of minority of nodes is increased. In addition, the pre-training model used in the invention can provide more accurate label information, thereby improving the distinguishing ability of the model in subsequent GCL training. Then, a new data augmentation technology is designed, in the node masking process, pseudo label information is utilized, information of minority class nodes is reserved preferentially, and meanwhile majority class nodes are masked; the method is helpful for the model to better capture minority class features in an unbalanced data set. And finally, a linear classifier is used for classification. According to the method, the unbalanced node classification performance under the self-supervision condition can be effectively improved.

The invention discloses a customer loss prediction method based on oversampling classification, and the method comprises the steps: obtaining a customer loss data set, carrying out the preprocessing, carrying out the data clustering, selecting and marking a cluster with a dominant minority of samples, calculating a sampling weight value, calculating the membership degree of the samples in the cluster with the sampling weight value larger than 0, and carrying out the space division, and selecting a sample according to a region division result, executing linear interpolation oversampling to generate a balanced data set, training a classifier, preferentially constructing a final customer loss prediction model, and outputting a customer loss prediction category and probability. According to the method, the customer loss data set is balanced through an oversampling method of clustering and membership region division, the safest samples are selected for synthesis, the risk of introducing noise and fuzzy boundaries is effectively reduced, data category distribution is balanced by generating high-quality minority-class samples, and the accuracy of data classification is improved. And therefore, the recognition performance of minority class samples can be improved, and the accuracy of customer loss prediction can be improved.

The invention discloses a Tokamak plasma rupture prediction method based on cWGAN and TCN, which takes AUC as a measurement index to evaluate the effectiveness of a model, and adopts a conditional Wasserstein generative adversarial network (cWGAN)-based minority class sample generation method for solving the problem that the number of minority class samples is limited. Gradient penalty (GP) is introduced in a training stage to improve the quality of generated samples and training stability, and an inverse frequency mini-batch minority class equilibrium training strategy is adopted to enable a generator to fully learn distribution characteristics of broken samples, so that high-quality minority class samples are generated and a training set is expanded. A Transform structure is introduced on the basis of a time convolutional network (TCN), local time sequence features are efficiently extracted by using the TCN, and the Transform enhances the expression ability of a model for a long-time dependency relationship and global features, so that the prediction effect is improved.

The application discloses a network intrusion detection method based on LSTM and an attention mechanism, and comprises the following steps: 1) data preprocessing and target sample construction; 2) WGAN model establishment and new training sample merging; 3) network intrusion detection model establishment based on LSTM and the attention mechanism; 4) training in the obtained network intrusion detection model; and 5) model performance evaluation. The method can improve the detection accuracy and the detection rate of minority class samples.

The invention relates to the technical field of oil-immersed transformer fault diagnosis, in particular to an oil-immersed transformer fault diagnosis method and system based on feature enhancement and deep isomerism. The method comprises the steps that a minority class oversampling method is used for synthesizing minority fault sample data into new samples conforming to physical constraints; an enhanced input space is constructed based on deep features generated by an auto-encoder VAE, and deep features of fault gas in a new sample are fully extracted; based on the extracted deep features, using a heterogeneous integrated model to simulate a lightweight gradient elevator Light GBM to capture a shallow relationship of the deep features; using a convolutional neural network 1D-CNN to identify local features of the fault gas; establishing a global dependency relationship on the basis of a Transform model; the robustness of the model is remarkably enhanced through a dynamic weighting mode, and therefore it is guaranteed that reliable and stable diagnosis results are continuously output in various complex and high-uncertainty practical application scenes.

The invention discloses a network intrusion detection method for enhancing minority class data identification capability, and belongs to the field of artificial intelligence and network security. The network intrusion detection model comprises three main components: an improved table-based generative adversarial network structure, a self-encoder structure combined with a self-attention mechanism, and a parallel one-dimensional convolution and bidirectional gating loop unit structure. Wherein the improved table generative adversarial network structure is used for enhancing minority class data in original data, and the self-encoder structure combined with the self-attention mechanism is used for performing dimension reduction on original traffic data. And finally, training the parallel one-dimensional convolution and bidirectional gating cycle unit structure by using the enhanced minority class data and the original data, thereby achieving the purposes of reducing the data dimension, accelerating the model convergence, enhancing the recognition capability of the model for the minority class data and improving the overall recognition precision of the model.

The application discloses a text sentiment analysis method based on class imbalance data. The method comprises the following steps: extracting semantic information of text by using a pre-trained classifier for a first data set, and then obtaining a sentiment classification boundary corresponding to each text, wherein the sentiment classification boundary is used to define a majority class and a minority class in the first data set; based on the classification boundary, generating pseudo samples of the minority class by using feature information in the majority class, and iteratively adjusting the classification boundary of the classifier by taking a set rejection function as an optimization target, wherein the rejection function is used to measure the quality of the generated pseudo samples; adding the generated pseudo samples to the first data set to form a second data set, and continuing to train the classifier by using the second data set. The application obtains hidden information of classification from data of the majority class, and performs oversampling on the minority class data according to the obtained classifier, so that a data class balancing strategy is realized.

The invention discloses an unbalanced classification method for electroencephalogram data in epilepsy detection, and relates to the technical field of data optimization and big data processing. According to the method, firstly, covariance matrixes of majority class samples and minority class samples are calculated, linear transformation is achieved through matrix decomposition, and the minority class samples inherit global distribution characteristics of the majority class samples; and then, in the transformed feature space, sorting samples based on mahalanobis distance and performing partition pairing, selecting sample pairs with large difference to generate convex combination synthesis samples, and ensuring sample diversity and boundary consistency. Experimental results show that on a CHB-MIT electroencephalogram data set, the method effectively solves the problems that a traditional oversampling technology is prone to expanding minority class decision boundaries and generated samples are lack of diversity, and the reliability of epileptic seizure detection is remarkably improved.

The invention discloses a short text unbalanced classification method based on semi-supervised clustering and dynamic resampling, which comprises the following steps of: S1, performing cleaning and feature extraction on original short text data, removing features of which the missing rate exceeds 20%, performing vectorization expression on residual texts, and constructing an initial feature space; s2, under a'compact cluster 'assumption, based on the guidance of marked data, performing iterative segmentation on majority class and minority class samples by adopting a semi-supervised hierarchical clustering algorithm, generating a plurality of discontinuous clusters, and revealing inter-class and intra-class unbalanced distribution characteristics; s3, based on a clustering result, performing dynamic under-sampling on majority of samples; s4, based on a clustering result, carrying out dynamic oversampling on minority class unmarked data; according to the method for introducing semi-supervised clustering into short text unbalanced classification for mixed sampling, the marked data and the unmarked data are clustered, the basic distribution characteristics of the short text data are effectively captured, and follow-up sampling is facilitated.

The invention discloses a user behavior analysis method and system.The method comprises the steps that on the basis of normal behavior data of a user, frequent item sets and rare item sets of normal behaviors are mined from mass unlabeled behavior data, and an accurate basis is provided for data classification; a rare item set is used as a standard to divide majority class and minority class data sets, and minority class data is anchored based on rare features, so that the processing accuracy is improved; performing N times of conversion on the majority class data set based on the rare item set, classifying the converted data into a minority class data set, dynamically balancing the two classes of data, and improving the sensitivity of the model to a minority class normal behavior mode; and constructing a user behavior graph and a graph neural network learning model based on the balanced training data, carrying out structured representation on a user complex behavior pattern, inputting graph neural network learning to obtain a graph structure, nodes and edge features of a normal behavior pattern, inputting actual user behavior data into the trained model, and outputting an accurate user behavior analysis result.

The application provides a perioperative patient sample dataset balancing method and a sample dataset acquisition system. The sample dataset balancing method comprises the following steps: S1, oversampling minority class label samples in a sample dataset of a perioperative patient to obtain synthetic samples, and generating a corresponding synthetic label set for the synthetic samples; the sample dataset comprises a plurality of samples and a sample corresponding classification label set; S2, adding the synthetic samples and the synthetic label set to the sample dataset to obtain a temporary sample dataset; and S3, cleaning the samples in the temporary sample dataset to obtain a balanced sample dataset. The minority class label samples in the sample dataset are oversampled to increase the number of minority class label samples, the majority class label samples and the minority class label samples are balanced, the noise samples are cleaned to improve the sample quality in the output balanced sample dataset, and the balanced sample dataset can improve the performance of a classification model during subsequent classification processing.

The invention provides a data enhancement method and device based on a variational auto-encoder and a generative adversarial network, and relates to the technical field of data enhancement. Iteratively updating a variational auto-encoder and a dual generative adversarial network by using a back propagation algorithm until the model converges, enhancing a non-equilibrium industrial data set by using an optimal data enhancement model, extracting key features of minority class samples through an encoder and a variational representation layer, and extracting the key features of the minority class samples. Reconstructing the extracted key features into samples by using a first decoder module, inputting labels of minority class samples and random noise vectors into a generator module to obtain fused feature variables, inputting the fused feature variables into a second decoder module, outputting generated new samples, mixing the reconstructed samples, the new samples and original samples, and outputting the reconstructed samples, the new samples and the original samples. According to the method, the authenticity and diversity of the generated data are improved, and the data enhancement quality is improved.

The application discloses a kind of class imbalance table data processing methods based on contrast constraint diffusion generation, belong to data processing technical field, it includes using binary classification table dataset to generate model training, obtain the generation model based on noise generation minority class's synthetic latent representation, generate model is randomly generated with several noise inputs, diffusion modeling branch is denoised to each noise and obtains the synthetic latent representation of minority class of each noise, using decoding module is decoded to each synthetic latent representation, obtains synthetic table sample;Data filtering is carried out to all synthetic table samples, then all synthetic table samples after filtering are merged with table dataset, and enhanced imbalance table data are obtained.The application introduces contrast learning branch in the process of generating minority class sample, effectively overcome the deficiency existing in class imbalance table data processing of existing oversampling method and general generation model.

An ensemble learning based method is for a binary classification on an imbalanced dataset. The imbalanced dataset has a minority class comprising positive samples and a majority class comprising negative samples. The method includes: generatively oversampling the imbalanced dataset by synthetically generating minority class examples, thereby generating a generated dataset; using the generated dataset to generate subsamples, and learning a base classifier on each of the subsamples to determine a plurality of base classifiers; and learning a weighted majority vote classifier by combining outputs of the base classifiers. Each of the base classifiers is assigned a weight in such a way that a diversity between the base classifiers on the positive samples is minimized.

The present invention relates to a system call sequence data enhancement method for host intrusion detection, which belongs to the field of computer and cyberspace security. The present invention first uses minority class samples to construct a system call sequence pair database; then converts the sequence data into graph structure data to better characterize the relationship between system calls; then uses a graph conversion network based on a multi-head attention mechanism and an LSTM network to respectively construct a graph encoder and a sequence decoder to learn the temporal dependencies contained in the behavioral semantics of the system call sequence; finally, based on the system call temporal dependencies, a minority class system call sequence samples are cyclically generated to achieve data enhancement of the minority class samples. The present invention can make up for the problem that the existing system call data enhancement method does not fully exploit and utilize sequence information, resulting in the semantic inconsistency between the generated sequence and the original sequence, and generates enhanced samples that reflect the sample sequence relationship, effectively improving the detection accuracy of host intrusion detection.

The invention discloses a credit data synthesis oversampling method and system based on noise-free loss distribution migration, and relates to the technical field of unbalanced classification in artificial intelligence and data mining, and the method comprises the steps: obtaining an original credit unbalanced data set; noise label sample filtering is carried out on the data set, the prediction probability of each sample in the data set is predicted through a model, and the noise-free loss value of each sample is calculated; dividing a plurality of loss intervals according to the value range of the noise-free loss value, distributing each sample to the corresponding loss interval, and determining noise-free loss distribution of majority-class and minority-class samples; migration is carried out based on noise-free loss distribution, synthetic sample distribution is determined, root samples and auxiliary samples are screened out, and minority class pseudo samples are synthesized through linear interpolation; and adding the minority class pseudo samples into the original credit unbalanced data set to obtain a class balanced credit data set. According to the optimized balanced data set, the recognition precision of majority-class samples and minority-class samples can be effectively improved.

This invention discloses a rotating machinery fault diagnosis method based on a conditional decoupled representation enhancement network. Addressing scenarios involving class imbalance and cross-domain fault diagnosis in rotating machinery, the method utilizes a deep convolutional feature extractor to capture fault features from both the known source domain and the unknown target domain in the acquired vibration signals. A decoupled representation enhancement module generates high-quality class semantic information to enhance minority class features in the high-dimensional feature space, thereby balancing the data distribution between classes and mitigating the impact of differences in the number of samples from different classes. Subsequently, a conditional discriminant contrastive loss is used to increase attention to the minority class and strengthen the compactness of intra-class features across domains. Finally, the desired imbalanced domain generalized mechanical fault diagnosis model is formed. This method not only overcomes the influence of class imbalance but also improves the generalization ability and robustness to unknown target domains, effectively enhancing the accuracy of rotating machinery fault diagnosis.

Class imbalance in a training dataset may negatively impact the accuracy of a machine-learning model in classifying rare events that are underrepresented in the training dataset. Training datasets comprising time-series data present a unique challenge. Accordingly, resampling techniques for up-sampling and / or down-sampling a training dataset of time series are disclosed. The up-sampling may respect the temporal correlation of time samples in the time series, while generating synthetic time series that mimic the feature values of time series belonging to the minority class. Down-sampling may be used to fine-tune the ratio of time series belonging to the minority class to the time series belonging to the majority class.

The application discloses a static software defect prediction method based on self-step oversampling integrated learning, according to label information, a training set is divided into majority class and minority class sets; the prediction result of an integrated classifier is used to estimate sample classification difficulty; according to the evaluation, bin processing is carried out on the two kinds of sample sets, and the average difficulty contribution of binning is determined; the self-step learning factor and the sampling weight of binning are updated based on the difficulty contribution; based on the sampling weight of binning, a training subset is obtained through weighted Bootstrap sampling; the training subset is subjected to SMOTE oversampling, and then a base classifier is trained; the prediction performance of the base classifier is used to determine the weight, and the integrated classifier is updated; the process is repeated until the integrated classifier of a specified size, namely a software defect prediction model, is obtained. The application overcomes the problems that the model training process lacks pertinence due to factors such as insufficient training data and class imbalance faced by the static software defect prediction task, and that overfitting occurs in the later model training period due to excessive attention to noise samples and abnormal samples, thereby affecting the defect prediction performance.

The present disclosure relates to a credit evaluation model generation method, device, equipment and medium, wherein the method comprises: obtaining a majority class sample set, a minority class sample set and a reference credit evaluation model; using the reference credit evaluation model to perform credit prediction on each first sample and each second sample respectively, to determine and delete a first target sample from the majority class sample set according to the prediction result; clustering each second sample to obtain a plurality of class clusters, and generating at least one second target sample according to the plurality of class clusters, to add the second target sample to the minority class sample set; and generating a target credit evaluation model according to the updated majority class sample set and the updated minority class sample set. Thus, the overfitting of the target credit evaluation model can be avoided, and the generalization performance of the target credit evaluation model can be improved.