Imbalanced Time Series Classification Research Articles

Minimum Recall-Based Loss Function for Imbalanced Time Series Classification

This paper deals with imbalanced time series classification problems. In particular, we propose to learn time series classifiers that maximize the minimum recall of the classes rather than the accuracy. Consequently, we manage to obtain classifiers which tend to give the same importance to all the classes. Unfortunately, for most of the traditional classifiers, learning to maximize the minimum recall of the classes is not trivial (if possible), since it can distort the nature of the classifiers themselves. Neural networks, in contrast, are classifiers that explicitly define a loss function, allowing it to be modified. Given that the minimum recall is not a differentiable function, and therefore does not allow the use of common gradient-based learning methods, we apply and evaluate several smooth approximations of the minimum recall function. A thorough experimental evaluation shows that our approach improves the performance of state-of-the-art methods used in imbalanced time series classification, obtaining higher recall values for the minority classes, incurring only a slight loss in accuracy.

A cost-sensitive attention temporal convolutional network based on adaptive top-k differential evolution for imbalanced time-series classification

Imbalanced time-series classification (ITSC) is ubiquitous in many real-world applications. In this study, a novel cost-sensitive deep learning framework, namely ACS-ATCN, is proposed for ITSC. With the framework of ACS-ATCN, first, weighted class costs are optimized jointly with the hyperparameters of an attention temporal convolutional network (ATCN). Second, an improved evolutionary algorithm, termed adaptive top-k differential evolution (ATDE), is presented for optimizing class costs as well as the network’s hyperparameter. Experiments on five data sets demonstrate that ACS-ATCN achieves a higher average G-mean than other cost-sensitive learning and oversampling algorithms while using much less computational time. Comparison between different deep learning frameworks also confirms its advantages over other existing benchmarking methods in ITSC. Experimental results also reveal that ATDE provides more accurate classification than the vanilla DE algorithm, and saves as high as 41.53% of average computational expense for convergence.

Minority oversampling for imbalanced time series classification

Many vital real-world applications involve time-series data with skewed distribution. Compared to traditional imbalanced learning problems, the classification of imbalanced time-series data is more challenging due to the high dimensionality and high inter-variable correlation. This paper proposes a structure-preserving Oversampling method to resolve the High-dimensional Imbalanced Time-series classification (OHIT). OHIT leverages a density-ratio-based shared nearest neighbor clustering algorithm to capture the modes of minority class in high-dimensional space. It for each mode applies the shrinkage technique of large-dimensional covariance matrix to obtain an accurate and reliable covariance structure. The structure-preserving synthetic samples are eventually generated based on the multivariate Gaussian distribution with the estimated covariance matrix. In addition, to further promote the performance of classifying imbalanced time-series data, we integrate OHIT into boosting framework to obtain a new ensemble algorithm OHITBoost. Extensive experiments on several publicly available time-series datasets (including unimodal and multimodal) demonstrate the effectiveness of OHIT and OHITBoost in terms of F1, G-mean, and AUC.

Optimizing shapelets quality measure for imbalanced time series classification

Time series classification has been considered as one of the most challenging problems in data mining and is widely used in a broad range of fields. A biased distribution leads to classification on minority time series objects more severe. A commonly taken approach is to extract or select the representative features to retain the structure of a time series object. However, when the data distribution is imbalanced, the traditional features cannot represent time series effectively, especially in multi-class environment. In this paper, Shapelets — a primitive time series mining technology — is applied to extract the most representative subsequences. Especially, we verify that IG (Information Gain) is unsuitable as a shapelet quality measure for imbalanced data sets. Nevertheless, we propose two quality measures for shapelets on imbalanced binary and multi-class problem respectively. Based on extracted shapelet features, we select the diversified top-k shapelets based on new quality measure to represent the top-k best features and achieve this procedure on map-reduce framework. Lastly, two oversampling methods based on shapelet features are proposed to re-balance the binary and multi-class time series data sets. We validated our methods on the benchmark data sets by comparing with the canonical classifiers and the state-of-the-art time series algorithms. It is verified that the proposed algorithms perform more competitive than the compared methods in statistical significance.

Cost-sensitive convolutional neural networks for imbalanced time series classification

Cost-sensitive convolutional neural networks for imbalanced time series classification

A parsimonious mixture of Gaussian trees model for oversampling in imbalanced and multimodal time-series classification.

We propose a novel framework of using a parsimonious statistical model, known as mixture of Gaussian trees, for modeling the possibly multimodal minority class to solve the problem of imbalanced time-series classification. By exploiting the fact that close-by time points are highly correlated due to smoothness of the time-series, our model significantly reduces the number of covariance parameters to be estimated from O(d(2)) to O(Ld), where L is the number of mixture components and d is the dimensionality. Thus, our model is particularly effective for modeling high-dimensional time-series with limited number of instances in the minority positive class. In addition, the computational complexity for learning the model is only of the order O(Ln+d(2)) where n+ is the number of positively labeled samples. We conduct extensive classification experiments based on several well-known time-series data sets (both single- and multimodal) by first randomly generating synthetic instances from our learned mixture model to correct the imbalance. We then compare our results with several state-of-the-art oversampling techniques and the results demonstrate that when our proposed model is used in oversampling, the same support vector machines classifier achieves much better classification accuracy across the range of data sets. In fact, the proposed method achieves the best average performance 30 times out of 36 multimodal data sets according to the F-value metric. Our results are also highly competitive compared with nonoversampling-based classifiers for dealing with imbalanced time-series data sets.

Integrated Oversampling for Imbalanced Time Series Classification

This paper proposes a novel Integrated Oversampling (INOS) method that can handle highly imbalanced time series classification. We introduce an enhanced structure preserving oversampling (ESPO) technique and synergistically combine it with interpolation-based oversampling. ESPO is used to generate a large percentage of the synthetic minority samples based on multivariate Gaussian distribution, by estimating the covariance structure of the minority-class samples and by regularizing the unreliable eigen spectrum. To protect the key original minority samples, we use an interpolation-based technique to oversample a small percentage of synthetic population. By preserving the main covariance structure and intelligently creating protective variances in the trivial eigen dimensions, ESPO effectively expands the synthetic samples into the void area in the data space without being too closely tied with existing minority-class samples. This also addresses a key challenge for applying oversampling for imbalanced time series classification, i.e., maintaining the correlation between consecutive values through preserving the main covariance structure. Extensive experiments based on seven public time series data sets demonstrate that our INOS approach, used with support vector machines (SVM), achieved better performance over existing oversampling methods as well as state-of-the-art methods in time series classification.