Feature selection, which is important for successful analysis of chemometric data, aims to produce parsimonious and predictive models. Partial least squares (PLS) regression is one of the main methods in chemometrics for analyzing multivariate data with inputXand responseYby modeling the covariance structure in theXandYspaces. Recently, orthogonal projections to latent structures (OPLS) has been widely used in processing multivariate data because OPLS improves the interpretability of PLS models by removing systematic variation in theXspace not correlated toY. The purpose of this paper is to present a feature selection method of multivariate data through orthogonal PLS regression (OPLSR), which combines orthogonal signal correction with PLS. The presented method generates empirical distributions of features effects uponYin OPLSR vectors via permutation tests and examines the significance of the effects of the input features onY. We show the performance of the proposed method using a simulation study in which a three-layer network structure exists in compared with the false discovery rate method. To demonstrate this method, we apply it to both real-life NIR spectra data and mass spectrometry data.
Measuring and managing the financial sustainability of the borrowers is crucial to financial institutions for their risk management. As a result, building an effective corporate financial distress prediction model has been an important research topic for a long time. Recently, researchers are exerting themselves to improve the accuracy of financial distress prediction models by applying various business analytics approaches including statistical and artificial intelligence methods. Among them, support vector machines (SVMs) are becoming popular. SVMs require only small training samples and have little possibility of overfitting if model parameters are properly tuned. Nonetheless, SVMs generally show high prediction accuracy since it can deal with complex nonlinear patterns. Despite of these advantages, SVMs are often criticized because their architectural factors are determined by heuristics, such as the parameters of a kernel function and the subsets of appropriate features and instances. In this study, we propose globally optimized SVMs, denoted by GOSVM, a novel hybrid SVM model designed to optimize feature selection, instance selection, and kernel parameters altogether. This study introduces genetic algorithm (GA) in order to simultaneously optimize multiple heterogeneous design factors of SVMs. Our study applies the proposed model to the real-world case for predicting financial distress. Experiments show that the proposed model significantly improves the prediction accuracy of conventional SVMs.
As technologies for storing time-series data such as smartwatches and smart factories become common, we are collectively accumulating a great deal of time-series data. With the accumulation of time-series data, the importance of time-series abnormality detection technology that detects abnormal patterns such as Cyber-Intrusion Detection, Fraud Detection, Social Networks Anomaly Detection, and Industrial Anomaly Detection is emerging. In the past, time-series anomaly detection algorithms have mainly focused on processing univariate data. However, with the development of technology, time-series data has become complicated, and corresponding deep learning-based time-series anomaly detection technology has been actively developed. Currently, most industries rely on deep learning algorithms to detect time-series anomalies. In this paper, we propose an anomaly detection algorithm with an ensemble of multi-point LSTMs that can be used in three cases of time-series domains. We propose our anomaly detection model that uses three steps. The first step is a model selection step, in which a model is learned within a user-specified range, and among them, models that are most suitable are automatically selected. In the next step, a collected output vector from M LSTMs is completed by stacking ensemble techniques of the previously selected models. In the final step, anomalies are finally detected using the output vector of the second step. We conducted experiments comparing the performance of the proposed model with other state-of-the-art time-series detection deep learning models using three real-world datasets. Our method shows excellent accuracy, efficient execution time, and a good F1 score for the three datasets, though training the LSTM ensemble naturally requires more time.