A Novel Distance Metric Based on Differential Evolution

Feature Transformation Framework for Enhancing Compactness and Separability of Data Points in Feature Space for Small Datasets

Compactness and separability of data points are two important properties that contribute to the accuracy of machine learning tasks such as classification and clustering. We propose a framework that enhances the goodness criteria of the two properties by transforming the data points to a subspace in the same feature space, where data points of the same class are most similar to each other. Most related research about feature engineering in the input data points space relies on manually specified transformation functions. In contrast, our work utilizes a fully automated pipeline, in which the transformation function is learnt via an autoencoder for extraction of latent representation and multi-layer perceptron (MLP) regressors for the feature mapping. We tested our framework on both standard small datasets and benchmark-simulated small datasets by taking small fractions of their samples for training. Our framework consistently produced the best results in all semi-supervised clustering experiments based on K-means and different seeding techniques, with regards to clustering metrics and execution time. In addition, it enhances the performance of linear support vector machine (LSVM) and artificial neural network (ANN) classifier, when embedded as a preprocessing step before applying the classifiers.

Fault Detection in the MSW Incineration Process Using Stochastic Configuration Networks and Case-Based Reasoning

Fault detection in the waste incineration process depends on high-temperature image observation and the experience of field maintenance personnel, which is inefficient and can easily cause misjudgment of the fault. In this paper, a fault detection method is proposed by combining stochastic configuration networks (SCNs) and case-based reasoning (CBR). First, a learning pseudo metric method based on SCNs (SCN-LPM) is proposed by training SCN learning models using a training sample set and defined pseudo-metric criteria. Then, the SCN-LPM method is used for the case retrieval stage in CBR to construct the fault detection model based on SCN-CBR, and the structure, algorithmic implementation, and algorithmic steps are given. Finally, the performance is tested using historical data of the MSW incineration process, and the proposed method is compared with typical classification methods, such as a Back Propagation (BP) neural network, a support vector machine, and so on. The results show that this method can effectively improve the accuracy of fault detection and reduce the time complexity of the task and maintain a certain application value.