Kernelized evolutionary distance metric learning for semi-supervised clustering

A Hybrid Particle Swarm Optimization Algorithm with Dynamic Adjustment of Inertia Weight Based on a New Feature Selection Method to Optimize SVM Parameters

Support vector machine (SVM) is a widely used and effective classifier. Its efficiency and accuracy mainly depend on the exceptional feature subset and optimal parameters. In this paper, a new feature selection method and an improved particle swarm optimization algorithm are proposed to improve the efficiency and the classification accuracy of the SVM. The new feature selection method, named Feature Selection-score (FS-score), performs well on data sets. If a feature makes the class external sparse and the class internal compact, its FS-score value will be larger and the probability of being selected will be greater. An improved particle swarm optimization model with dynamic adjustment of inertia weight (DWPSO-SVM) is also proposed to optimize the parameters of the SVM. By improving the calculation method of the inertia weight of the particle swarm optimization (PSO), inertia weight can decrease nonlinearly with the number of iterations increasing. In particular, the introduction of random function brings the inertia weight diversity in the later stage of the algorithm and the global searching ability of the algorithm to avoid falling into local extremum. The experiment is performed on the standard UCI data sets whose features are selected by the FS-score method. Experiments demonstrate that our algorithm achieves better classification performance compared with other state-of-the-art algorithms.

Reinforcement learning based metric filtering for evolutionary distance metric learning

Data collection plays an important role in business agility; data can prove valuable and provide insights for important features. However, conventional data collection methods can be costly and time-consuming. This paper proposes a hybrid system R-EDML that combines a sequential feature selection performed by Reinforcement Learning (RL) with the evolutionary feature prioritization of Evolutionary Distance Metric Learning (EDML) in a clustering process. The goal is to reduce the features while maintaining or increasing the accuracy leading to less time complexity and future data collection time and cost reduction. In this method, features represented by the diagonal elements of EDML matrices are prioritized using a differential evolution algorithm. Further, a selection control strategy using RL is learned by sequentially inserting and evaluating the prioritized elements. The outcome offers the best accuracy R-EDML matrix with the least number of elements. Diagonal R-EDML focusing on the diagonal elements is compared with EDML and conventional feature selection. Full Matrix R-EDML focusing on the diagonal and non-diagonal elements is tested and compared with Information-Theoretic Metric Learning. Moreover, R-EDML policy is tested for each EDML generation and across all generations. Results show a significant decrease in the number of features while maintaining or increasing accuracy.