Particle Swarm Optimization (PSO) is a well known technique for solving various kinds of combinatorial optimization problems including scheduling, resource allocation and vehicle routing. However, basic PSO suffers from premature convergence problem. Many techniques have been proposed for alleviating this problem. One of the alternative approaches is hybridization. Genetic Algorithms (GAs) are one of the possible techniques used for hybridization. Most often, a mutation scheme is added to the PSO, but some applications of crossover have been added more recently. Some of these schemes use dynamic parameterization when applying the GA operators. In this work, dynamic parameterized mutation and crossover operators are combined with a PSO implementation individually and in combination to test the effectiveness of these additions. The results indicate that all the PSO hybrids with dynamic probability have shown satisfactory performance in finding the best distance of the Vehicle Routing Problem With Time Windows.
<span lang="EN-US">Automated machine learning is a promising approach widely used to solve classification and prediction problems, which currently receives much attention for modification and improvement. One of the progressing works for automated machine learning improvement is the inclusion of evolutionary algorithm such as Genetic Programming. The function of Genetic Programming is to optimize the best combination of solutions from the possible pipelines of machine learning modelling, including selection of algorithms and parameters optimization of the selected algorithm. As a family of evolutionary based algorithm, the effectiveness of Genetic Programming in providing the best machine learning pipelines for a given problem or dataset is substantially depending on the algorithm parameterizations including the mutation and crossover rates. This paper presents the effect of different pairs of mutation and crossover rates on the automated machine learning performances that tested on different types of datasets. The finding can be used to support the theory that higher crossover rates used to improve the algorithm accuracy score while lower crossover rates may cause the algorithm to converge at earlier stage.</span>
Reseachers have acknowledged that machine learning is useful to be utilized in many different domains of complex real life problem. However, to implement a complete machine learning model involves some technical hurdles such as the steep learning curve, the abundance of the programming skills, the complexities of hyper-parameters, and the lack of user friendly platform to be used for the implementation. This paper provides an insight of a rapid software framework for implementing machine learning. This paper also demonstrates the empirical research results of machine learning classification models from the rapid software framework. Additionally, this paper explains comparisons of results between two platforms of rapid software; the proposed software and Python program. The machine learning model in the two platforms were tested on breast cancer and tax avoidance datasets with Decision Tree algorithm. The results indicated that although the software framework is easier than the programming platform for implementing the machine learning model, the results from the software framework were highly accurate and reliable. Keywords- Software framework, rapid, implementation, machine learning
This paper addresses the performances of machine learning classification models for the detection of tax avoidance problems. The machine learning models employed automated features selection with hybrid two metaheuristics algorithms namely particle swarm optimization (PSO) and genetic algorithm (GA). Dealing with a real dataset on the tax avoidance cases among companies in Malaysia, has created a stumbling block for the conventional machine learning models to achieve higher accuracy in the detection process as the associations among all of the features in the datasets are extremely low. This paper presents a hybrid meta-heuristic between PSO and adaptive GA operators for the optimization of features selection in the machine learning models. The hybrid PSO-GA has been designed to employ three adaptive GA operators hence three groups of features selection will be generated. The three groups of features selection were used in random forest (RF), k-nearest neighbor (k-NN), and support vector machine (SVM). The results showed that most models that used PSO-GA hybrids have achieved better accuracy than the conventional approach (using all features from the dataset). The most accurate machine learning model was SVM, which used a PSO-GA hybrid with adaptive GA mutation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.