Machine Learning
Building predictive models that learn from data to solve real-world problems, optimizing accuracy and efficiency through advanced algorithms.
This study involves performing clustering analyses using different clustering methods (k-means, PAM, and hierarchical agglomerative clustering) on two different datasets. The project aims to explore the clustering tendency of datasets, determine the optimal number of clusters, validate clusters, and answer two research questions related to clustering.
The aim of this study is to replicate the research conducted by Abdollahi and Moghaddam (2022) to gain a deeper understanding of various feature selection methods, specifically ReliefF, FCBF, and genetic algorithms. This research focuses on improving the accuracy of heart disease diagnosis by using an ensemble classification model that incorporates a genetic algorithm and feature selection. The proposed model achieved a high accuracy of 97.57%, demonstrating its potential for effective implementation in healthcare settings.
Electric Vehicle Lithium-ion Battery Ageing Analysis Under Dynamic Condition: A Machine Learning Approach
The study by Swarnkar, et al. investigates the challenges of monitoring and predicting the health of lithium-ion batteries, which are critical for electric vehicles. The paper highlights the importance of a robust battery management system (BMS) due to the non-linear degradation of batteries influenced by factors such as charge/discharge rates, temperature, and other operational conditions. By analyzing battery performance under various dynamic conditions, the study applies machine learning techniques such as Multiple Linear Regression, M-SVM and Neural Network (LSTM-GPR), to predict the battery's state of health (SOH).
