Principal component analysis (PCA) is an unsupervised machine learning algorithm that plays a vital role in reducing the dimensions of the data in building an appropriate machine learning model. It is a statistical process that transforms the data containing correlated features into a set of uncorrelated features with the help of orthogonal transformations. Unsupervised machine learning is a concept of self-learning method that involves unlabelled data to identify hidden patterns. PCA converts the data features from a high dimensional space into a low dimensional space. PCA also acts as a feature extraction method since it transforms the ‘n’ number of features into ‘m’ number of principal components (PCs; m < n). Mobile Malware is increasing tremendously in the digital era due to the growth of android mobile users and android applications. Some of the mobile malware are viruses, Trojan horses, worms, adware, spyware, ransomware, riskware, banking malware, SMS malware, keylogger, and many more. To automate the process of detecting mobile malware without human intervention, machine learning methods are applied to discover the malware more precisely. Specifically, unsupervised machine learning helps to uncover the hidden patterns to detect anomalies in the data. In discovering hidden patterns of malware, PCA is an important dimensionality reduction technique that can be applied to transform the features into PCs containing important feature values. So, by implementing PCA, the correlated features are transformed into uncorrelated features automatically to explore the anomalies in the data effectively. This book chapter explains all the variants of the PCA, including all linear and non-linear methods of PCA and their suitability in applying to mobile malware detection. A case study on mobile malware detection with variants of PCA using machine learning techniques in CICMalDroid_2020 dataset has been experimented. Based on the experimental results, for the given dataset, normal PCA is suitable to detect the malware data points and forms an optimal cluster.
Part of the book: Advances in Principal Component Analysis