Machine learning techniques: A systematic review

Artificial Neural Network based machine learning has achieved great success in the field of computer science. As the availability of data is increasing day by day, Machine Learning uses smart data analysis (for good cause) to make computers learn from old data and predict future outcomes. Neural Networks are very efficient in capturing the nonlinearity in dynamical systems, which is otherwise almost next to impossible to capture using traditional methods. Supervised and Semi-Supervised Learning are the branches of machine learning with a goal of developing a model using historical data to predict future instances. This paper presents an overview of Neural Network based machine learning and discusses supervised and semi-supervised learning in detail. Also, it highlights the types of supervised and semi-supervised learning techniques along with its algorithms. A detailed comparison of supervised and semi- supervised techniques is also discussed.