The use and development of specific machine learning (ML) algorithms are gaining popularity. ML is the study of teaching computers to learn and act like people by giving those data and knowledge without directly programming them. ML algorithms are trained using training data. They can make precise predictions and judgments based on previous data when new data is received. A range of learning strategies offered for various intrusion detection challenges may be divided into two major categories such as unsupervised (anomaly detection and clustering) and supervised (classification). Therefore, this paper reviews the characteristics of supervised and unsupervised machine learning to review their employment in different places. In addition to that this paper also compares the supervised and unsupevise3d machine learning which can further be used by data scientists as a primer for their study.