Machine learning (ML) is emerging as an upcoming future technology for an enhanced cybersecurity measures with artificial intelligence (AI). It involves statistics and algorithms on the models so that machines can learn from data and take decisions without being specifically programmed. ML models can handle huge data, identify patterns, detect anomalies which can further prevent various cyber-attacks. Applying ML in cybersecurity could also be challenging because they are complex, and one require high knowledge to implement them. Furthermore, ML models are influenced by the quality of data; poor or noisy data may give false results. Decoding such challenges effectively is a critical task while applying ML in Cyber security. The objective of this review paper is to provide an idea of different ML techniques used in Cybersecurity with an aim to encourage researchers and practitioners to discover different areas for future research and development by giving them the current state of the field and how ML in cybersecurity is preventing ever-evolving cyber threats.