Among all agricultural inputs which include seeds, fertilizer, plant protection, machinery and credit, irrigation assumes a very importance place. Irrigation means watering the fields by any means other than rain. Indian agriculture still dependent on monsoon rains. Over the last one-decade monsoon has been affected due to climate change factors. The ground water has been depleting and untimely rain has led to flood. Therefore, to enhance the Indian irrigation system adaptation of cutting-edge technologies like machine learning is the need of hour. Machine learning techniques in irrigation is one of the most effective means in upgrading of land and water management and increasing food production. Adoption of machine learning techniques and automation enhanced water use efficiency in irrigation, increases yield per land and improved produce quality. Machine learning technology plays important role in smart irrigation system. It helps to use the water in efficient manner and reduces water wastage. It uses computational method to learn agriculture data. The focus of the learning process is to learn from training data to perform a given task. In this study we did a review on various machine algorithms like Decision Tree, SVM, KNN (K- Nearest Neighbors), Deep learning, CNN etc used for smart irrigation. This study also discussed these machine learning algorithms and their applications in irrigation system their characteristics and functionalities.