Education is very important for the future growth of the students. Educational institutions are taking efforts to provide the quality education. Faculty members play key role in the performance and development of the students. Educational field contains large amount of data that needs to be used for predicting the learning behaviour of the students. Data analytics need to be used effectively to predict learning behaviour of the students. The prediction results can be used to identify the students at risk, as well as who are likely to be dropouts who need timely intervention. It is helpful to the faculty members to make use of these learning behaviours in improving teaching learning process. For the present research work, data is collected from UG (B.Com., BBA, BBA (CA), B.Sc., (CS)) students. The prediction of learning behaviour of the students is done using various machine learning algorithms through classification technique. The comparison of the algorithms is represented graphically. The interpretations of research findings of the present study are also discussed.