Academic information systems are a place of storage for data collection, processing, analysis and reporting of educational information. Improving the quality of education in higher education can be seen from high rates of student success and low failure rates of students. That has become a challenge for Technical education institutions to maintain their reputation and business continuity. Some of the educational organizations have multi-education paths such as Diploma, Engineering and Medicine collages. In such colleges, the behavior of the student in the preparatory year determines which education path the student will join in the future. The improper selection of courses would seriously affect the students’ achievements, which enforces students to drop out the improperly selected courses. Therefore, there is an urgent need to develop course recommendation system. In order to solve such an issue, proposed course recommendation system by using Condensed  Nearest Neighbour (CNN) with Random Forest (RF) Classification. Here the  dataset are collected from school passed out registered for admission to various diploma Courses. The process of data pre-processing takes place by using natural language processing to prepare the data in the correct object oriented format. The process of feature extraction to describe future prediction and recommendation  results of object oriented data analysis and interpretation of prediction models. The experimental result shows that the proposed CNN with RF is effective by achieving accuracy of 98.67%. Whereas, the existing K-Nearest Neighbor (KNN)  method showed accuracy of 92.52% for course recommendation.

Key words: Condensed  Nearest Neighbour , Course Recommendation,  Classification, Random Forest, K-Nearest Neighbor.