: Indian Technical institutes create naval knowledge and support social communities along with regular academic requirements. They play a significant role to increase engineering competitiveness from local to national level. National Institute Ranking framework (NIRF), India evaluates all technical institutions based on teaching, Learning & Resources, Research, Professional Practice and Collaborative Performance, Graduation Outcomes, Outreach and Perception. Sentiment analysis has wide scope in many domains including education, medical etc. Several research reports provide the state of the applications of sentiment analysis in industry, business, social and educational performance. But no / few work focused on ranking of Engineering College ranking using natural language processing, deep learning and machine learning solutions. The aim of the research work was to develop a sentiment analysis of the NIRF in order to enhance the performance of the ranking method. The work investigates the effect of NIRF five factors on ranking of engineering college values and brand attachments based on stakeholders such as students, parents and industries sentimental values. The work used 5002, 3051, 2821, 4252 and 3625 Twitter data for Learning & Resources, Research, Professional Practice and Collaborative Performance, Graduation Outcomes, Outreach and Perception respectively. This research work has applied Natural Language Processing (NLP) operations such as pre-processing, stop-word elimination, tf-idf transformation and n-gram model to bring the textual data to machine learning understandable format. Later state-of-the-art Machine Learning (ML) algorithms had to be applied following the topic modeling and extraction of the sentiment. This research work mostly focused on the features and key terms which will influence the prospective ranking of the educational institutions with their percentage of the contribution. performed machine learning on the 26678 tweets followers of 25 institutes were considered for the ranking process and conducted statistical verification with NIRF rankings.

Sentiment analysis, Natural Language Process, Machine Learning, Institute ranking