Smart Feedback Summarization Using Machine Learning Techniques

Abstract

Currently, the use of the Textual Feedback developed by the end-users has started growing rapidly in institutes, and this has led to various inefficiencies in analysis and decision making tasks. However, the main drawbacks of the existing systems are not specialized but generalized systems for use by academic institutions. Thus, to overcome this research gap, this research has proposed an Intelligent System for Academic Feedback Summarization by utilizing ML and NLP techniques. The proposed analysis and processing approach for analysis and processing includes the following phases such as preprocessing, TF-IDF Feature Extraction, Sentiment-based Classification, and Extensive Summarization techniques to design efficient feedback tools with compact information. Certain classifications are employed owing to their high efficiency with respect to the effective Sentiment analysis tasks. From the experimental results, it is evident that it is highly efficient and effective as it correctly classified the feedback and summarized the complete text to form a compact piece of information, which is of immense use in designing the key elements of academics for the administration and faculty staff of institutions to enhance the quality and services provided in academics.