Sentiment Analysis of Visual, Auditory, and Kinesthetic Learning Styles Using Logistic Regression

Learning styles influence the way individuals learn and what they can learn. The visual, auditory, and kinesthetics (VAK) learning style theory categorizes learners based on the mode of receiving information. Conventional methods of determining learning styles are usually time-consuming and non-automated. This research focuses on creating a web-based system that classifies users' dominant learning styles according to their responses through Logistic Regression (LR) for sentiment analysis. It consists of a friendly and interactive user interface to identify their dominant learning style. This project uses the Modified Waterfall Model, which entails requirement collection, system design, implementation, testing, and deployment. Feature extraction methods which are Count Vectorizer (CV) was used to transform text data, hence facilitating effective classification. The classification model is trained on labelled data in a manner that facilitates accuracy in classifying learning styles based on user inputs. The classification model was trained on labelled datasets, achieving an accuracy score of 96 % in predicting learning styles based on user responses. Future work will focus on enhancing the system by incorporating additional machine learning models to improve classification performance and expanding the dataset for better generalization.