ABSTRACT
The widespread of Corona virus in Malaysia has led to open distance learning (ODL) for every student in continuing their study. However, the student readiness in ODL is unknown for the university. Thus, university institutions must always monitor what the university says about social media and the student readiness on ODL. This project aimed at developing a sentiment classification by means of a Naïve Bayes algorithm in displaying the readiness index of students by extracting Twitter tweets. A standard sentiment analysis performance measurement was used in evaluating the developed sentiment analysis model. About 98.8% of the extracted tweet are negative tweet about the online learning which indicate that the students are not ready for the ODL. The classifier model's precision, recall and f1-measure for each category on online learning readiness are obtained. The best f1-score of the classifier model is in online category with 29.0% for negative tweets and 31.7% for positive tweets compare to the other category with a precision of 27.9% for negative tweets and 30.7% for positive tweets while for the recall value is 29.6% for negative tweets and 32.2% for positive tweets. However, there are room for improvements by implement of the autocorrect function for the sentiment analysis model, implementation of a live update function in dynamically for student readiness index live when new tweets appear, implement a machine classification model, providing dynamic database application and used higher performance computer for data scrapping. © 2022 IEEE.