Sentiment Analysis on Students' Satisfaction of the Tanauan City College using Naïve Bayes Algorithm

Abstract

Sentiment analysis (SA) is a technique for determining how people feel or think about the products or services offered by a particular organization. It is the study of people's thoughts, feelings, judgments, attitudes, and emotions as expressed in written language. In Higher Education Institutions (HEIs), it is important to identify the sentiments of each stakeholder to ensure its continuous improvement. One of the stakeholders that should not be ignored and whose opinion may greatly impact the institution is the students. Their opinions may impact HEI's brand image. Tanauan City College (TCC) is one of the HEIs recognized by the Commission on Higher Education (CHED) under the category of Local Universities and Colleges (LUCs). At TCC, the students openly express their sentiments through social media platforms or even through the use of Google Forms. Mining those sentiments may be the technique of the institution to understand its stakeholders' experiences and may serve as inputs to decision-making to improve its services and the quality of education it provides. The study's main objective is to analyze the sentiments of TCC stakeholders using the Naïve Bayes (NB) algorithm. Data collection, preprocessing, categorization, validation and assessment, and NB integration are the proposed incremental processes used in the study. Findings of the study revealed that the implementation of the NB algorithm returned an accuracy of 85.41% for the sentiment analysis of TCC students. Therefore, the study confirms that NB is one of the choices for text categorization and could be an efficient and optimum solution for SA problems.