Abstract: Sentiment analysis has been extensively researched for the purpose of analyzing qualitative data using a computation technique. However, there are only a few research papers that focus on analyzing sentiments in terms of teachers’ evaluation. Analyzing comments on teacher evaluations can lead to understanding more of what faculty development programs can be provided to improve teachers’ academic performance. Thus, this study presents an opinion-mining system utilizing the Bayesian technique of the Naïve Bayes algorithm. The descriptive research method was used in this study, with a questionnaire serving as the instrument for testing the acceptability of the application. One hundred (100) evaluators were surveyed for the teachers’ evaluations. The application performance attributes are defined using the functionality, usability, reliability, performance and security (FURPS) model. The mean formula was used to analyze the data. The usability and security were evaluated as perfectly acceptable, with a weighted mean of 4.64, and 4.56, respectively. Furthermore, the functionality, reliability, and performance were evaluated at acceptable evaluation ratings with a weighted mean of 4.29, 4.02, and 4.11. The overall quality of the system was given an acceptable rating with a weighted mean of 4.32, indicating that the application provided and managed ratings and comments on individual teachers’ performance.