This paper presented the development and evaluation of an Intelligent Tutoring System (ITS) for learning Java threads. The system utilized a Bayesian Network to predict student certainty in Computer Science students at FEU Institute of Technology. The main objective of the system is to help students better understand the concept of Java threads and improved their performance in this area. The research approach follows a Descriptive-Developmental study design, which evaluated the impact of the ITS on student learning outcomes. The system provided a concept thread related activities, assesses student answers, and tracks performance data. It benefits instructors by offering insights into student knowledge, automating tasks, and providing analytics for decision-making. To prove the effectiveness of the system for tutoring computer science students. The researchers conducted a survey involving 74 respondents, including students and instructors. The software evaluation received positive feedback regarding functional suitability, usability, reliability, and efficiency. The ITS achieved a high Likert scale average rating of 4.4 signified those respondents were "Very Satisfied" with its overall performance. It effectively supported tutoring and enhanced student learning that highlighted the potential of Bayesian Networks in Intelligent Tutoring Systems.