Prospects and Limitations of Machine Learning in Computer Science Education

Abstract

The study examined the prospects and limitations of Machine Learning (ML) in Computer Science education. Thus, it explored the concept of ML, ML algorithms, categorization, prospects and limitations of ML application in computer science education. The review indicated that machine learning offers enormous opportunities that could improve the quality of teaching and learning of Computer Science, particularly in the areas of development of intelligent tutoring systems and educational software/apps, monitoring student’s  learning progress and prediction of their learning styles and outcomes, automated assessment and grading, customized lecture, student placement decision, teachers’ training, sentiment analysis, academic translations, detection and diagnostic supports for at-risk students. Also, it highlights several machine learning algorithms like; K-Nearest Neighbour (KNN), Naïve Bayes, Neural Network, Random Forest, Decision tree and many others that have prognostic capabilities that are of great interest to educators and are being used to advance teaching excellence and assist students and educators to access online education resources and improve their experiences. Thus, it concluded that machine learning is a promising pedagogical technology that can change the narratives in the education sector in no distant future both in terms of content development and delivery, mode of instruction, learning, feedback and evaluation pattern.