XGBoost and Deep Neural Network Comparison: The Case of Teams’ Performance Filippos Giannakas (B ) , Christos Troussas, Akrivi Krouska, Cleo Sgouropoulou, and Ioannis Voyiatzis Department of Informatics and Computer Engineering, University of West Attica, Egaleo, P. O. 12243, Athens, Greece {fgiannakas,ctrouss,akrouska,csgouro,voyageri}@uniwa.gr Abstract. In the educational setting, working in teams is considered an essential collaborative activity where various biases exist that influ- ence the prediction of teams performance. To tackle this issue, machine learning algorithms can be properly explored and utilized. In this con- text, the main objective of the current paper is to explore the ability of the eXtreme Gradient Boosting (XGBoost) algorithm and a Deep Neu- ral Network (DNN) with 4 hidden layers to make predictions about the teams’ performance. The major finding of the current paper is that shal- low machine learning performed better learning and prediction results than the DNN. Specifically, the XGBoost learning accuracy was found to be 100% during teams learning and production phase, while its pre- diction accuracy was found to be 95.60% and 93.08%, respectively for the same phases. Similarly, the learning accuracy of the DNN was found to be 89.26% and 81.23%, while its prediction accuracy was found to be 80.50% and 77.36%, during the two phases. Keywords: Deep Neural Network · Machine learning · Comparison · XGBoost · Adamax · Team performance 1 Introduction Nowadays, Machine Learning (ML) gains applicability in various domains where different applications were developed for addressing e.g., sentiment analysis, image recognition, natural language processing, speech recognition, and others. Especially in the educational setting, various ML applications were developed and incorporated on Intelligent Tutoring Systems (ITS), that among others, sup- port and increase learners’ engagement, retention, motivation, towards improv- ing their learning outcomes [12–14]. The learning process is becoming more challenging when learners are engaged in a team-based learning experience. This is because collaboration and commu- nication activities among participants seem to significantly influence students’ c  Springer Nature Switzerland AG 2021 A. I. Cristea and C. Troussas (Eds.): ITS 2021, LNCS 12677, pp. 343–349, 2021. https://doi.org/10.1007/978-3-030-80421-3_37