Open AccessExtracting and Composing a Dataset of Competitive Counter-Strike Global Offensive Matches
Author(s) -
E. J. Johnny Rocha,
Henrique Maio,
Daniel Sadoc Menasché,
Claudio Miceli
Publication year - 2021
Language(s) - English
Resource type - Conference proceedings
DOI - 10.5753/dsw.2021.17412
Subject(s) - offensive , computer science , analytics , data science , adversary , process (computing) , data visualization , construct (python library) , visualization , enhanced data rates for gsm evolution , data mining , artificial intelligence , computer security , operations research , engineering , programming language , operating system
There is a growing necessity for insightful and meaningful analyticswithin eSports: be it to entertain spectators as they watch their favorite teamscompete, to automatically identify and catch cheaters or even to gain a com-petitive edge over an opponent, there is a plethora of potential applicationsfor analytics within the scene. It follows then, that there is also a necessityfor well structured and organized datasets that enable efficient data explorationand serve as the foundation for the visualization and analytics layers. Becauseof this, the entire process - from data collection at the source to the means ofaccessing the desired information - need to be planned out to address thoseneeds. Our work provides the means by which to construct such a dataset forthe Counter-Strike Global Offensive (CS:GO) game, thus opening up a range ofpossible applications on top of the data