Detecting Predatory Behaviour in Online Game Chats

Elin Rut Gudnadottir; Alaina K. Jensen; Yun-Gyung Cheong; Julian Togelius; Byung Chull Bae; Christoffer Holmgård Pedersen

Detecting Predatory Behaviour in Online Game Chats

Elin Rut Gudnadottir, Alaina K. Jensen, Yun-Gyung Cheong, Julian Togelius, Byung Chull Bae, Christoffer Holmgård Pedersen

Center for Digital Play

Research output: Contribution to conference - NOT published in proceeding or journal › Paper › Research › peer-review

Abstract

This paper describes a machine learning approach to detect
sexually predatory behaviour in the massively multiplayer online game for children, MovieStarPlanet. The goal of this work is to take a chat log as an input and outputs its label as either the predatory category or the non-predatory category. From the raw in-game chat logs provided by MovieStarPlanet, we ﬁrst prepared three sub datasets via extensive preprocessing. Then, two machine learning algorithms, naive Bayes and Decision Tree, were employed to model the predatory behaviour using diﬀerent feature sets. Our evaluation has revealed that the proposed
approach achieved high accuracies in detecting predatory chats

Original language	English
Publication date	9 Nov 2013
Number of pages	11
Publication status	Published - 9 Nov 2013
Event	The 2nd Workshop on Games and NLP: Workshop at the 6th International Conference on Interactive Digital Storytelling - Bahcesehir University Galata Campus (Animation Lab), Istanbul, Turkey Duration: 9 Nov 2013 → 9 Nov 2013 Conference number: 6 http://gamesandnarrative.net/icids2013/call-to-participate-in-workshops

Workshop

Workshop	The 2nd Workshop on Games and NLP
Number	6
Location	Bahcesehir University Galata Campus (Animation Lab)
Country/Territory	Turkey
City	Istanbul
Period	09/11/2013 → 09/11/2013
Internet address	http://gamesandnarrative.net/icids2013/call-to-participate-in-workshops

Keywords

NLP
predator
game
text classification

Access to Document

SIREN: Social games for conflIct REsolution based on natural iNteraction
Yannakakis, G. (PI), Togelius, J. (CoI), Cheong, Y.-G. (CoI), Khaled, R. (CoI), Grappiolo, C. (CoI), Liapis, A. (CoI) & Holmgård, C. (CoI)
European Commission
01/09/2010 → 31/08/2013
Project: Research

Cite this

@conference{e8e415ccf05a497bbc43b45ccb944ad3,

title = "Detecting Predatory Behaviour in Online Game Chats",

abstract = "This paper describes a machine learning approach to detectsexually predatory behaviour in the massively multiplayer online game for children, MovieStarPlanet. The goal of this work is to take a chat log as an input and outputs its label as either the predatory category or the non-predatory category. From the raw in-game chat logs provided by MovieStarPlanet, we ﬁrst prepared three sub datasets via extensive preprocessing. Then, two machine learning algorithms, naive Bayes and Decision Tree, were employed to model the predatory behaviour using diﬀerent feature sets. Our evaluation has revealed that the proposedapproach achieved high accuracies in detecting predatory chats",

keywords = "NLP, predator, game, text classification, NLP, predator, game, text classification",

author = "Gudnadottir, {Elin Rut} and Jensen, {Alaina K.} and Yun-Gyung Cheong and Julian Togelius and Bae, {Byung Chull} and Pedersen, {Christoffer Holmg{\aa}rd}",

year = "2013",

month = nov,

day = "9",

language = "English",

note = "The 2nd Workshop on Games and NLP : Workshop at the 6th International Conference on Interactive Digital Storytelling, GAMNLP-13 ; Conference date: 09-11-2013 Through 09-11-2013",

url = "http://gamesandnarrative.net/icids2013/call-to-participate-in-workshops",

}

TY - CONF

T1 - Detecting Predatory Behaviour in Online Game Chats

AU - Gudnadottir, Elin Rut

AU - Jensen, Alaina K.

AU - Cheong, Yun-Gyung

AU - Togelius, Julian

AU - Bae, Byung Chull

AU - Pedersen, Christoffer Holmgård

N1 - Conference code: 6

PY - 2013/11/9

Y1 - 2013/11/9

N2 - This paper describes a machine learning approach to detectsexually predatory behaviour in the massively multiplayer online game for children, MovieStarPlanet. The goal of this work is to take a chat log as an input and outputs its label as either the predatory category or the non-predatory category. From the raw in-game chat logs provided by MovieStarPlanet, we ﬁrst prepared three sub datasets via extensive preprocessing. Then, two machine learning algorithms, naive Bayes and Decision Tree, were employed to model the predatory behaviour using diﬀerent feature sets. Our evaluation has revealed that the proposedapproach achieved high accuracies in detecting predatory chats

AB - This paper describes a machine learning approach to detectsexually predatory behaviour in the massively multiplayer online game for children, MovieStarPlanet. The goal of this work is to take a chat log as an input and outputs its label as either the predatory category or the non-predatory category. From the raw in-game chat logs provided by MovieStarPlanet, we ﬁrst prepared three sub datasets via extensive preprocessing. Then, two machine learning algorithms, naive Bayes and Decision Tree, were employed to model the predatory behaviour using diﬀerent feature sets. Our evaluation has revealed that the proposedapproach achieved high accuracies in detecting predatory chats

KW - NLP

KW - predator

KW - game

KW - text classification

KW - NLP

KW - predator

KW - game

KW - text classification

M3 - Paper

T2 - The 2nd Workshop on Games and NLP

Y2 - 9 November 2013 through 9 November 2013

ER -

Detecting Predatory Behaviour in Online Game Chats

Abstract

Workshop

Keywords

Access to Document

Fingerprint

Projects

SIREN: Social games for conflIct REsolution based on natural iNteraction

Cite this