Automated Identification of Child Abuse in Chat Rooms by Using Data Mining

Automated Identification of Child Abuse in Chat Rooms by Using Data Mining

Mohammadreza Keyvanpour (Alzahra University, Iran), Mohammadreza Ebrahimi (Concordia University, Canada), Necmiye Genc Nayebi (École de Technologie Supérieure (ÉTS), Canada), Olga Ormandjieva (Concordia University, Canada) and Ching Y. Suen (Concordia University, Canada)
Copyright: © 2016 |Pages: 30
DOI: 10.4018/978-1-5225-0463-4.ch009
OnDemand PDF Download:
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Providing a safe environment for juveniles and children in online social networks is considered as one of the major factors of improving public safety. Due to the prevalence of the online conversations, mitigating the undesirable effects of child abuse in cyber space has become inevitable. Using automatic ways to combat this kind of crime is challenging and demands efficient and scalable data mining techniques. The problem can be casted as a combination of textual preprocessing in data/text mining and pattern classification in machine learning. This chapter covers different data mining methods including preprocessing, feature extraction and the popular ways of feature enrichment through extracting sentiments and emotional features. A brief tutorial on classification algorithms in the domain of automated predator identification is also presented through the chapter. Finally, the discussion is summarized and the challenges and open issues in this application domain are discussed.
Chapter Preview
Top

1. Introduction

The ease of access and anonymity of the Internet users facilitate child exploitation, online bullying and cyber sexual abuse. This has been a major concern in developed countries with a high rate of Internet access in which children are basically the most vulnerable Internet stakeholders. Automated Online Predator Identification (OPI) is a proactive way to counteract the undesirable effects causing by aforementioned crimes. Recently in the literature, this is also known as Sexual Predator Identification (SPI) or Sexual Predator Detection (SPD). Although practical OPI problem involves dealing with textual data and images, textual data are considerably more convenient to be used for automation purposes rather than the imagery data. Accordingly, dealing with textual data is meant throughout the chapter, wherever the OPI is mentioned in general. This part highlights the importance of Online Predator Identification as an effective action toward improving public safety in society. The target audience of this chapter is the researchers working in the area of crime data mining and readers who want to have an overall grasp of the OPI field. In sections 1-1 and 1-2, we present the key concepts about public safety, OPI domain and the domain related aspects, whereas Section 1-3 will provide further information in regards to the relationship between data mining and OPI.

1.1 Public Safety and OPI

The ease of access and anonymity of the Internet users has made child abuse easier than the past. According to Kierkegaard (2008), sexual solicitations of 89% of youth are made in chat rooms. This implies the vital need for mining large volumes of anonymous chat logs in order to combat this kind of social crime. Providing a safe environment for juveniles and children in online social networks is considered as one of the major factors of improving public safety. Due to the prevalence of the online conversations, mitigating the undesirable effects of juvenile abuse in cyber space has become critical.

1.2 Domain Concepts

This part contains the essential information about legal and psychological aspects of online predator identification.

Although all of the legislative and regulatory provisions regarding online child sexual abuse aim to combat and mitigate the impact of this threat, they may vary in different countries or even for different jurisdictions in the same country. According to Kierkegaard (2008), while virtual child porn using avatars is generally considered illegal in the European Union, it might not necessarily be treated as such in the United States (p. 44). Similarly, images which are illegal to view in the USA may not be illegal to view in Germany (p. 41). The same situation exists in the concept of age disparity between the adult and minor. Hence, there have been countless writings on legal aspects of child sexual abuse in online environment which go beyond the scope of this chapter.

1.2.2 Psychological Aspects

The most effective and also naïve psychological aspects of “predatorhood” might be those defined by Morris in his master of science thesis (Morris, 2013). The author defined predatorhood as having two major components: age disparity and inappropriate intimacy. The former relates to the psychological immaturity of the victim compared to that of predator (adult) which may differ in various countries by law. The latter corresponds to the attempt of adult to establish an intimate conversation that usually involves implicit or explicit sexual comments.

One of the most practical psychological theories which is widely used in online predator identification is known as luring communication theory (Olson, Daggs, Ellevold, & Rogers, 2007). The theory comprises three main phases needed for committing a predatory act:

  • 1.

    Gaining access to the victim.

  • 2.

    Entrapping and grooming until the victim accepts sexual advances.

  • 3.

    Initiating and maintaining the abusive relationship.

On the Internet, the most common way for gaining access to the victim is through online conversations in chat rooms. The second stage can be distinguished by observation of the predator’s attempt to desensitize the child to inappropriate intimacy. Finally, the third step involves explicit sexual exploitation of the minor. When an explicit exploitation is about to occur, a reliable OPI system can flag the conversation for the attention of law enforcement in order to prevent the predator from approaching the victim.

Complete Chapter List

Search this Book:
Reset