• Corpus ID: 18331862

Comparison of Rule-based to Human Analysis of Chat Logs

  title={Comparison of Rule-based to Human Analysis of Chat Logs},
  author={April Kontostathis and Lynne Edwards and Jennifer Bayzick and India McGhee and Amanda Leatherman and Kristina Moore},

Tables from this paper

Detecting Child Grooming Behaviour Patterns on Social Media

This paper proposes to characterise such stages of child grooming by a series of features covering sentiment polarity, content, and psycho-linguistic and discourse patterns, vital for building robust systems for detecting grooming conversations and potential predators on social media.

Automatic Feedback System for Collaborative Learning using Chats and Forums

A discourse model for the analysis of chat conversations and discussion forums for discovering implicit links between utterances that build into conversation threads is proposed, based on Mikhail Bakhtin’s dialogic theory.

A Human-Centered Systematic Literature Review of the Computational Approaches for Online Sexual Risk Detection

A comprehensive literature review to analyze 73 peer-reviewed articles on computational approaches utilizing text or meta-data/multimedia for online sexual risk detection found that the majority of work has focused on identifying sexual predators after-the-fact, rather than taking more nuanced approaches to identify potential victims and problematic patterns that could be used to prevent victimization before it occurs.

Exposing patterns of adult solicitor behaviour: towards a theory of control within the cybersexual abuse of youth

An evolved grounded theory analysis of 90 Perverted Justice transcripts is used to extract the behavioural patterns that emerge when online solicitors use instant messaging to enact behaviours of grooming, predation, offending, and control.

Using Luring Communication Theory to Analyze the Behavior of Online Sexual Offenders

The prevalence of sexual offenders (SOs) contacting children online attempting to engage in sexual exploitation has increased dramatically in recent years likely due to more potential victims and

Data quality measures for identity resolution

This thesis describes and grounds a comprehensive matching schema of profile attributes, and defines data quality measures which are important for identity resolution, and measures the availability, consistency and uniqueness of the schema’s contents.

The Dilemmatic Nature of Luring Communication: An Action-Implicative Discourse Analysis of Online Predator and P-J Member Interaction

institutional context and it is ever-present and implicates everyone within a given society. All people within a given society are assigned to the role of “citizen” and are expected to adhere to the

A Systematic Survey of Online Data Mining Technology Intended for Law Enforcement

A systematic mapping study describing online data-mining literature which visibly targets law enforcement applications is described, using evidence-based practices in survey making to produce a replicable analysis which can be methodologically examined for deficiencies.

ReaderBench: Automated evaluation of collaboration based on cohesion and dialogism

Two computational models for assessing collaboration are proposed and validated based on a cohesion graph and a cohesion-based model of discourse, which can be perceived as a longitudinal analysis of the ongoing conversation and enabling a transversal analysis of subsequent discussion slices.



Entrapping the Innocent: Toward a Theory of Child Sexual Predators’ Luring Communication

This paper examines an unexplored form of communication deviance—the communicative process of entrapment used by child sexual predators to lure their victims into an ongoing sexual relationship.

Supporting Law Enforcement in Digital Communities through Natural Language Analysis

An approach to provide automated support for the detection of child sexual abuse related activities in digital communities is presented and an overview of how this approach can be extended to police chat and social networking communities is given.

Topic Detection and Extraction in Chat

Preliminary methods of topic detection and topic thread extraction that augment a typical TF-IDF-based vector space model approach with temporal relationship information between posts of the Chat dialog combined with WordNet hypernym augmentation are presented.

Toward Spotting the Pedophile Telling victim from predator in text chats

  • N. Pendar
  • Computer Science
    International Conference on Semantic Computing (ICSC 2007)
  • 2007
This paper presents the results of a pilot study on using automatic text categorization techniques in identifying online sexual predators. We report on our SVM and k-NN models. Our distance weighted

ChatCoder: Toward the Tracking and Categorization of Internet Predators

The software used to facilitate analysis of chat log transcripts and the development of a communicative theory of online predation is presented, which shows that one can distinquish between predator and victim communication, but not as reliably as one would like to.

Luring language and virtual victims: Coding cyber-predators online communicative behavior

  • Luring language and virtual victims: Coding cyber-predators online communicative behavior
  • 2009