Framework for surveillance of instant messages in instant messengers and social neworking sites using data mining and ontology

Abstract

Innumerable terror and suspicious messages are sent through Instant Messengers (IM) and Social Networking Sites (SNS) which are untraced, leading to hindrance for network communications and cyber security. We propose a Framework that discover and predict such messages that are sent using IM or SNS like Facebook, Twitter, LinkedIn, and others. Further, these instant messages are put under surveillance that identifies the type of suspected cyber threat activity by culprit along with their personnel details. Framework is developed using Ontology based Information Extraction technique (OBIE), Association rule mining (ARM) a data mining technique with set of pre-defined Knowledge-based rules (logical), for decision making process that are learned from domain experts and past learning experiences of suspicious dataset like GTD (Global Terrorist Database). The experimental results obtained will aid to take prompt decision for eradicating cyber crimes.