Title
On the early detection of threats in the real world based on open-source information on the internet
Author
Bouma, H.
Rajadell Rojas, O.
Worm, D.T.H.
Versloot, C.A.
Wedemeijer, H.
Publication year
2012
Abstract
Many threats in the real world can be related to activities of persons on the internet. Internet surveillance aims to detect threats in an early stage and to assist in finding suspects based on information from the web. However, the amount of data on the internet rapidly increases and it is time consuming to monitor many open sources. In this paper, we present a method to find abnormal behavior on the internet and give an early warning for threats. The system was tested on Twitter data. The results show that it can successfully analyze the content of messages and recognize abnormal changes in sentiment and threatening content.
Subject
Anomaly detection
Internet surveillance
Data mining
Cyber security
Pattern recognition
Safety and Security
Defence, Safety and Security
Physics & Electronics ; Organisation ; Communication & Information
II - Intelligent Imaging ; NO - Networked Organisations ; MNS - Media & Network Services
TS - Technical Sciences ; BSS - Behavioural and Societal Sciences
To reference this document use:
http://resolver.tudelft.nl/uuid:4ab1af72-b6dd-46f5-b091-638f9d11252c
TNO identifier
464757
Source
International Conference on Information Technologies and Security ITSEC-2012, 15-16 October 2012 , Chisinau, Republic of Moldova
Document type
conference paper