Recent advances in the Semantic Web technologies and the abundance of highly heterogeneous and disparate information on the web opens a new path to the analysis of terrorist networks. Numerous systems have been collaboratively developed between agencies and academia applying different metaphors in order to untangle terrorist activities. Among them, the network metaphor is the most popular, and social network analysis, visualization and data mining methods are employed to assist analysts in the discovery of structural characteristics and patterns of interaction. Although, Natural Language Processing and Information Extraction have matured to the point where the conversion of freeform documents to these diagrams can be largely automated, existing approaches provide a restrictive view of the entities and relations using simplified statistical approaches. In addition, constructed networks are considered static, thus inhibiting the discovery of temporal patterns. This paper focuses on the network construction process by utilizing advanced NLP and IE techniques. The Parmenides framework proposed a set of techniques and a methodology to the consistent and efficient gathering, organization and dissemination of relevant information towards the effective construction and analysis of terrorist networks.