The increase in the amount of data generated by modern technologies has resulted in an increasing need for context awareness. Context provides the boundaries within which we can transition from data to relevant information although context may not be explicitly present in the data itself. The interpretation of data which leads to the extraction of information, changes or varies when the context changes.

In this thesis we examine data and create a contextual model that seamlessly combines data elements of a domain in order to effectively locate and provide the most appropriate information for the user according to his or her needs. Much of this contextual information becomes specialized or tailored to one special domain or environment and hence becomes completely unusable for other domains or environments. We propose a generic system design for modeling and representing any contextual model for any domain. We also propose and implement an automatic solution for creating contextual models for a particular domain or a business environment.

We demonstrate the use of contextual information and semantic techniques with the implementation of a prototype in the application domain of identifying potential threats associated with the shipments from the perspective of U.S. Federal agencies. We implemented a set of tools and technologies to assist in the process of creation of a contextual model as well as semantic networks. The experimental evaluation of our methodology shows that our techniques are promising and they produce better precision results compared to the scenario when these techniques are not considered.