Panos Kostakos

The ‘good’, the ‘bad’ and the ‘Charlie’: the business of cocaine smuggling in Greece

This article examines the social organisation of cocaine smuggling in Greece. Emphasis is placed on the involvement of professionals from the shipping industry and actors from the ‘upper society echelons’ who play a pivotal role in the transportation and importation of cocaine to Western Europe and Greece. After considering empirical evidence from a variety of sources, our findings indicate that the cocaine market in Greece is ‘organised’ by a system of collaborative relationships between state, business and civil society actors. It is suggested that to better understand the nature of this illegal market, further research is required to take a closer look into the economic, socio-cultural and political incentives of these actors.

Inferring social networks from physical interactions: a feasibility study

Purpose – The purpose of this paper is to present a prototype system that can be used to capture longitudinal socialising processes by recording peoples encounters in space.Design/methodology/approach – The paper presents the results of a longitudinal study, carried out with members of the public, which demonstrates the capabilities of the system.Findings – The findings show that community structure can be inferred from physical interactions, and that different locations exhibit varying community structures.Originality/value – The paper argues that such a system can usefully be deployed in environments where people interact and socialise, as a mechanism for inferring the underlying network structure of those peoples relationships.

On Web Based Sentence Similarity for Paraphrasing Detection.

Semantic similarity measures play vital roles in information retrieval, natural language processing and paraphrasing detection. With the growing plagiarisms cases in both commercial and research community, designing efficient tools and approaches for paraphrasing detection becomes crucial. This paper contrasts web-based approach related to analysis of snippets of the search engine with WordNet based measure. Several refinements of the web-based approach will be investigated and compared. Evaluations of the approaches with respect to Microsoft paraphrasing dataset will be performed and discussed.