​​​​PavlosParaskevopoulos​

I work in the fields of Big Data Analytics, Data Mining and Knowledge Discovery, and my interests are currently concentrating on Text Mining for domain experts and on the discovery of patterns and hidden connections in Heterogeneous Information Networks (HIN). More precisely, the target of my work is the development of methods that can be used by the domain experts for the fast and accurate revealing of connections and patterns in a large, uncertain and potentially heterogeneous dataset.

​

The application of my methods on the appropriate datasets could lead to the identification of important non-obvious connections and prediction of undesirable situations in evolving environments, allowing the domain experts to make fast and more accurate decisions. An example of the applications that my methods could be useful for is the connection of genomes with later manifested illnesses, allowing the doctors to get decisions for the prevention of the illnesses.

​

The techniques that I apply on the data are mainly machine learning and data mining algorithms. My target is the development of a framework that can be applied on any type of data, quickly providing accurate non-obvious insights.

I am currently working on data describing the financial and social connections of persons and companies. The dataset was provided to us by the Schar School of Policy and Government in the scope of the multidisciplinary research project DIVES and target of this collaboration is the development of methods that are going to help the domain experts to identify patterns of interaction of key players in the finance and government sector.

​

During my PhD, I have worked on the extraction of interesting human behaviors, based on mobile phone and social media activities. More precisely, the target of my work was the event identification and prediction of undesirable situations (e.g. traffic jams), by using datasets containing calling activity (e,g, number of calls, SMS, etc.) and social activity (e.g tweets).

​

I have participated at the “Transportation and Mobility Hackathon” organized by Fairfax County and Virginia Department of Transportation (Nov 2017), serving as a mentor of the winning team. In the past, I have participated at the “Data for Development (D4D) Challenge” organized by Orange (2012) and I am co-author of the study “Identification and Characterization of Human Behavior Patterns from Mobile Phone Data” presented at the International Conference on the Analysis of Mobile Phone Datasets (NetMob), Special Session on the Data for Development (D4D) Challenge (May 2013).

​

​