Giorgos Stamatelatos

Software Engineer

Software Engineer

• Angular JS, Vue JS, Python Backend, Drupal
• Business Plan app
• Automatic job postings aggregation app

Research Assistant

Funded project PROMOTE: Tools for Promoting Tourism Experience

Social Network Analysis, Tourist Personalized Recommendations, Graph Analysis Algorithms, Data Mining from Social Media, Development of full stack application for data mining, presentation of points-of-interest, personalized recommendations, and access through API: Promote Web UI.

Teaching Assistant

• Data Structures
• Algorithms and Complexity

Research Assistant

Funded project AKSEID: Assessment of News Reliability in Social Networks of Influence

Social Network Analysis, Assessment of the political orientation of politicians and news media, Graph theoretic algorithms, Social network mining, Design and implementation of automatic monitoring of social media.

Research Assistant

• Social Network Analysis, Games in Social Networks, Influence Maximization in Social Graphs.
• Development of social network analysis software.

Network Administrator

Network operation responsibilities during educational certification exams.

Research Assistant

Funded project ATLAS: Advanced Tourism pLAnning System

ATLAS seeks to create a step change in tourism services by advancing the state-of-the-art of the algorithms for fast geo-aware social media tagging and annotation and geo-aware multimedia information retrieval and recommendation. In particular, ATLAS aims at an innovative breakthrough in combining the just mentioned technologies for practical tourism services at a quality level that meets the expectations of the users.

• Tagging and commenting of data from social media, development of algorithms for personalized points-of-interest recommendation.
• Development of full stack application for personalized recommendations, maps, visualizations, implementation of algorithms and exposure via API.

Democritus University of Thrace

Algorithmic Analysis of User Behavior in Social Media

In the work presented in this thesis we use both established and novel structural methodology social network analysis in order to recognize the online behavior of users and utilize the observations in order to attempt to solve problems that commonly arise in online social networks. Our assumption, that is confirmed –to an extent– through our analysis, is that users will behave consistently with their interests, beliefs or tastes. This phenomenon gives rise to patterns and structure to real social networks as users have the tendency to create connections with other users or other entities based on criteria of interest, taste or exposure. Our methods are exclusively structural, i.e. they rely on the links formed in an explicit or implicit network between the relevant entities, which may not be necessarily formed among real users. In summary, our analysis is twofold: (1) structural analysis and (2) based on the expected behavior of users.

Democritus University of Thrace

Algorithmic Techniques for Influence Dynamics in Social Networks

In this thesis, a non-cooperative, zero-sum game of influence between two firms on a social network is defined and examined. The social network is represented with a directed graph of n individual agents with an initial opinion on a particular subject. The DeGroot model defines the opinion dynamics and the rules of social interaction, according to which the belief of each individual is influenced by the opinions of their neighbors. The players of the game are two external entities who have diametrically opposed opinions. Each player tries to manipulate the community in their own opinion’s favor, by actively interfering with the social graph structure. We investigate possible equilibria, identify basic principles for a successful strategy and compare the performance of several algorithms implementing player strategies.

Democritus University of Thrace

Distributed Search among Agents with Enhanced Privacy

In this thesis, a new platform, called Quantum, for distributed computations among independent agents, is presented. Quantum is capable to operate on infrastructures that are formed by massive numbers of agents communicating over the Internet, and is tolerant of adding/removing peers. Furthermore, the new platform is intended to support the privacy of the peers which participate in each distributed computation. In the proposed solution we have chosen a decentralized network architecture and exploited technologies of peer-to-peer networks.