After graduating from computer science at University of Ljubljana in 2002 Peter was employed as a researcher at the Department of Knowledge Technologies at Jožef Stefan Institute for four years. Following this he collaborated on three European IST projects: Sol-Eu-Net, cInQ, and Ecolead. His main focus was data analysis, which included tasks such as mining 21 years of UK Traffic accident data, modeling trust in networked organizations and automatic ontology construction from the set of textual documents. He developed a state-of-the-art method named CIPER for inducing polynomial equations, published numerous scientific papers (IEEE Transactions), attended several international scientific conferences covering machine learning and data mining, and was assisting the social networking course on International postgraduate school.

From 2006 on Peter has been running his own company which was subcontracted by CME (Central European Media Enterprises). There he worked as team leader to develop news portals for TV stations in several countries (24ur.com, dnevnik.hr, www.1plus1.ua). Then he was working on videolectures.net, is happy that he wasted lots of precious time competing on the Netflix prize, was (and still is) developing recommender systems for websites aliiike.com, and is currently helping with Slovene news aggregator times.si. Recently he finished developing a web portal for personalized printing and selling prints named eggotrip.

His main interests are: computer science, machine learning, data mining, knowledge representation, social network analysis, natural language processing, web scraping, web development. Favorite tools and platforms are the following:

• programming languages: c, c++, python, php
• platforms: linux, django, apache, mysql
• data analysis tools: R, Weka, Cluto, Pajek

More you can find on his personal blog.