While I haven't done any serious research in the last few years, I'm interested in natural language processing and semantics. During my Masters, I worked on distributional semantic models.
I published two papers:
-
2009 - 2011: DSM Methods for Measuring Semantic Similarity
For M.Sc., worked on research and implementation of distributional semantic models (DSMs) for Croatian
language. DSMs dynamically build semantic word representations in the form of multi-dimensional vector
spaces through a statistical analysis of the contexts in which words occur. Geometrical relations
between vectors correlate with semantic relations between represented words.
-
2011: Hyperlink Constituency
Project lead: Michael Yoshitaka Erlewine, MIT
Investigating the syntactic notion of constituency by studying inline hyperlinks in a hypertext corpus,
the project hypothesis being that inline links are constituents in their host sentences.
-
2011: Corpus Explorer
Faculty of Electrical Engineering and Computing, University of Zagreb, Croatia
Mentor: Prof. Bojana Dalbelo-Bašić, Ph.D.
Implemented an explorer and visualizer tool for a large corpora set of Croatian journal articles
provided by Croatian News Agency.
-
2010: Assessment of multicore programming languages
Mentor: Ivan Skuliber, Ph.D.
During Ericsson Summer camp worked on evaluating multicore programming libraries for C++, Java and
Erlang by implementing and testing parallel text processing, process management and message parsing.
Implemented universal GLR parser in Erlang.
-
2009: PROTO: Multicore SIP parser
Mentor: Ivan Skuliber, Ph.D.
During Ericsson Summer camp designed and implemented a 100% compliant SIP parser for use in multicore
processor systems.