Her research project aims to overcome the problem caused by data sparsity in the automatic development of language resources, based on distributional information. She investigates a method to discover more possible co-occurring words for low-frequent words, besides the co-occurrences existent in corpus. Her approach relies on a similarity measure between words, based on statistical information of the words usage in context. She expects that this measure approves a transfer of context from a frequent word to a low-frequent word.
- 2009-2010: Master Degree specialized in “Web Intelligence”, Université “Jean Monnet” Saint-Etienne, FRANCE
- 2005-2008: Undergraduate Degree, – Universitatea “Transilvania” Brasov, ROMANIA
- Silvia Necsulescu, Núria Bel, Muntsa Padró, Montserrat Marimon and Eva Revilla: Towards the Automatic Merging of Language Resources. In Proceedings of WoLeR 2011. Ljubljana, Slovenia. pdf
- Muntsa Padró, Núria Bel, Silvia Necsulescu: Towards the Automatic Merging of Lexical Resources:Automatic Mapping. In Proceedings of RANLP 2011. Hissar, Bulgaria.
- Silvia Necsulescu: Automatic acquisition of co-occurrences for low frequent words, In Proceedings of RANLP 2011 Student Research Workshop, Hissar, Bulgaria.
- Muntsa Padró, Núria Bel, Silvia Necsulescu: A Method Towards the Fully Automatic Merging of Lexical Resources. In Proceedings of LRTS Sharing Workshop 2011, Chiang Mai, Thailand
silvia.necsulescu at upf.edu