About the system

A traditional search engine, given a text query, provides as a results a list of related documents. This system is a "lexico-semantic search engine". Given a query it returns a list of related words. This way you can discover meaning of words in an interactive manner, search for related words and more. This project is a result of a colloboration between Universite catholique de Louvain (UCL) and Moscow State Technical University (BMSTU). Several people contributed to the project: Alexander Panchenko, Alexey Romanov, Andrey Philippovich, Cedrick Fairon, Hubert Naets, Laeticia Browers, Olga Morozova, Sergey Adeykin, Pavel Romanov, Yuri Philippovich, Otto Vale, and Artem Lukanin. We would like to thank the following organizations for support: Wallonie-Brussels International, Center for Natural Language Processing of UCL, IT-CLAIM of BMSTU and Russian Humanities Foundation. Currently the system is maintained by Language Technology Group of Hamburg University and hosted using the generous support of UCL. Original design of the logo and the web site was developed by Maxim Kolesov (Midstripe Teurgy).

If you would like to know more about the system or would like to refer to it in a publication, please refer to the following paper: Panchenko et al. (2013) Serelex: Search and visualization of semantically related words. In Proceedings of the European Conference on Information Retrieval, ECIR'2013. Springer.

@inproceedings{panchenko2013serelex, 
   title={Serelex: Search and visualization of semantically related words},
   author={Panchenko, Alexander and Romanov, Pavel and Morozova, Olga and Naets, Hubert and Philippovich, Andrey and Romanov, Alexey and Fairon, Cedrick},
   booktitle={European Conference on Information Retrieval},
   pages={837--840},
   year={2013},
   organization={Springer}
}
The paper mentioned above provides a general introduction of the system. Further information about variations of the systems for different languages featuring a detailed descriptions of the PatternSim similarity measure can be found in the publications below: