Natural Language Processing Laboratory

FarsNet is an ongoing project to develop a lexical ontology to cover Persian words and phrases. It is designed to contain a Persian WordNet in its first phase and grow to cover verbs' argument structures (thematic roles and their selectional restrictions) in its second phase. Thus FarsNet consists of two main components:

•  Persian WordNet 
• Persian Net of verb frames (PeVNet)

Phase one: Persian WordNet

The first release of FarsNet 1.0 which just contains Persian WordNet will be available by the end of 2009. It contains lexical, syntactic and semantic knowledge about more than 15000 Persian words and phrases organized in about 10000 synsets of nouns, adjectives and verbs. The included words and phrases are selected according to BalkaNet base concepts and the most frequent Persian words and phrases in our corpora. FarsNet 1.0 relates synsets in each POS category by the set of WordNet 2.1 relations. In the current version inter-POS relations are not implemented.

FarsNet has also inter-lingual relations connecting Persian synsets to English ones (in Princeton WordNet 3.0). Persian WordNet goes closely in the lines and principles of Major WordNets such as Princeton WordNet, EuroWordNet and BalkaNet to maximize its compatibility to these WordNets and to be connected to the other WordNets in the world to enable cross lingual tasks such as MT, multilingual IR and developing multilingual dictionaries and thesauri.

 Team Members

• Dr. Mehrnoush Shamsfard (Head)

 Active members (currently working on the project)

• Dr. Mostafa Assi
• Dr. Niloofar Mansoori
• Dr. Ali Famian
• Akbar Hesabi
• Hakimeh Fadaee
• Somayeh Bagherbeigi 
• Elham Fekri
• Maliheh Monshizadeh
• Bahram Jahanshahi

Passive members (has worked on the project for a period of time)

• Rahim Dehkharghani 
• Masoud Rouhizadeh 
• Mahsa Arab Yarmohammadi

Sponser

This first phase of FarsNet is partly funded by Iran Telecommunication Research Center (ITRC) under contract No T/500/19231

Publications:

1. Shamsfard M., Hesabi A., Fadaei H., Mansoory N., Famian A., Bagherbeigi S., Fekri E., Monshizadeh M., Assi M.: Semi Automatic Development of FarsNet; The Persian WordNet, Global WordNet Conference, Mumbai, India, 2010.
2. Akbar Hesabi, Mostafa Assi, Mehrnoush Shamsfard, Mahsa A. Yarmohammadi, Design and Development of Persian Noun Wordnet, International Journal of Information and Communication Technology (IJICT), 2009.
3. Rahim Dehkharghani Mehrnoush Shamsfard, Mapping Persian Words to English Synsets, To be appeared at International Journal of Interactive Multimedia and Artificial Intelligence (IJMAI).
4. Rahim Dehkharghani Mehrnoush Shamsfard, Automatic Alignment of Persian and English Lexical Resources: A Structural-Linguistic Approach, The 6th International Workshop on Natural Language Processing and Cognitive Science (NLPCS 2009) - Milan, Italy, 6-7 May, 2009.
5. Somayeh Bagherbeigi, Mehrnoush Shamsfard, A Comparative Study of WordNets: A Proposal for Persian Lexical Network Structure, Third International Conference of Cognitive Science, Tehran, Iran (2009).
6. Shamsfard, M., Developing FarsNet: A Lexical Ontology for Persian, 4th Global WordNet Conference (GWC’08), 2008.
7. Rouhizadeh, M., Shamsfard, M., Yarmohammadi, M. A., Building a WordNet for Persian Verbs, 4th Global WordNet Conference (GWC’08), 2008.
8. Shamsfard, M., Towards Semi Automatic Construction of a Lexical Ontology for Persian, The sixth international conference on Language Resources and Evaluation (LREC’2008), Morocco., Morocco, 2008.