ABSTRACT
This paper describes the approach we adopted during official submission of FIRE-2013 Shared Task on Transliterated Search along with few other approaches that we experimented post-submission. The techniques solve the problem of language labeling, by identifying query word as English or Hindi (E or H) term in mixed language sentence queries. Manual and machine learning algorithms are used. For the transliteration of H labeled word we use manual (dictionary based), generative (grapheme based) and combination of both in different algorithms. We observe that learning based classification improves labeling accuracy. Extraction based transliteration gives better result than Generation based when the terms are available in bilingual dictionary. But it may lead to incorrect transliteration if terms are wrongly aligned and the approach fails for out-of-dictionary words. In this case transliteration by generation is the only alternative. But generation alone does not perform well because of spelling variation in transliterated terms. During evaluation we also observe that transliteration systems are generally corpus-biased. Although our performance in the official submission was moderate, we obtain better results during our post-submission experiments.
- Chinnakotla, M. K., Damani, O. P., and Satoskar, A. Transliteration for resource-scarce languages. ACM Transactions on Asian Language Information Processing (TALIP) 9, 4 (2010), 14. Google ScholarDigital Library
- Chinnakotla, M. K., Ranadive, S., Damani, O. P., and Bhattacharyya, P. Hindi to english and marathi to english cross language information retrieval evaluation. In Advances in Multilingual and Multimodal Information Retrieval. Springer, 2008, pp. 111--118. Google ScholarDigital Library
- Choudhury, M., Majumder, P., Roy, R. S., and Agarwal, K. Fire shared task on transliterated search. http://research.microsoft.com/en-us/events/fire13_st_on_transliteratedsearch/default.aspx, 2013. Online; accessed 10-09-2013.Google Scholar
- Dale, R. Language technology. Slides of HCSNet Summer School Course. Sydney (2007).Google Scholar
- Das, A., Ekbal, A., Mandal, T., and Bandyopadhyay, S. English to hindi machine transliteration system at news 2009. In Proceedings of the 2009 Named Entities Workshop: Shared Task on Transliteration (2009), Association for Computational Linguistics, pp. 80--83. Google ScholarDigital Library
- El-Kahky, A., Darwish, K., Aldein, A. S., El-Wahab, M. A., Hefny, A., and Ammar, W. Improved transliteration mining using graph reinforcement. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (2011), Association for Computational Linguistics, pp. 1384--1393. Google ScholarDigital Library
- Gupta, K., Choudhury, M., and Bali, K. Mining hindi-english transliteration pairs from online hindi lyrics. In LREC (2012), pp. 2459--2465.Google Scholar
- Karimi, S., Scholer, F., and Turpin, A. Machine transliteration survey. ACM Computing Surveys (CSUR) 43, 3 (2011), 17. Google ScholarDigital Library
- Khapra, M. M., and Bhattacharyya, P. Improving transliteration accuracy using word-origin detection and lexicon lookup. In Proceedings of the 2009 Named Entities Workshop: Shared Task on Transliteration (2009), Association for Computational Linguistics, pp. 84--87. Google ScholarDigital Library
- King, B., and Abney, S. Labeling the languages of words in mixed-language documents using weakly supervised methods. In Proceedings of NAACL-HLT (2013), pp. 1110--1119.Google Scholar
- Klein, D. The stanford classifier. http://http://nlp.stanford.edu/software/classifier.shtml, 2003. Online; accessed 19-02-2014.Google Scholar
- Knight, K., and Graehl, J. Machine transliteration. Computational Linguistics 24, 4 (1998), 599--612. Google ScholarDigital Library
- Kumaran, A., Khapra, M. M., and Bhattacharyya, P. Compositional machine transliteration. ACM Transactions on Asian Language Information Processing (TALIP) 9, 4 (2010), 13. Google ScholarDigital Library
- Malik, A., Besacier, L., Boitet, C., and Bhattacharyya, P. A hybrid model for urdu hindi transliteration. In Proceedings of the 2009 Named Entities Workshop: Shared Task on Transliteration (2009), Association for Computational Linguistics, pp. 177--185. Google ScholarDigital Library
- McCallum, A. K. Mallet: A machine learning for language toolkit. http://mallet.cs.umass.edu/download.php, 2002. Online; accessed 19-02-2014.Google Scholar
- Rama, T., and Gali, K. Modeling machine transliteration as a phrase based statistical machine translation problem. In Proceedings of the 2009 Named Entities Workshop: Shared Task on Transliteration (2009), Association for Computational Linguistics, pp. 124--127. Google ScholarDigital Library
- Roy, R. S., Choudhury, M., Majumder, P., and Agarwal, K. Overview and datasets of fire 2013 track on transliterated search. In Pre-proceedings of the FIRE 5th workshop (2013). Google ScholarDigital Library
- Sharma, S., Bora, N., and Halder, M. English-hindi transliteration using statistical machine translation in different notation. Training 20000, 297380 (2012).Google Scholar
- Singh, P. RomaDeva: English(roman) to hindi(devanagri) transliteration tool. https://code.google.com/p/romadeva/downloads/list, 2012. Online; accessed 19-02-2014.Google Scholar
- Sowmya, V., Choudhury, M., Bali, K., Dasgupta, T., and Basu, A. Resource creation for training and testing of transliteration systems for indian languages. In LREC (2010).Google Scholar
- Yoon, S.-Y., Kim, K.-Y., and Sproat, R. Multilingual transliteration using feature based phonetic method. In Annual Meeting-Association for Computational Linguistics (2007), vol. 45(1), Citeseer, pp. 112--119.Google Scholar
Index Terms
- ISM@FIRE-2013 Shared Task on Transliterated Search
Recommendations
IIIT-H System Submission for FIRE2014 Shared Task on Transliterated Search
FIRE '14: Proceedings of the 6th Annual Meeting of the Forum for Information Retrieval EvaluationThis paper describes our submission for FIRE 2014 Shared Task on Transliterated Search. The shared task features two sub-tasks: Query word labeling and Mixed-script Ad hoc retrieval for Hindi Song Lyrics.
Query Word Labeling is on token level language ...
ISM@FIRE-2014: Shared Task on Transliterated Search
FIRE '14: Proceedings of the 6th Annual Meeting of the Forum for Information Retrieval EvaluationThis paper describe approaches we used for the Shared Task on Transliterated Search in FIRE-2014. The approaches solve identification of native languages of given terms/words and their labeling. MaxEnt a supervised classifier is used for the ...
Overview of the FIRE 2013 Track on Transliterated Search
FIRE '12 & '13: Proceedings of the 4th and 5th Annual Meetings of the Forum for Information Retrieval EvaluationIn this paper, we provide an overview of the FIRE 2013 track on transliterated search and describe the datasets released as part of the track. This was the first year that the track was organized. We had proposed two subtasks as part of the challenge. ...
Comments