In this paper, we assess the challenges for multi-domain, multi-lingual question answering, create necessary resources for benchmarking and develop a baseline model. We curate 500 articles in six different domains from the web. These articles form a comparable corpora of 250 English documents and 250 Hindi documents. From these comparable corpora, we have created 5; 495 question-answer pairs with the questions and answers, both being in English and Hindi. The question can be both factoid or short descriptive types. The answers are categorized in 6 coarse and 63 finer types. To the best of our knowledge, this is the very first attempt towards creating multi-domain, multi-lingual question answering evaluation involving English and Hindi. We develop a deep learning based model for classifying an input question into the coarse and finer categories depending upon the expected answer. Answers are extracted through similarity computation and subsequent ranking. For factoid question, we obtain an MRR value of 49:10% and for short descriptive question, we obtain a BLEU score of 41:37%. Evaluation of question classification model shows the accuracies of 90:12% and 80:30% for coarse and finer classes, respectively.
@InProceedings{GUPTA18.826, author = {Deepak Gupta and Surabhi Kumari and Asif Ekbal and Pushpak Bhattacharyya}, title = "{MMQA: A Multi-domain Multi-lingual Question-Answering Framework for English and Hindi}", booktitle = {Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)}, year = {2018}, month = {May 7-12, 2018}, address = {Miyazaki, Japan}, editor = {Nicoletta Calzolari (Conference chair) and Khalid Choukri and Christopher Cieri and Thierry Declerck and Sara Goggi and Koiti Hasida and Hitoshi Isahara and Bente Maegaard and Joseph Mariani and Hélène Mazo and Asuncion Moreno and Jan Odijk and Stelios Piperidis and Takenobu Tokunaga}, publisher = {European Language Resources Association (ELRA)}, isbn = {979-10-95546-00-9}, language = {english} }