Transformer based Answer-Aware Bengali Question Generation
Question generation (QG), the task of generating questions from text or other forms of data, a significant and challenging subject, has recently attracted more attention in natural language processing (NLP) due to its vast range of business, healthcare, and education applications through creating qu...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
KeAi Communications Co., Ltd.
2023-06-01
|
| Series: | International Journal of Cognitive Computing in Engineering |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666307423000311 |
| _version_ | 1797377840535044096 |
|---|---|
| author | Jannatul Ferdous Ruma Tasmiah Tahsin Mayeesha Rashedur M. Rahman |
| author_facet | Jannatul Ferdous Ruma Tasmiah Tahsin Mayeesha Rashedur M. Rahman |
| author_sort | Jannatul Ferdous Ruma |
| collection | DOAJ |
| description | Question generation (QG), the task of generating questions from text or other forms of data, a significant and challenging subject, has recently attracted more attention in natural language processing (NLP) due to its vast range of business, healthcare, and education applications through creating quizzes, Frequently Asked Questions (FAQs) and documentation. Most QG research has been conducted in languages with abundant resources, such as English. However, due to the dearth of training data in low-resource languages, such as Bengali, thorough research on Bengali question generation has yet to be conducted. In this article, we propose a system for producing varied and pertinent Bengali questions from context passages in natural language in an answer-aware input format using a series of fine-tuned text-to-text transformer (T5) based models. During our studies with various transformer-based encoder-decoder models and various decoding processes, along with delivering 98% grammatically accurate questions, our fine-tuned BanglaT5 model had the highest 35.77 F-score in RougeL and 38.57 BLEU-1 score with beam search. Our automated and human evaluation results show that our answer-aware QG models can create realistic, human-like questions relevant to the context passage and answer. We also release our code, generated questions, dataset, and models to enable broader question generation research for the Bengali-speaking community. |
| first_indexed | 2024-03-08T19:58:09Z |
| format | Article |
| id | doaj.art-6ac0a7bdff784c67988fd60e27214247 |
| institution | Directory Open Access Journal |
| issn | 2666-3074 |
| language | English |
| last_indexed | 2024-03-08T19:58:09Z |
| publishDate | 2023-06-01 |
| publisher | KeAi Communications Co., Ltd. |
| record_format | Article |
| series | International Journal of Cognitive Computing in Engineering |
| spelling | doaj.art-6ac0a7bdff784c67988fd60e272142472023-12-24T04:46:43ZengKeAi Communications Co., Ltd.International Journal of Cognitive Computing in Engineering2666-30742023-06-014314326Transformer based Answer-Aware Bengali Question GenerationJannatul Ferdous Ruma0Tasmiah Tahsin Mayeesha1Rashedur M. Rahman2Department of Electrical & Computer Engineering, North South University, Dhaka 1229, BangladeshDepartment of Electrical & Computer Engineering, North South University, Dhaka 1229, BangladeshCorresponding author.; Department of Electrical & Computer Engineering, North South University, Dhaka 1229, BangladeshQuestion generation (QG), the task of generating questions from text or other forms of data, a significant and challenging subject, has recently attracted more attention in natural language processing (NLP) due to its vast range of business, healthcare, and education applications through creating quizzes, Frequently Asked Questions (FAQs) and documentation. Most QG research has been conducted in languages with abundant resources, such as English. However, due to the dearth of training data in low-resource languages, such as Bengali, thorough research on Bengali question generation has yet to be conducted. In this article, we propose a system for producing varied and pertinent Bengali questions from context passages in natural language in an answer-aware input format using a series of fine-tuned text-to-text transformer (T5) based models. During our studies with various transformer-based encoder-decoder models and various decoding processes, along with delivering 98% grammatically accurate questions, our fine-tuned BanglaT5 model had the highest 35.77 F-score in RougeL and 38.57 BLEU-1 score with beam search. Our automated and human evaluation results show that our answer-aware QG models can create realistic, human-like questions relevant to the context passage and answer. We also release our code, generated questions, dataset, and models to enable broader question generation research for the Bengali-speaking community.http://www.sciencedirect.com/science/article/pii/S2666307423000311Question GenerationNatural Language ProcessingText-to-text transformerBanglaT5 |
| spellingShingle | Jannatul Ferdous Ruma Tasmiah Tahsin Mayeesha Rashedur M. Rahman Transformer based Answer-Aware Bengali Question Generation International Journal of Cognitive Computing in Engineering Question Generation Natural Language Processing Text-to-text transformer BanglaT5 |
| title | Transformer based Answer-Aware Bengali Question Generation |
| title_full | Transformer based Answer-Aware Bengali Question Generation |
| title_fullStr | Transformer based Answer-Aware Bengali Question Generation |
| title_full_unstemmed | Transformer based Answer-Aware Bengali Question Generation |
| title_short | Transformer based Answer-Aware Bengali Question Generation |
| title_sort | transformer based answer aware bengali question generation |
| topic | Question Generation Natural Language Processing Text-to-text transformer BanglaT5 |
| url | http://www.sciencedirect.com/science/article/pii/S2666307423000311 |
| work_keys_str_mv | AT jannatulferdousruma transformerbasedanswerawarebengaliquestiongeneration AT tasmiahtahsinmayeesha transformerbasedanswerawarebengaliquestiongeneration AT rashedurmrahman transformerbasedanswerawarebengaliquestiongeneration |