Open domain question answering system

Deep learning methods have drawn tremendous attention from both the research community and the industrial practitioners thanks to their undeniable power in learning feature representation in higher dimensions without manual, handcrafting features. An application of deep learning that arises naturall...

Full description

Saved in:
Bibliographic Details
Main Author: Hoang, Nghia Tuyen
Other Authors: Joty Shafiq Rayhan
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2022
Subjects:
Online Access:https://hdl.handle.net/10356/156955
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-156955
record_format dspace
spelling sg-ntu-dr.10356-1569552023-02-28T23:17:28Z Open domain question answering system Hoang, Nghia Tuyen Joty Shafiq Rayhan Pun Chi Seng School of Physical and Mathematical Sciences cspun@ntu.edu.sg, srjoty@ntu.edu.sg Engineering::Computer science and engineering::Information systems::Information storage and retrieval Science::Mathematics Deep learning methods have drawn tremendous attention from both the research community and the industrial practitioners thanks to their undeniable power in learning feature representation in higher dimensions without manual, handcrafting features. An application of deep learning that arises naturally is question answering, in which a question answering system must answer questions posed by humans. One of its sub-fields, opendomain question answering, attempts to answer questions about nearly anything, without being given relevant reference texts. Despite its impactful applications in search engines, chatbots and factual correction, research work in open-domain question answering is relatively under-explored due to its complex and large-scale nature. In this work, we aim to advance the progress of recent open-domain question answering systems by developing various mathematical-driven methods. More specifically, in the first part of this thesis, we introduce the widely adopted two-stage paradigm in opendomain question answering and perform comprehensive error analysis on state-of-the-art models. Based on this, we are then able to formulate and develop methods aiming specifically at overcoming these weaknesses in the second part of the thesis. These approaches range from simple methods such as parameter sharing and data augmentation to more sophisticated methods such as designing new objective functions or pseudo data synthesis and semi-supervised learning. Finally, we unify these developed methods into a single framework that outperforms state-of-the-art models by a significant margin on common benchmarking datasets. The code to reproduce our experiments is released at https://github.com/hnt4499/DPR. Bachelor of Science in Mathematical Sciences 2022-05-04T03:33:54Z 2022-05-04T03:33:54Z 2022 Final Year Project (FYP) Hoang, N. T. (2022). Open domain question answering system. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/156955 https://hdl.handle.net/10356/156955 en application/pdf Nanyang Technological University
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Computer science and engineering::Information systems::Information storage and retrieval
Science::Mathematics
spellingShingle Engineering::Computer science and engineering::Information systems::Information storage and retrieval
Science::Mathematics
Hoang, Nghia Tuyen
Open domain question answering system
description Deep learning methods have drawn tremendous attention from both the research community and the industrial practitioners thanks to their undeniable power in learning feature representation in higher dimensions without manual, handcrafting features. An application of deep learning that arises naturally is question answering, in which a question answering system must answer questions posed by humans. One of its sub-fields, opendomain question answering, attempts to answer questions about nearly anything, without being given relevant reference texts. Despite its impactful applications in search engines, chatbots and factual correction, research work in open-domain question answering is relatively under-explored due to its complex and large-scale nature. In this work, we aim to advance the progress of recent open-domain question answering systems by developing various mathematical-driven methods. More specifically, in the first part of this thesis, we introduce the widely adopted two-stage paradigm in opendomain question answering and perform comprehensive error analysis on state-of-the-art models. Based on this, we are then able to formulate and develop methods aiming specifically at overcoming these weaknesses in the second part of the thesis. These approaches range from simple methods such as parameter sharing and data augmentation to more sophisticated methods such as designing new objective functions or pseudo data synthesis and semi-supervised learning. Finally, we unify these developed methods into a single framework that outperforms state-of-the-art models by a significant margin on common benchmarking datasets. The code to reproduce our experiments is released at https://github.com/hnt4499/DPR.
author2 Joty Shafiq Rayhan
author_facet Joty Shafiq Rayhan
Hoang, Nghia Tuyen
format Final Year Project
author Hoang, Nghia Tuyen
author_sort Hoang, Nghia Tuyen
title Open domain question answering system
title_short Open domain question answering system
title_full Open domain question answering system
title_fullStr Open domain question answering system
title_full_unstemmed Open domain question answering system
title_sort open domain question answering system
publisher Nanyang Technological University
publishDate 2022
url https://hdl.handle.net/10356/156955
_version_ 1759857127808040960