Knowledge and data integration for deep learning under small data
This research addresses limited training data in deep learning, where data volume, quality, and diversity significantly influence model performance. The availability of diverse and abundant data is crucial for effective training models. However, in many real-world scenarios, obtaining such varied da...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/176309 |
| _version_ | 1811680686127448064 |
|---|---|
| author | Teo, Hazel Kai Xin |
| author2 | Mao Kezhi |
| author_facet | Mao Kezhi Teo, Hazel Kai Xin |
| author_sort | Teo, Hazel Kai Xin |
| collection | NTU |
| description | This research addresses limited training data in deep learning, where data volume, quality, and diversity significantly influence model performance. The availability of diverse and abundant data is crucial for effective training models. However, in many real-world scenarios, obtaining such varied data can be challenging, potentially leading to biased models, particularly affecting minority classes. Recent literature and research by various scholars emphasize data augmentation techniques as a promising solution to mitigate data scarcity and enhance model accuracy without exhaustive labeling efforts.
This study explores the potential of data augmentation, particularly text augmentation, in alleviating the dependency on extensive training data. The aim is to enhance the effectiveness and accuracy of deep learning models, especially in the context of natural language processing (NLP). We investigate the benefits of employing synonym replacement as a primary text augmentation technique, assessing its ability to generate supplementary data and improve model performance. |
| first_indexed | 2024-10-01T03:28:59Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/176309 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T03:28:59Z |
| publishDate | 2024 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1763092024-05-17T15:45:18Z Knowledge and data integration for deep learning under small data Teo, Hazel Kai Xin Mao Kezhi School of Electrical and Electronic Engineering EKZMao@ntu.edu.sg Engineering Synoynm replacement This research addresses limited training data in deep learning, where data volume, quality, and diversity significantly influence model performance. The availability of diverse and abundant data is crucial for effective training models. However, in many real-world scenarios, obtaining such varied data can be challenging, potentially leading to biased models, particularly affecting minority classes. Recent literature and research by various scholars emphasize data augmentation techniques as a promising solution to mitigate data scarcity and enhance model accuracy without exhaustive labeling efforts. This study explores the potential of data augmentation, particularly text augmentation, in alleviating the dependency on extensive training data. The aim is to enhance the effectiveness and accuracy of deep learning models, especially in the context of natural language processing (NLP). We investigate the benefits of employing synonym replacement as a primary text augmentation technique, assessing its ability to generate supplementary data and improve model performance. Bachelor's degree 2024-05-15T23:28:09Z 2024-05-15T23:28:09Z 2024 Final Year Project (FYP) Teo, H. K. X. (2024). Knowledge and data integration for deep learning under small data. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/176309 https://hdl.handle.net/10356/176309 en A1084-231 application/pdf Nanyang Technological University |
| spellingShingle | Engineering Synoynm replacement Teo, Hazel Kai Xin Knowledge and data integration for deep learning under small data |
| title | Knowledge and data integration for deep learning under small data |
| title_full | Knowledge and data integration for deep learning under small data |
| title_fullStr | Knowledge and data integration for deep learning under small data |
| title_full_unstemmed | Knowledge and data integration for deep learning under small data |
| title_short | Knowledge and data integration for deep learning under small data |
| title_sort | knowledge and data integration for deep learning under small data |
| topic | Engineering Synoynm replacement |
| url | https://hdl.handle.net/10356/176309 |
| work_keys_str_mv | AT teohazelkaixin knowledgeanddataintegrationfordeeplearningundersmalldata |