Named entity recognition for unaccompanied children based on deep learning
Since 2020, the pandemic has not only brought huge losses to airlines, but also caused great inconvenience to passengers. Compared with adults, children's travel is more significantly affected. Among them, unaccompanied children who travel by air is facing greater difficulties and challenges....
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis-Master by Coursework |
| Language: | English |
| Published: |
Nanyang Technological University
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/159276 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Since 2020, the pandemic has not only brought huge losses to airlines, but also caused great inconvenience to passengers. Compared with adults, children's travel is more significantly affected. Among them, unaccompanied children who travel by air is facing greater difficulties and challenges.
This dissertation mainly uses the python crawler framework to extract and obtain an unaccompanied children dataset from the official websites of world-famous airlines. After labeling the corpus with Label-Studio, popular deep learning based models, LSTM/LSTM-CRF, BiLSTM/BiLSTM-CRF and BERT/BERT-CRF are applied to test the strength of those models in named entity recognition on the newly built unaccompanied children dataset. Experimental study has been conducted and comparisons have been made on this dataset. The performance analysis on those models is reported in the dissertation. |
|---|