Keyword and named entity recognition on emergency call texts

• Main Author: Hu, Wanyu
• Other Authors: Chng Eng Siong
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
• Online Access: https://hdl.handle.net/10356/144506
• Institution: Nanyang Technological University

This report summarizes the work that has been done in the Final Year Project on the topic Keyword and Named Entity Recognition on Emergency Call Texts. With the development of Artificial Intelligence, much more attention than ever before has been paid to the idea of AI- Oriented systems that can be used to accomplish tasks. This report utilize one of the methods of Artificial Intelligence (AI), specifically the Named Entity Recognition (NER) technique on Emergency Call Texts. The model used for this project is the Bidirectional Representations from Transformers (BERT) model. 75% of the Emergency Call Texts dataset is obtained online. The remaining 25% is generated using a custom Data Generation Model design using BERT. The dataset is fed into the model and the result is evaluated by confusion matrix and F1 score. The predicted result is visualized using a Web Application design using Python Streamlit package. The named entities will be highlighted based on the result from the model and then presented in the web application.