Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Shayona@SMM4H23: COVID-19 Self diagnosis classification using BERT and LightGBM models

Rushi Chavda, Darshan Makwana, Vraj Patel, Anupam Shukla

TL;DR

The Transformer model (BERT) in combination with the LightGBM model for both tasks has leveraged the Transformer model (BERT) in combination with the LightGBM model for both tasks.

Abstract

This paper describes approaches and results for shared Task 1 and 4 of SMMH4-23 by Team Shayona. Shared Task-1 was binary classification of english tweets self-reporting a COVID-19 diagnosis, and Shared Task-4 was Binary classification of English Reddit posts self-reporting a social anxiety disorder diagnosis. Our team has achieved the highest f1-score 0.94 in Task-1 among all participants. We have leveraged the Transformer model (BERT) in combination with the LightGBM model for both tasks.

Shayona@SMM4H23: COVID-19 Self diagnosis classification using BERT and LightGBM models

TL;DR

The Transformer model (BERT) in combination with the LightGBM model for both tasks has leveraged the Transformer model (BERT) in combination with the LightGBM model for both tasks.

Abstract

This paper describes approaches and results for shared Task 1 and 4 of SMMH4-23 by Team Shayona. Shared Task-1 was binary classification of english tweets self-reporting a COVID-19 diagnosis, and Shared Task-4 was Binary classification of English Reddit posts self-reporting a social anxiety disorder diagnosis. Our team has achieved the highest f1-score 0.94 in Task-1 among all participants. We have leveraged the Transformer model (BERT) in combination with the LightGBM model for both tasks.
Paper Structure (8 sections, 1 figure, 1 table)

This paper contains 8 sections, 1 figure, 1 table.

Table of Contents

  1. Abstract
  2. Introduction
  3. Data
  4. System
  5. Results
  6. Discussion
  7. Conclusions
  8. References

Figures (1)

  • Figure 1: Figure 1: An overview of the proposed method