Andrew Nedilko


2023

pdf
TEAM BIAS BUSTERS@LT-EDI-2023: Detecting Signs of Depression with Generative Pretrained Transformers
Andrew Nedilko
Proceedings of the Third Workshop on Language Technology for Equality, Diversity and Inclusion

This paper describes our methodology adopted to participate in the multi-class classification task under the auspices of the Third Workshop on Language Technology for Equality, Diversity, Inclusion (LT-EDI) in the Recent Advances in Natural Language Processing (RANLP) 2023 conference. The overall objective was to employ ML algorithms to detect signs of depression in English social media content, classifying each post into one of three categories: no depression, moderate depression, and severe depression. To accomplish this we utilized generative pretrained transformers (GPTs), leveraging the full-scale OpenAI API. Our strategy incorporated prompt engineering for zero-shot and few-shot learning scenarios with ChatGPT and fine-tuning a GPT-3 model. The latter approach yielded the best results which allowed us to outperform our benchmark XGBoost classifier based on character-level features on the dev set and score a macro F1 score of 0.419 on the final blind test set.

pdf
Team Bias Busters at WASSA 2023 Empathy, Emotion and Personality Shared Task: Emotion Detection with Generative Pretrained Transformers
Andrew Nedilko | Yi Chu
Proceedings of the 13th Workshop on Computational Approaches to Subjectivity, Sentiment, & Social Media Analysis

This paper describes the approach that we used to take part in the multi-label multi-class emotion classification as Track 3 of the WASSA 2023 Empathy, Emotion and Personality Shared Task at ACL 2023. The overall goal of this track is to build models that can predict 8 classes (7 emotions + neutral) based on short English essays written in response to news article that talked about events perceived as harmful to people. We used OpenAI generative pretrained transformers with full-scale APIs for the emotion prediction task by fine-tuning a GPT-3 model and doing prompt engineering for zero-shot / few-shot learning with ChatGPT and GPT-4 models based on multiple experiments on the dev set. The most efficient method was fine-tuning a GPT-3 model which allowed us to beat our baseline character-based XGBoost Classifier and rank 2nd among all other participants by achieving a macro F1 score of 0.65 and a micro F1 score of 0.7 on the final blind test set.

pdf
Generative Pretrained Transformers for Emotion Detection in a Code-Switching Setting
Andrew Nedilko
Proceedings of the 13th Workshop on Computational Approaches to Subjectivity, Sentiment, & Social Media Analysis

This paper describes the approach that we utilized to participate in the shared task for multi-label and multi-class emotion classification organized as part of WASSA 2023 at ACL 2023. The objective was to build mod- els that can predict 11 classes of emotions, or the lack thereof (neutral class) based on code- mixed Roman Urdu and English SMS text messages. We participated in Track 2 of this task - multi-class emotion classification (MCEC). We used generative pretrained transformers, namely ChatGPT because it has a commercially available full-scale API, for the emotion detec- tion task by leveraging the prompt engineer- ing and zero-shot / few-shot learning method- ologies based on multiple experiments on the dev set. Although this was the first time we used a GPT model for the purpose, this ap- proach allowed us to beat our own baseline character-based XGBClassifier, as well as the baseline model trained by the organizers (bert- base-multilingual-cased). We ranked 4th and achieved the macro F1 score of 0.7038 and the accuracy of 0.7313 on the blind test set.
Search
Co-authors
Venues