Dante Contella
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Dante Contella has a total of 2 co-authors. Here are the top co-authors by number of shared publications:
- Charles Mutigwe
College of Business Western New England University Springfield MA USA
1 shared publication - Johnson Kinyua
Pennsylvania State University
1 shared publication
Latest Publications
- Open Access
- 8 September 2025
- Published by SETSCI
- Vol. 24, 39-44
Enhancing Financial Market Sentiment Analysis Using Dataset Augmentation
The SETSCI Conference Proceedings is an Open Access publication created for the storage and indexing of proceedings | We developed a series of financial social media sentiment analysis models starting with PyFin-Sentiment, a recent state-of-the-art domain-specific large language model for the financial domain. We focused on label accuracy and improvements in the quality and quantity of the dataset by collecting more data. We were able to improve model accuracy from 0.70 to 0.93, supporting a well-known maxim that models are just as good as the data they are trained on. This created a consistently accurate model that outperformed other existing financial sentiment models. We also used nuanced approaches such as separating social media data into short and long comments which drastically improved the accuracy of our model.
- 8 September 2025
- Published by SETSCI
- Vol. 24, 39-44
Open AccessEnhancing Financial Market Sentiment Analysis Using Dataset Augmentation
The SETSCI Conference Proceedings is an Open Access publication created for the storage and indexing of proceedings | We developed a series of financial social media sentiment analysis models starting with PyFin-Sentiment, a recent state-of-the-art domain-specific large language model for the financial domain. We focused on label accuracy and improvements in the quality and quantity of the dataset by collecting more data. We were able to improve model accuracy from 0.70 to 0.93, supporting a well-known maxim that models are just as good as the data they are trained on. This created a consistently accurate model that outperformed other existing financial sentiment models. We also used nuanced approaches such as separating social media data into short and long comments which drastically improved the accuracy of our model.