Financial Text Classification With Deep Learning Using FinBERT
This article is a comprehensive overview of the application of the FinBERT pre-trained model on financial text data classification tasks
5 min readJan 6, 2022
Introduction
Managing information from the financial market is not an easy task, especially in today’s fast and complex ecosystem. Financial actors might want to predict the change in stock price or identify companies that better fit their investment requirements; those are some of the challenges amongst a plethora of challenges that financial actors have been trying to tackle.
However, using previous financial text information, actors can increase their chance of finding the best company to invest in or predict the right change in stock price. This article will try to use deep learning in order to classify is given financial news headlines using FinBERT.
About the Dataset
- This is the Financial Phrase Bank data, downloadable from Kaggle. contains the sentiments for financial news headlines from the perspective of a retail investor. Further details about the dataset can be found in: “Good debt or bad debt: Detecting semantic orientations in economic texts.”. There are two main columns:
