yingpengma/FinTrust
Code for ACL-2023 paper "Measuring Consistency in Text-based Financial Forecasting Models"
This project helps financial analysts and researchers evaluate the reliability of AI models that predict financial outcomes from text. It takes earnings call transcripts and generates specialized test sets. The output allows you to assess how consistently a forecasting model performs when faced with subtle variations in text data. It's for anyone building or using AI models to make financial predictions from text.
No commits in the last 6 months.
Use this if you need to rigorously test the consistency and robustness of text-based financial forecasting models, especially those analyzing earnings call transcripts.
Not ideal if you're looking for a tool to build a financial forecasting model from scratch, or if your primary data source isn't earnings call transcripts.
Stars
16
Forks
4
Language
Python
License
—
Category
Last pushed
Jan 22, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/yingpengma/FinTrust"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
Shubxam/Nifty-500-Live-Sentiment-Analysis
Live Sentiment Analysis dashboard of NIFTY 500 universe of stocks using plotly and streamlit
lefterisloukas/edgar-crawler
The only open-source toolkit that can download SEC EDGAR financial reports and extract textual...
yya518/FinBERT
A Pretrained BERT Model for Financial Communications. https://arxiv.org/abs/2006.08097
shirosaidev/stocksight
Stock market analyzer and predictor using Elasticsearch, Twitter, News headlines and Python...
databricks-industry-solutions/esg-scoring
In this solution, we offer a novel approach to sustainable finance by combining NLP techniques...