FNSPID: A Comprehensive Financial News Dataset in Time Series
By Zihan Dong, Xinyu Fan, Zhiyuan Peng
Rating
1149
Battle Count: 318
Relevance
9/10
Highly relevant for developing and testing quantitative trading strategies based on news sentiment and stock price data
Implementation Complexity
6/10
Moderate complexity due to large dataset and multiple data types, but well-documented methodology
Reproducibility
4/5
Detailed data collection and processing methods provided, code and documentation available on GitHub
About this paper
Methodology: Data Collection and Processing. Problem types: Time Series Forecasting, Sentiment Analysis, Regression.
The interactive Everscope explorer (charts, battles, favorites) loads below.