Newswire: A Large-Scale Structured Database of a Century of Historical News

By Emily Silcock, Abhishek Arora, Luca D'Amico-Wong, Melissa Dell

Rating

1155
Battle Count: 82

Relevance

3/10
While not directly applicable to quantitative trading, the dataset and methods could be adapted for historical financial news analysis

Implementation Complexity

8/10
Complex pipeline involving multiple deep learning models and data processing steps

Reproducibility

4/5
Detailed methodology provided, models and dataset publicly available on Hugging Face

About this paper

Methodology: Deep Learning Pipeline. Problem types: Natural Language Processing, Computer Vision, Information Retrieval, Entity Disambiguation, Topic Classification, Named Entity Recognition.

The interactive Everscope explorer (charts, battles, favorites) loads below.