YangLinyi / FinNLP-Progress

NLP progress in Fintech. A repository to track the progress in Natural Language Processing (NLP) related to the domain of Finance, including the datasets, papers, and current state-of-the-art results for the most popular tasks.

Tracking Progress in FinNLP

News

Updates for ACL 2022:

• FiNER: Financial Numeric Entity Recognition for XBRL Tagging
• Guided Attention Multimodal Multitask Financial Forecasting with Inter-Company Relationships and Global and Local News
• Incorporating Stock Market Signals for Twitter Stance Detection
• Buy Tesla, Sell Ford: Assessing Implicit Stock Market Preference in Pre-trained Language Models

FinNLP@EMNLP2021:

• FinQA: A Dataset of Numerical Reasoning over Financial Data

FinNLP@ACL2021:

• Exploring the Efficacy of Automatically Generated Counterfactuals for Sentiment Analysis
• Multimodal Multi-Speaker Merger & Acquisition Financial Modeling: A New Task, Dataset, and Neural Baselines
• TAT-QA: A Question Answering Benchmark on a Hybrid of Tabular and Textual Content in Finance

FinNLP@AAAI2021:

• Coupling Macro-Sector-Micro Financial Indicators for Learning Stock Representations with Less Uncertainty
• Modeling the Momentum Spillover Effect for Stock Prediction via Attribute-Driven Graph Attention Networks
• Stock Selection via Spatiotemporal Hypergraph Attention Network: A Learning to Rank Approach
• DeepTrader: A Deep Reinforcement Learning Approach for Risk-Return Balanced Portfolio Management with Market Conditions Embedding
• Deep Portfolio Optimization via Distributional Prediction of Residual Factors
• Commission Fee Is Not Enough: A Hierarchical Reinforced Framework for Portfolio Management

Table of contents

• Financial Index Forecasting
• Financial Event Prediction
• Financial Risk Analysis
• Financial Text Mining
• Fraud Detection
• Blockchain

Open-source Datasets

• Multi-modal Earnings Conference Call Datasets
• Pre-trained Word Embedding on Financial Communication Text
• Reuters TRC2 dataset
• Financial Phrase Bank

Open-source Datasets (In Chinese)

数据名称	数据字段	样本量	总量	下载链接
企业工商信息	名称,公司名称,公司介绍,工商,地址,工商注册id,成立时间,法人代表,注册资金,统一信用代码,网址	1万	50万 - (上市及中小型企业)	下载
金融讯息新闻	title-新闻标题,content-新闻内容,pub_ts-发稿日期	2万	210万	下载
专栏资讯	title-新闻标题,content-新闻内容,pub_ts-发稿日期	1万	58万	下载
投资机构信息	机构名称,介绍,行业,规模,轮次	1K	3万	下载
投资事件	事件资讯,投资方,融资方,融资事件,轮次,金额	2K	7万	下载
36氪新闻	title-新闻标题,content-新闻内容,url-网址	1万	11万	下载

Research Topics

Financial Index Forecasting (Financial News/Social Media/Professional Documents/Earning Conference Call/10K-10Q Report)

Tasks:

1. Classification (Binary/Triple Classification)
2. Regression (MSE): Volatility Prediction; Return Prediction

Financial Documents Analysis (Professional Documents)

Tasks: Correlation (ACL-19: Financial Analysts Rating-Earning Conference Call)

Investor Sentiment Analysis (Social Media/Financial News)

Tasks: Sentiment Analysis (Binary Classification / Five-class Classification)

Financial Event Prediction (Bankrupt/IPO/M&A)

Tasks: Classification (Event Prediction) Sentiment Analysis (Market Sentiment Prediction)