【文獻分享】FinBERT: Financial Sentiment Analysis with Pre-trained Language Models

Date

August 27, 2023

類型

文獻分享

程度

高階

領域

資料科學機器學習/深度學習資料分析

單元要先準備事項

建議看過 Unit 21. 機器學習超入門方法 - 一元線性回歸與評估指標以瞭解評估指標
建議看過 Unit 29. 邊寫邊理解 Logistic Regression （LR Model Part. 1）＆ Unit 30. Logistic Regression + 技術指標做預測

選這篇的原因

認識消息面或許可以從什麼方面切入
延伸學習下 NLP 的子任務（畢竟太夯了(x)）
認識極性分析、怎麼做把一個通用模型做 domain 訓練

文獻資源

出處（這篇直接看的到）：

arxiv.org

影片內容

Dataset

Reuters Corpora @ NIST

In 2000, Reuters Ltd made available a large collection of Reuters News stories for use in research and development of natural language processing, information retrieval, and machine learning systems. This corpus, known as "Reuters Corpus, Volume 1" or RCV1, is significantly larger than the older, well-known Reuters-21578 collection heavily used in the text classification community.

trec.nist.gov

(PDF) FinancialPhraseBank-v1.0

PDF | On Jul 23, 2013, Pekka Malo and others published FinancialPhraseBank-v1.0 | Find, read and cite all the research you need on ResearchGate

www.researchgate.net

FiQA - 2018

Open Challenge - WWW 2018 LYON, FRANCE (23 - 27 April 2018) Summary The growing maturity of Natural Language Processing (NLP) techniques and resources is drastically changing the landscape of many application domains which are dependent on the analysis of unstructured data at scale. The financial

sites.google.com

Github & Hugging Face

Embed GitHub

ProsusAI/finbert · Hugging Face

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

huggingface.co