FinBERT-ABSA
所属分类：人工智能/神经网络/深度学习
开发工具：Python
文件大小：0KB
下载次数：0
上传日期：2023-12-01 10:20:01
上 传 者：
sh-1993
说明： 用于分析金融新闻的NLP模型-提取新闻中提到的公司的ESG得分。
(NLP model for the analysis of financial news - Extraction of ESG scores for companies mentioned in the news.)
文件列表:
NLP and the study of financial news: The contribution of artificial intelligence to decision-making.pdf (1050248, 2023-12-01)
News Analysis/ (0, 2023-12-01)
News Analysis/Pré-processing - Model.py (6859, 2023-12-01)
News Analysis/Sentiment_analysis_final_ESG.py (17035, 2023-12-01)
News Analysis/Summary + Translation.py (6795, 2023-12-01)
Training/ (0, 2023-12-01)
Training/NER on Training Dataset.py (3958, 2023-12-01)
Training/Pré-processing - Model - Training.py (6382, 2023-12-01)
Training/training_model.py (15510, 2023-12-01)
# FinBERT-ABSA
NLP model for the analysis of financial news - Extraction of ESG scores for companies mentioned in the news.
# Abstract
This paper presents a project aimed at extracting Environmental, Social, and Governance (ESG) scores for companies mentioned in financial news, providing a decision support tool for fund managers. The approach involves manual annotation of a financial news dataset comprising approximately 2000 training samples. The FinBERT model is employed and fine-tuned for the Aspect-Based Sentiment Analysis (ABSA) task, focusing on isolating scores related to a specific aspect in each sentence, namely, the mentioned company. The methodology showcases the adaptation of pre-trained language models for targeted ESG assessment in the financial domain, contributing to the advancement of sentiment analysis applications in the context of investment decision-making.
近期下载者：
相关文件：
收藏者：