Negative Event Extraction from News Text

[jazzwang]

• 演講主題 Talk Topic: Negative Event Extraction from News Text
• 講者姓名 Name: 宋名傑
• 單位名稱 Organization & Job Title: 玉山銀行數位金融處 機器學習工程師
• 講者簡介 Self Introduction:

半路出家踏入資料分析/機器學習的領域，近期技能點放在NLP；興趣是爬山、陪小孩放電，希望未來能帶著小孩上山同時放電。

• 講題摘要 Abstract:

With emerging threats and enhancing regulatory requirements of money-laundering in banking industry, countless checking/monitoring affairs are taking place every day in commercial banks. However, in ESUN bank, we use state-of-the-art deep learning models to mitigate some tedious daily routine. In this session, for example, we will introduce how we apply pre-trained model in NLP(Natural Language Processing) to classify negative news which depicts financial crimes/criminals from various sources of news. Moreover, we further extract the suspicious people with their accused crime in the news text to effectively alleviate human reading/operation times.