期刊
ENTERPRISE INFORMATION SYSTEMS
卷 17, 期 4, 页码 -出版社
TAYLOR & FRANCIS LTD
DOI: 10.1080/17517575.2021.1959652
关键词
Supply chain; risk assessment; social media; fuzzy logic; natural language processing
In this study, a novel methodology is developed to identify supply chain disruption events using Twitter feeds in real time. The approach incorporates advances in Natural Language Processing (NLP) and machine learning, and demonstrates improved performance over existing models.
In this study, we develop a novel methodology to identify supply chain disruption events using Twitter feeds in real time. Underpinned by advances in Natural Language Processing (NLP) and machine learning, we propose an approach that includes a state-of-the-art variant of Conditional Random Field (CRF) model for event annotation, location-based clustering of the annotated events, and a fuzzy inference system to evaluate supply chain risk. We validate the new approach through a text corpus derived from a Twitter data stream, which is a popular method in NLP. The results show that the proposed model outperforms the baseline model.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据