Wed 08 Nov 2023 - Named entity recognition (NER) is a fundamental task in natural language processing, concerned with identifying spans of texts belonging to a set of pre-defined categories (e.g. person, organisation and location names). In the financial and insurance domain, apart from using the generic NER system to extract person, organisation and location names, more often, the task is altered to predict domain-specific categories, such as fiscal year, currency and expense. Those systems are tailored to extract useful information from financial documents like annual reports and insurance claims. Moreover, NER also plays an important role in other crucial financial pipelines, such as search engines, knowledge graph extraction and aspects-based sentiment analysis.
The talk will first give a brief introduction to different NER applications in the financial and insurance domains, including real-world examples to show the impact of NER in those areas. Next, an industry system is used as an illustration to show the differences between the industry and academia NER systems. Finally, the last part of the talk will focus on the challenges and opportunities of the NER task, including topics.
Speaker: Dr Juntao Yu
Chair: Dr Lei Fang
You can watch the recording of this webinar on the IFoA's Virtual Learning Environment (VLE). Watch recording.
Dr Lei Fang is a Lecturer in the School of Mathematical Sciences, Queen Mary, University of London (QMUL). Her research interests include, but are not limited to, Systemic Risk Management and Machine Learning Application in Finance. She enjoys teaching Actuarial Science, Risk Management, and Financial Mathematics.
Lei was awarded the Sir Edward Johnston Prize by the Institute and Faculty of Actuaries (IFoA), and the Award of the Dimitris N. Chorafas Foundation. In addition to being the Deputy Head of the Centre for Mathematical Education in QMUL, Lei is currently the Chair of the Natural Language Processing working party in IFoA.
Dr. Juntao Yu is a lecturer at the Queen Mary University of London. He is a co-investigator of an EPSRC project (Annotating Reference and Coreference In Dialogue Using Conversational Agents in games) and a KTP project working with Open Medical on information extraction in the biomedical domain. Before that, he was a lecturer at the University of Essex.
Before becoming a lecturer, he was a post-doctoral researcher at the Queen Mary University of London, working with Professor Massimo Poesio on his five-year DALI (ERC-2015-AdG) project.
He did his PhD at the University of Birmingham, working with Dr. Bernd Bohnet on out-of-domain dependency parsing. His research interests include Deep Learning for NLP, Information Extraction, Coreference Resolution, Conversational AI, Dependency Parsing, Domain Adaptation, Semi-supervised Learning, and Multi-task Learning.