Risk, Emotions and Information Detection in Text and Speech

Name: Risk, Emotions and Information Detection in Text and Speech
Start: 2019-12-06T17:30:00+01:00
End: 2019-12-06T20:30:00+01:00
Location: Swiss Re Insurance Company Ltd

Hosted By

Kornelia P.

Risk, Emotions and Information Detection in Text and Speech

Details

It is time for our yearly NLP Christmas together! We would love to welcome you at the 22nd NLP Zurich event 6 December 2019 at Swiss Re Next! We invited three amazing NLP experts to share their research and industrial experience with the community. Each presentation is 20 minutes followed by a short discussion time. Please RSVP on the meetup site.

Agenda:
17:30 Registration (closes 17:55)
18:00 Welcome
18:10 Jochen L. Leidner (Director of Research, R&D, Refinitiv Labs): Automating the Risk Register – Computer-Assisted Identification of Company Risk
18:35 Pankaj Gupta (Research Scientist, Siemens AG Munich): Neural Representation Learning Beyond Sentence Boundaries for Information Extraction and Document Topic Modeling
19:00 Piero Pierucci (Director Speech Tech, Telepathy Labs): Multimodal Emotion Detection – Foundations and State of the Art
19:40 Networking

Jochen L. Leidner: Automating the Risk Register – Computer-Assisted Identification of Company Risk

In this presentation, Risk Mining is presented, a project that led to the first automatic capability for extracting a risk register from journalistic news. I cover how risks can be extracted from unstructured data sources using a combination of natural language processing and weakly supervised as well as supervised machine learning processes. The approach demonstrates the induction of a weakly-supervised risk taxonomy in combination with dependency tree analysis, in order to perform large-scale company-risk relationship classification.

Dr. Jochen L. Leidner is Director of Research, R&D at Refinitiv Labs London and Royal Academy of Engineering Visiting Professor in Data Analytics at the University of Sheffield. For a number of years, he has also guest lectured on the topic of big data at the University of Zurich. He holds a Master's degree in engineering with a focus on machine learning from the University of Cambridge and a Ph.D. in information extraction from the University of Edinburgh. He is a Fellow of the Royal Geographical Society.

Pankaj Gupta: Neural Representation Learning Beyond Sentence Boundaries for Information Extraction and Document Topic Modeling

Dr. Pankaj Gupta is a research scientist, leading AI-NLP portfolio in the Machine Intelligence research group at Siemens AG Munich, Germany. His research contributions span from information extraction, textual representation learning to document analysis, including entity-relation extraction, neural topic modeling, semantic textual similarity and interpretability of neural networks. Beyond research, he also applies the in-house built state-of-the-art technologies into industrial applications at Siemens. https://sites.google.com/view/gupta-pankaj/

Piero Pierucci (Director Speech Tech, Telepathy Labs): Multimodal Emotion Detection – Foundations and State of the Art

In human perception and understanding, a number of different and complementary cues are adopted according to different modalities.
Various emotional states in communication between humans reflect this
variety of cues across modalities. Recent developments in multi-modal emotion recognition utilize deep-learning techniques to achieve remarkable performances. This talk will introduce the topic, and present some recent findings based on cross-modal fusion techniques over deep learning models for emotion detection from spoken audio and corresponding transcripts.