What we're about

Recent advances in Artificial Intelligence and Machine Learning have noticeable impact on the way the ongoing stream of world news is being created, processed and utilized. While computational approach to text has already been applied for some time to news – including document classification, named entity recognition, event detection, sentiment analysis and other tasks – recent advances (that often go under the title of “deep learning”) open new possibilities and horizons. We would like to dedicate these meetings to the exploration of those new directions: Can we significantly improve on the above mentioned tasks? What additional tasks are made possible? What are financial institutes expecting to see? How can we help them to digest the ever-growing quantities of information? Can we define and derive further news-based actionable insights?

We would like to introduce the above challenges and the computational techniques that underlie them to researchers, engineers and developers that are interested in this domain, as well as to the broader community of interested people, including academics of less thorough computational background (from disciplines such as Economics, Political Sciences, Communication and Media…), managers and visionaries from the high-tech industry and so on.

This meetup group is jointly sponsored by the Data Science Institute at Bar-Ilan University (http://dsi.biu.ac.il/) and Refinitiv (http://refinitiv.com/) (formerly Thomson-Reuters).

Past events (4)

Deep Entities and Convolutional Sentiments

Thomson-Reuters offices

Latest Research by the Bar-Ilan NLP Lab

Gonda Building (901)

Embedding-based Term Sets & Dynamic Ontologies

Thomson-Reuters offices

DIY Classification / Dear Data Scientist, Label Data or Die

Thomson-Reuters offices

Photos (9)