Sisense NLP Meetup

Join us for an NLP Meetup hosted by Sisense.

Schedule:
18:30 - 19:00 - Gathering and Refreshments
19:00 - 19:45 - IBM Project Debater — Building AI That Can Debate Humans — Roy Bar-Haim
19:45 - 20:00 - Break
20:00 - 20:45 - An Anatomy of a Natural Language Query (NLQ) Interface — Amnon Lotan

Location: Jabotinsky Rd 2, Ramat Gan, 16th Floor, on 17.12.2019

Details:
IBM Project Debater – Building AI That Can Debate Humans

Argumentation and debating are fundamental capabilities of human intelligence. They are essential for a wide range of everyday activities that involve reasoning, decision making or persuasion. A recent milestone in this field is Project Debater, the first demonstration of a live competitive debate between an AI system and a human debate champion. Project Debater is an IBM Research AI’s grand challenge, developed for over six years by a large team of NLP and ML researchers and engineers, and was demonstrated in February 2019 at the IBM Think conference in San Francisco, attracting massive media coverage. This significant research effort has resulted in nearly 40 scientific papers and many datasets.

In this talk I will first give a high-level view of the project and its core technologies. I will also discuss possible applications for these technologies. Specifically, I will present the Speech by Crowd platform, which supports the collection of free-text arguments from large audiences on debatable topics to generate meaningful narratives out of numerous independent contributions.

Project Debater achieves its goal by orchestrating dozens of engines, solving a variety of NLP tasks. In the second part of the talk, I will deep dive into one such task, and present our ACL 2019 paper on Debate Topic Expansion - finding related topics that can enrich our arguments and strengthen our case when debating a given topic.

Bio:
Roy Bar-Haim is a research staff member at IBM Research AI. He also serves as a co-chair of the Natural Language Processing Professional Interest Community (NLP PIC) at IBM Research. Since joining Project Debater in 2013, he has been leading research teams working on stance classification, sentiment analysis and argument mining. He received the IBM Research Division Award (2018) and the IBM Corporate Award (2019) for his contributions to the project. He has published in leading NLP and AI conferences and journals, including ACL, AAAI, EMNLP, COLING, EACL, JAIR and JNLE. He regularly reviews for top NLP and AI conferences, and serves as a member of the TACL elite reviewer team. Roy received his B.Sc and M.Sc degrees from the Technion, and his Ph.D from Bar-Ilan University, all in computer science.

An Anatomy of a Natural Language Query (NLQ) Interface

Sisense has become one of the world’s leading Business Intelligence (BI) platforms, on which thousands of business analysts use a friendly point-and-click GUI to generate insightful reports and charts from otherwise incomprehensible heaps of data. This year, we’ve set our sights on enabling our users to let the mouse go. Users can now describe what they want by typing words (like in a search engine), which will get translated to an SQL type query, and be answered.

The challenge of translating any NLQ within an elaborate BI platform forces us to serve a high variety, velocity, and volume of NLQs — as well as customer-customization and privacy issues, at scale.

In this talk, I would like to share our hard-earned experience. We will first go over the broader NLQ field in academia and industry, assess the leading Deep Learning and Rule-Based approaches, and then discuss our choices, results, and what to consider in your next NLQ system.

Bio:
Amnon Lotan is a data scientist with the Sisense AI team. He has over 7 years of experience in building products in NLP and Cyber. He’s worked in startups, corporations, and as an independent consultant. He graduated from BIU with a second degree in Linguistics and NLP.