Sava Kalbachou

Paul Anderson photo

AI Research Engineer
Lucidworks

Sava is AI Research Engineer at Lucidworks, whose main focus is researching, developing and integrating cutting-edge NLP algorithms. As a part of AI-Lab team, he has been working closely on question answering systems, chatbots and neural search engines as well as on enriching search results with additional information like sentiment, topics, extracted named entities etc. He’s also actively participating in variety of Machine Learning competitions on Kaggle and other platforms, where he was able to achieve winning places by applying state-of-the-art Deep Learning techniques.

Sava Kalbachou is speaking at the following session/s

Question Answering and Virtual Assistants with Deep Learning

Thursday | 2:20PM - 3:00PM | Columbia 3/4

Information Retrieval (IR) based question answering systems have many applications in the real world. Recent advances in Deep Learning (DL) give us a huge possibility to improve IR apps and engines, which can be used as a backbone for Virtual Assistants and Chatbots. In this talk, we will demo and go through details of our Question Answering systems and show how it can be integrated with a Chatbot. We will also explain how we addressed some of the common implementation issues and limitations, such as lack of training data, query time performance of dense vectors search, answer span selection, etc.

Attendee Takeaway
Learn how and why Deep Learning powered Question Answering systems might be built and integrated to existing solutions.

Intended Audience
Engineers, data scientists and machine learning enthusiasts who wants to enrich their products with deep learning powered question answering systems and chatbots. Prerequisite knowledge isn't necessary although might be useful to understand some concepts in deep.

Level:
All Levels