Knowledge about an organization’s products, processes, and services is often buried within terse and bulky knowledge repositories. The ability to bring out the required information efficiently, quickly and reliably to the relevant members of the organization goes a long way in improving efficiency and productivity thereby reducing costs. Keyword-based search solutions return a list of documents ranked by relevance, and the user has to read one or more documents to get an answer to the query.

This session presents an introduction to deep-learning based question-answer models. By virtue of the underlying transfer learning layer (using contextualized word embeddings such as BERT) off the shelf, question-answer models can easily find exact answers to factoid questions without the need for any training. End-users of an organization’s products often require support on procedures for different tasks. The talk presents the latest models in goal-oriented messaging that can be used to query the steps required to execute a task.


Outline/Structure of the Experience Report

  1. Using machine comprehension for intelligent search (4 minutes)
  2. Introduction to Question-Answer models (4 minutes)
  3. Introduction to goal-oriented Question-Answer model (4 minutes)
  4. Demo of an open-source pre-trained Question-Answer model (4 minutes)
  5. Challenges in building a Question-Answer model based intelligent search solution (4 minutes)

Learning Outcome

The hack session will enable understanding of Question-Answer models to build intelligent search solutions for their business requirements. The session will also introduce the audiences to a powerful application of word embeddings driven transfer learning for a real-life problem, and customizing word embedding for their domain.

Target Audience

Executives keen on simplifying the knowledge management and access process in their organisation. Data scientists keen on learning about latest natural language understanding techniques.

Prerequisites for Attendees

Introductory knowledge of Natural Language Processing, and word embeddings.

schedule Submitted 1 year ago

Public Feedback