Unlock an organization’s knowledge using machine comprehension

abhishek jha (~abhishek88)




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 single-turn as well as multi-turn(conversational) question-answer models. By virtue of the underlying transfer learning layer (using contextualized word embeddings such as BERT, XLNET, ALBERT etc.) off the shelf, question-answer models can easily find exact answers to factoid questions. End-users of an organization’s products often require support on procedures for different tasks.

Outline/Structure of the Talk

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

Slides Link



Knowledge of Machine Learning, Natural Language Processing and Python

Speaker Info:

Abhishek Jha(abhishek.a.jha@ericsson.com) is a Data Scientist working at Global Artificial Intelligence Accelerator (GAIA) at Ericsson India. He has over five years of extensive data science experience building predictive and prescriptive models for the retail and telecom domain. He is a Natural Language Processing (NLP) enthusiast and has worked on building production quality solutions for sentiment analysis in retail. He is a graduate from the prestigious Indian Institute of Technology, Kharagpur.

Atul Singh (Ph.D.)(atul.b.singh@ericsson.com) is a Principal Data Scientist working at Global Artificial Intelligence Accelerator (GAIA) at Ericsson India. He is a data science enthusiast with over sixteen years of software industry work experience in product development, research, and innovation.

He has demonstrated capability of using advanced machine learning techniques to solve complex business problems in the retail, finance and telecom domain.

He has a Ph.D. in Computer Science from Trinity College Ireland and has done his graduation from the Indian Institute of Technology Kanpur. He has eleven granted US patents, eleven pending US patent applications, and over twenty research publications in various international forums. He is also an alumnus of the Business Analytics and Intelligence course from IIM Bangalore. His interests include Natural Language Processing (NLP), geospatial analytics, and reinforcement learning.

Section: Data Science, Machine Learning and AI
Type: Talks
Target Audience: Advanced
Last Updated: