The Fifth Elephant 2019

The eighth edition of India's best data conference

Participate

Extract calendar events from free-form text (chats/emails) to automate scheduling

Submitted by Adarsh Dattatri (@adarsh-dattatri) on Monday, 15 April 2019

Abstract

Sales teams activities include scheduling meetings with prospects for product demos, resolving queries and doubts about the product, initial setup. At Freshworks, we use NLU to automatically detecting meeting intent within emails/chat and generate a calendar event. This talk is about the pipeline and tools used to engineer this system.

Outline

Input Source:
Email Chat Source Type:
Raw Text Domain Specific Text Text processing pipeline:
Intent classification Entity recognition English text to calendar entry generation

Speaker bio

I am a Lead Data Scientist at Freshworks. My background is in building ML systems from prototype to production. My previous experience was building ML systems for Fintech companies and a bulge-bracket investment bank.

Slides

https://www.slideshare.net/secret/e8NqWPm2mad9jJ

Comments

  • Zainab Bawa (@zainabbawa) Reviewer a month ago

    The slides are too thin to assess:

    1. What is the problem you are trying to solve?
    2. How generalizable is this problem beyond Freshworks? Participants who come to The Fifth Elephant do not want to hear about a particular company’s problems and its solution. They want to either hear a case study and/or war story, and what patterns/anti-patterns can be gleaned from case studies about building ML pipelines, designing data engineering platforms, etc.
    3. It is also unclear from the slides and the abstract as to why NLU was used as a solution to the problem. Did you consider any other approach? Why did you narrow down on this?

Login with Twitter or Google to leave a comment