The Fifth Elephant 2017

On data engineering and application of ML in diverse domains

Credit where Credit is due: Using data science to lend to customers without a credit history

Submitted by Vanitha DSilva (@vanithadsilva) on Tuesday, 11 April 2017

videocam
Preview video

Technical level

Intermediate

Section

Crisp talk for data engineering track

Status

Submitted

Vote on this proposal

Login to vote

Total votes:  +16

Abstract

Traditional loans are based on banking history leaving a large segment of people ineligible. These however, represent a highly untapped segment representing large purchasing potential. How do you deem if someone is trustworthy when you have no information to base your decision on? This session will detail methods of evaluating people and extending loans irrespective through leveraging technology and data in today’s digital world

Outline

1) Potential for lending in developing countries such as India
-Current Scenario -Untapped Potential that exists 2) How do you make an informed decision with no traditional sources of data
-What are the traditional sources of data -What are the alternate sources -How can Big Data and Data Science help? 3) Algorithms to deem credit-worthiness
-Credit-scoring architecture -ML techniques employed for different data sources and types -Feedback loops to come full circle

Speaker bio

Driven by identifying patterns, deriving insights and problem solving, data science and Vanitha is a natural fit. She has been instrumental in Oxigen’s evolution to a data-driven organization backed by a decade of experience in gleaning actionable insight and scalable data science solutions. She and her team enable key business decisions using ML, Big Data technology and sometimes just plain grit. When not working she is analysing random trivia usually over chai. She can be reached at vanitha.dsilva@gmail.com

Analyzing numbers, trends and visualisation is like breathing to Neelu, be it in industry specific domain analytics or identifying insights from polling numbers each election season as a hobby. As a core member of Oxigen’s Data Science Team, Neelu has built a company culture of utilizing data-driven insights for key business decisions. She has more than 7 years of experience in the industry across domains and can be reached at singh.neelu26@gmail.com.

Slides

https://docs.google.com/presentation/d/14yqaaUr4jVcs0P3iD_Bpopuq279wINg966DjdbCAryU/edit?usp=sharing

Preview video

https://youtu.be/0EfSyUAGhLg

Comments

  • 1
    Zainab Bawa (@zainabbawa) Reviewer a year ago

    Will you address privacy concerns involved in such a process? We had a similar talk at The Fifth Elephant 2016 about this, and the question of privacy immediately came up.

    • 1
      Vanitha DSilva (@vanithadsilva) Proposer a year ago

      We will talk about privacy and data that can and cannot be used in the sources of data segment. The focus segment here is rural India, which is an overlooked segment in lending since dearth of data is a great challenge

      • 1
        Zainab Bawa (@zainabbawa) Reviewer a year ago

        Please send the draft slides for this talk, detailing the content you will cover, and upload a two-min preview video about your talk, explaining what this talk is about and what is the takeaway for participants.

        • 1
          Vanitha DSilva (@vanithadsilva) Proposer a year ago

          Hi Zainab, updated the page with both links, please let me know if you face any issues viewing or have any further questions.

Login with Twitter or Google to leave a comment