The Fifth Elephant 2017

On data engineering and application of ML in diverse domains

How to read a user's mind? Designing algorithms for contextual recommendations

Submitted by Bharath Mohan (@bharathmohan) on Monday, 22 May 2017


Technical level



Crisp talk for data engineering track



Vote on this proposal

Login to vote

Total votes:  +5


The human mind is going through thousands of thoughts everyday. A perfect recommender system needs to know what is going on and suggest something useful - at all times, without being perceived as intrusive or noisy. After slicing every possible sensor within the reach of a digital system - from the GPS, Accelerometer, Time of day, Temperature, Browsing History, TV Viewing, Sound, a “perfect recommender system” should learn which ones to give more importance to, predict the state of mind - in order to make the most effective recommendations. This talk takes an exemplary approach to derive some heuristics that can drive these algorithms.


Speaker bio

Bharath Mohan is CEO of Sensara, a company that provides Smarter Live TV User Experience & Monetization. Sensara offers Smart Remotes, TV Search & Personalization, Smart TV Interfaces and TV Deep Data Mining. Bharath earlier worked for Google, and was part of the launch team in Google Finance, and also a technical lead for Google News. Bharath has a PhD in Computer Science from the Indian Institute of Science.



  • 1
    Abhishek Balaji (@booleanbalaji) Reviewer a year ago

    Hi Bharath,

    Please upload a two-min preview video explaining what the talk is about and what is the key takeaway for participants. We need this information by 29 May to evaluate your proposal.

Login with Twitter or Google to leave a comment