Going out for dinner in Mumbai during an extended stay, or planning for a long road-trip across the wild west of Rajasthan, the first thing one looks at is Maps, that informs the relative distance, estimated time and congestion areas of different routes for the drive. Zendrive built state-of-the-art technologies on its huge cache of driving data from smartphones and OBD, to add a significant dimension to the route mapping of Google, that is safety risk of the route. Essentially the technology is built on millions of drivers zipping through the route or segments thereof.
Automobile Insurance expands in UBI- where it has been established that tracking a driver’s behavior behind the wheels (like Hard Brake, Speeding etc) can predict significant differences between their chances of collisions. Looking at the same event data from the road perspective, aggregating the relative event density on road stretches also predict the relative chances of collision on that segment. We have used map matching using GIS techniques, parametric density estimation and rare event modeling using quasi-Poisson GLM to analyze our data, build the models and finally implement the scoring system across the GIS route maps.
Key learnings :
Relation between dangerous driving events and collisions
Route risk as an aggregate of all the drivers (or sample thereof) and their driving risk.
The route you take for commute may determine your auto insurance.
Usage Based Insurance : relation between collision rates and dangerous driving.
Driving events : aggressive acceleration, hard brake, speeding, phone use, aggressive turns
Poisson GLM modeling to predict collision rates using driving data
Events on a road segment : map-matching using GIS techniques to split trips along road stretches, and aggregate such events along the spatio-temporal dimension across all drivers.
Route risk of the road segment and any route comprising such segments.
Driving risk along such routes and corresponding collision risks using transfer of the GLM model.
Assignment of risks to drivers on their daily route of commute, to be used in UBI.
Knowledge Talk Room
How to popularize cutting edge research and use of artificial intelligence in India?
India has one of the largest student numbers in STEM field. It also has vast number of engineers in corporate and government sector. Despite that India is known for providing lower end services and cutting edge international products are a rarity. In addition, despite tremendous growth in the start-up sphere, India is still nowhere in the list of top innovative countries. How can student and corporate training change that? Also what kind of products and policies should be pushed on a priority, to make the needed difference? In my talk, I would use my experience of citizen science in USA and India to chart out a territory. I would emphasize needed skills for corporate re-skilling, so India is prepared for the fourth Industrial revolution. Also, I would discuss a couple of examples from my dozen or so partnerships with start-ups involved in AI sphere. I would emphasize how development of few key AI technologies can promote efficient, ethical, and egalitarian use of social media, healthcare, and education to transform India.
Amazon Alexa, the cloud based voice service that powers Amazon Echo, provides access to thousands of skills that enable customers to voice control their world - whether it’s listening to music, controlling smart home devices, listening to the news or even ordering a pizza. This talk will serve as an introduction to the Amazon Alexa and the Alexa Skills Kit while illustrating why Voice is the next major disruption in computing. The talk includes a hands-on demo of the entire skill building process. You will also learn how the Alexa cloud service works as well as best practices in voice design.