You are viewing content from a past/completed QCon

Presentation: Uber's Spatio-Temporal Data

Track: Sequential Data: Natural Language, Time Series, and Sound

Location: Cyril Magnin I

Duration: 11:00am - 11:40am

Day of week: Wednesday

Share this on:

Abstract

Uber’s Marketplace is the algorithmic brains and decision engine behind our ride-sharing services. Marketplace Forecasting builds and deploys spatio-temporal models and forecasts to enable hyperlocal decision making. To model the physical world requires us to reimagine how we look at the basic problem of forecasting.

 

We will discuss the challenges of modeling the influence of external signals, such as global news events and holidays in our Marketplace. In the majority of cases, there is limited historical data, and in cases where cities have just launched, there is no data at all. We will briefly cover how different techniques ranging from linear to deep learning models, generalized embeddings and cutting edge AI to help us forecast the future states of the Marketplace and even predict the onset of extreme events before they occur!

Speaker: Chintan Turakhia

Engineering Manager @Uber

Chintan Turakhia leads engineering for Uber’s Marketplace Forecasting, Global Intelligence, and Marketplace Experimentation teams. They focus on building production level systems and algorithms for real-time spatio-temporal forecasting, capturing open market signals that impact market dynamics, and designing continuous market-level experimentation systems for Uber. He has over 14 years of industry experience in large-scale system design and mass-event planning (think SuperBowl or Olympics) and is always looking for ways to blend data, engineering, and science to solve physical world problems.

Find Chintan Turakhia at

2019 Tracks

  • ML in Action

    Applied track demonstrating how to train, score, and handle common machine learning use cases, including heavy concentration in the space of security and fraud

  • Deep Learning in Practice

    Deep learning use cases around edge computing, deep learning for search, explainability, fairness, and perception.

  • Handling Sequential Data Like an Expert / ML Applied to Operations

    Discussing the complexities of time (half track) and Machine Learning in the data center (half track). Exploring topics from hyper loglog to predictive auto-scaling in each of two half-day tracks.

    Half-day tracks