Data Science and the Internet of Things - Deep Analytics on Traffic Data

Data Science
Location: King's Suite - Balmoral Level: Advanced
Average rating: ***..
(3.75, 12 ratings)

Automotive navigation systems are highly dependent on information about travel speeds along different routes. For this purpose, several methods exist which report traffic jams like TMC or using cell phone data. Instead of just reporting current traffic conditions, it would be great to know in advance if street segments will become crowded. This information is vital to plan routes to far-away destinations.

Predicting travel speeds is one possible application that is based on deep analysis of traffic data. Having a more detailed understanding of past traffic flows enables more precise predictions of future traffic conditions. However, dealing with traffic data poses a number of challenges; a number of bias effects have to be identified and filtered out accordingly.

Alexander Kagoshima presents various ways to visualize traffic flows and methods for deep analysis of these datasets. He will focus on the process of developing these algorithms and shows different interesting results from their application. It is a nice example how different machine learning methods can be combined to derive deep insights into a problem.

Interestingly enough, one applied method comes to identify car groups with different velocities comes from bioinformatics, which is a great example of how machine learning methods can be applied successfully to entirely new domains.

Alexander Kagoshima


Alexander Kagoshima received a M.S. in Economics and Engineering from TU Berlin in 2012. In graduate school his focus was on machine learning and statistics. In his bachelor thesis, he applied Gaussian Processes to currency exchange rates for prediction. For his master thesis, Alex developed and evaluated a change-point detection algorithm that operates on wind data, to enable new kinds of intelligent wind-turbine control systems. He gained practical experience during graduate school as a working student: at Volkswagen, he applied machine learning methods to data of a test fleet of fuel-cell cars. Since December 2012, Alex works at Greenplum (now Pivotal) as the first Data Scientist of the EMEA team. In his spare time, he tries to find new ways to analyze soccer games through statistics.

Photo of Noelle Sio

Noelle Sio


Noelle Sio has a background in mathematics, statistics, and data mining with an emphasis on digital media. She is currently a Principal Data Scientist at Pivotal (formerly Greenplum). Her work has mainly focused on helping companies across multiple industry verticals extend their analytical capabilities by exploring and modeling digital data, specifically to create an underlying analytics framework to optimize a consumer’s experience. Her previous projects include enabling multiple media companies to optimize their online campaigns via dynamic targeting and pricing, predicting online clicks and conversions, gaming/sports analytics, and developing traffic prediction methods for route optimization.

Previously, she worked as a researcher at eHarmony and Fox Interactive Media, where she leveraged massive datasets up to the petabyte level for marketing optimization, fraud detection, and ad monetization products. During her time at Greenplum, she was among the first contributors to MADlib , an in-database machine learning library. Noelle holds an A.B. From Washington University in St. Louis in Applied Mathematics and Physical Anthropology and a M.S. in Applied Mathematics from Cal Poly Pomona.


Sponsorship Opportunities

For exhibition and sponsorship opportunities, contact Susan Stewart at

Media Partner Opportunities

For information on trade opportunities with O'Reilly conferences email mediapartners

Press & Media

For media-related inquiries, contact Maureen Jennings at

Contact Us

View a complete list of Strata contacts