Cities, IoT and Analytics
Software platform to load and enrich data from various Big Data sources
Dataiku is a startup with the ambition to revolutionize the daily work of data scientists with a new product, launched in 2014 : DATA SCIENCE STUDIO (DSS). DSS is a software platform that enables you to load and enrich data from various Big Data sources (Hadoop,SQL,Cassandra…), explore and visualize data in a user-friendly interface and build predictive models.
It aggregates all the steps and tools necessary to get from raw data to production ready applications. The tool is used by several customers in e-retail contexts (e.g. Vente-privee.com ), but is also increasingly used in the context of smart-city / smart traffic.
- Blablacar (car sharing leader in France) would use it to optimize matching algorithm,
- Chronopost (subsidiary of French postal service) to predict last-mile delivery needs.
- Parkeon to build a mobile app which could predict where drivers can park their car when they want to park it.
Predictive modelling for traffic is tough and challenging: it involves large and noisy time series stored in different databases, scalability issues, geographical data manipulations and stacking several predictive layers. We will present our current understanding of what works and what does not.