Predicting Fuel Prices and PDFs - Machine Learning to uncover decisive factors


First Talk:

Fuel pricing is a highly interesting data theme as it is influenced by a plethora of factors. We will be exploring it through the use of state of the art Data Science models and tools!

Planning holidays in France? Taking the car to head there? Look no further, this is THE talk you must attend before leaving. If not, don’t worry, there’s also space for you for what will surely be a highly interesting talk!

What will this talk be about?

Based around advanced statistical modeling and the use of graphical tools, our very own Louis-Philippe will show you how external factors can impact gas prices. Using multiple indicators such as localisation, brand and its surrounding competition, he will highlight the different ties between each and how they can merge to predict and explain fuel prices in France.

Speaker Bio:

Louis-Philippe Kronek ( is an accomplished Data Scientist working at Dataiku ( as VP Data science. He is in charge of running the team of (20 and growing) Data Scientists based in Paris, London and NYC. He holds a PhD in Data analysis and is currently working in our New York office, helping customers to build and deploy predictive applications.

Second Talk:

Unsupervised machine learning to cluster PDF's based on the text content.

With an estimated 80% of data ever produced being unstructured, coupled with an ever-growing demand for more value to be produced from data, it becomes increasingly important to use the tools and techniques around us to better our understanding of these unstructured formats, be it free text, images, sound clips or video.

Using language processing libraries and graphing databases, we will explore the possibilities and use cases for free text analysis, classification and clustering.

Speaker Bio:

Ali Kokaz ( is the current head of analytics projects at Kubrick Group ( and is responsible for driving analytics innovation through real life use cases.
Ali holds a first-class degree in astrophysics where he used statistical models to recreate and analyze early universe environments.

This meet-up is for:

- Data geeks of all ages willing to discover how multiple factors which seemingly have nothing in common, can come together to provide insights and forecasts to predict fuel prices in precise locations.

- Data professionals from all horizons (any role, any sector)

As anyone who has attended our meet-ups before will testify, you can expect excellent talks, discussion, and complimentary pizza and beer for everyone!

/Make sure you give your email for the RSVP, it is required by our venue/