Comparing different ML approaches on the NY taxi dataset

Mindified Christmas Edition -

The machine learning field has an extensive set of algorithms and tools that you can apply on a specific problem. In order to see how some different machine learning algorithms and feature selection approaches compare when applied to a regression problem with a fairly large data set, we took on the Kaggle New York Taxi data set: New York City Taxi Trip Duration | Kaggle .

We will share some of our results using classical machine learning algorithms as well as neural networks. Further, we will discuss some of the feature engineering that can be applied to improve results. Finally, as we were setting up a new team with different backgrounds in this project, we will also touch on the team aspect and the skill sets needed to run a machine learning project.

After the presentation there will be snacks, beers and time for discussions.

Giuseppe is a Biomedical Engineer with a PhD at Lund University. His work focused on the application of image analysis to digital pathology. He has now embarked on a new adventure with Mindified

Lars has a MSc in Engineering Physics and has a background from software and product development mainly with focus on IoT applications. He joined Mindified in October.