Automating the data science cycle is one of the most current trends in machine learning. This method is typically based on three components: feature engineering and selection, model selection, and lastly, fine tuning the model’s parameters.
As we believe in openness at KNIME, our approach to automating machine learning is open, too. In our talk we will walk you through a workflow for automated machine learning consisting of modular building blocks, which can be adapted and exchanged to your needs. We show how this workflow can be customised by integrating a variety of tools such as H2O, Apache Spark, Deep Learning or R and Python scripts or adjusted for more complex data types. Furthermore, to make this usable for everyone, we sprinkle Guided Analytics into the mix.