I am using machine learning for Manufacturing data, we have a lot of data and lots of features, it’s so huge that it’s even a bit overwelming, can you suggest some data preproocessing and feature engineering technique that i can use to clean up the data and figure out the top feautres (or dimension reduction)?
Thanks a million
The Pandas package has some nice tools for this.
So does the R platform, with the advantage of tight integration to lots of visualization capabilities. For exploratory data analysis (EDA), it’s quite useful.
This article describes a conceptual R EDA workflow that might be helpful…
This one is a survey of EDA packages available in the R universe…
Let us know what you learn!