Machine Learning in R

It’s been a while that I want to get into machine learning. Given that my “base” in programming is R, it was natural that I would start there.

Looking online for packages, the e1071 seemed to have what I need (mainly, a Support Vector Machine or SVM classifier) and this straightforward tutorial gave me the basic idea. Of course it uses the iris dataset as an example, which is a very common dataset to learn R programming and data manipulation.

Machine learning can be useful in many contexts in which we would like to make certain predictions on new data based on previous observations, otherwise known as regression, or to classify new observations in set categories.

Why to stop there, I thought!? Since I’ve been experimenting with Shiny apps, let’s make a simple interface to predict iris species based on the four parameters in the dataset. You can check out the result here. You could input the different parameters of an iris flower and see which category does it probably fit into.

Take a look and make sure to write me your questions or comments. Also, if you think you need to apply a classifier on your data or you need an online interface to do some data processing, feel free to contact me via this contact form, or on social media below.

UPDATE (26.11.2018): The link to the shiny app has been changed. I will be adding other machine learning parts to the app, like the weight-height regression predictor.