Predicting ad click-through with a decision tree

After several examples, it is now time to predict ad click-through using the decision tree algorithm you have just thoroughly learned about and practiced with. We will use the dataset from a Kaggle machine learning competition, Click-Through Rate Prediction (https://www.kaggle.com/c/avazu-ctr-prediction). The dataset can be downloaded from https://www.kaggle.com/c/avazu-ctr-prediction/data.

Only the train.gz file contains labeled samples, so we only need to download this and unzip it (it will take a while). In this chapter, we will focus on only the first 300,000 samples from the train.csv file unzipped from train.gz.

The fields in the raw file are as follows:

Figure 3.12: Description and example values of the dataset

We take a glance at the head of the file by running the following command:

head train | sed 's/,,/, ,/g;s/,,/, ,/g' | column -s, -t

Rather than a simple head train, the output is...