Feature Engineering Made Easy

An example of unstructured data – server logs

As an example of unstructured data, we have pulled some sample server logs from a public source and included them in a text document. We can take a glimpse of what this unstructured data looks like, so we can recognize it in the future:

# Import our data manipulation tool, Pandasimport pandas as pd# Create a pandas DataFrame from some unstructured Server Logslogs = pd.read_table('../data/server_logs.txt', header=None, names=['Info'])# header=None, specifies that the first line of data is the first data point, not a column name# names=['Info] is me setting the column name in our DataFrame for easier access

We created a DataFrame in pandas called logs that hold our server logs. To take a look, ...