
158 Text Mining and Visualization: Case Studies Using Open-Source Tools
using Python or any other scripting language, and the output can be represented as a set
of feature vectors for each session, as shown in Figure 7.3.
FIGURE 7.3: A set of feature vectors from the AOL log.
There are of course a great many other features we could extract or derive from this log,
and the process of feature selection is something to which careful thought should be given.
But for now let us consider another key question: What are the most effective techniques
for revealing common patterns in such data? One such approach is to use an unsupervised
learning technique, such ...