Week 5: Book Summary Classifier

This week we tried some new methods to acquire better results and continued experimenting with different classification methods.

Because of the systems inability to process our tags as their given format, we converted them into an array format with values designated by us simply by the order in which they appear.

Before the process

After the values are converted

On this weeks experiment we produced a TF-IDF, which calculates the relevance of a word in a given tag in our data.

This time around we decided to drop the stop words in the data, so the TF IDF better calculates the relevant words.

This shows the first five words related to the genre given by our classification.

Members:

Batuhan Mete — Medium

Mehmet Akoguz — Medium

MTUSTA — Medium