Turkish keyphrase extraction using KEA

Keyphrases provide semantic metadata to summarize and characterize documents. Unfortunately, there are many digital documents especially on the Internet that do not have a list of assigned keyphrases. Assigning keyphrases to these documents manually is a tedious process and requires knowledge of the subject. Automatic keyphrase extraction solves this problem. In this paper, we present implementation of keyphrase extraction algorithm (KEA) for Turkish as well as extending it with new features to improve its performance.