Combining Distributional and Morphological Information for Part of Speech Induction

In this paper we discuss algorithms for clustering words into classes from unlabelled text using unsupervised algorithms, based on distributional and morphological information. We show how the use of morphological information can improve the performance on rare words, and that this is robust across a wide range of languages.

• Type: Conference
• Year: 2003
• Where: EACL
• Authors: Alexander Clark
• Keywords: EACL 2003, Morphological Information, Natural Language Processing, Unlabelled Text, Unsupervised Algorithms