Automatic Classification of WordNet Morphosemantic Relations

This paper presents work in progress on a machine learning method for classification of morphosemantic relations between verb and noun synsets. The training data comprises 5,584 verb–noun synset pairs from the Bulgarian WordNet, where the morphosemantic relations were automatically transferred from the Princeton WordNet morphosemantic database. The machine learning is based on 4 features (verb and noun endings and their respective semantic primes). We apply a supervised machine learning method based on a decision tree algorithm implemented in Python and NLTK. The overall performance of the method reached F1-score of 0.936. Our future work focuses on automatic identification of morphosemantically related synsets and on improving the classification.