Learning Shallow Syntactic Dependencies from Imbalanced Datasets: A Case Study in Modern Greek and English

The present work aims to create a shallow parser for Modern Greek subject/object detection, using machine learning techniques. The parser relies on limited resources. Experiments with equivalent input and the same learning techniques were conducted for English, as well, proving that the methodology can be adjusted to deal with other languages with only minor modifications. For the first time, the class imbalance problem concerning Modern Greek syntactically annotated data is successfully addressed.