Rich Morphological Modeling for Multi-lingual HLT Applications - 2016

Explore advanced techniques for improving human language technology applications across diverse languages through morphological modeling in this comprehensive talk by Dr. Christo Kirov. Delve into projects including the creation of a large-scale morphological paradigm database from Wiktionary, consensus-based morphology transfer via cross-lingual projection, and innovative approaches to lemmatization and morphological analysis using recurrent neural network architectures. Gain insights into the UniMorph project at CLSP, supported by DARPA LORELEI, and its contributions to advancing multi-lingual HLT applications. Learn from Dr. Kirov's expertise in combining machine learning with traditional linguistics to represent and learn morphological systems across world languages, and discover how this knowledge can be leveraged in Machine Translation, Information Extraction, and other HLT tasks.