Distributed word embeddings have shown superior performances in numerous Natural Language Processing (NLP) tasks.
However, their performances vary significantly across different tasks, implying that the word embeddings learnt by those methods capture complementary aspects of lexical semantics.
Therefore, we believe that it is important to combine the existing word embeddings to produce more accurate and complete meta-embeddings of words.
For this purpose, we propose an unsupervised locally linear meta-embedding learning method that takes pre-trained word embeddings as the input, and produces more accurate meta embeddings.
Unlike previously proposed meta-embedding learning methods that learn a global projection over all words in a vocabulary, our proposed method is sensitive to the differences in local neighbourhoods of the individual source word embeddings.
Moreover, we show that vector concatenation, a previously proposed highly competitive baseline approach for integrating word embeddings, can be derived as a special case of the proposed method.
Experimental results on semantic similarity, word analogy, relation classification, and short-text classification tasks show that our meta-embeddings to significantly outperform prior methods in several benchmark datasets, establishing a new state-of-the-art for meta-embeddings.
Danushka Bollegala is an associate professor in natural language processing at the Department of Computer Science, University of Liverpool.
He obtained his PhD in 2009 from the University of Tokyo. His research interests are natural language processing, machine learning, and data mining.
He has published over 80 peer reviewed papers in top international venues related to these fields such as ACL, IJCAI, WWW, EMNLP and AAAI.
He has published several books on Machine Learning and Web Mining.
He has received several best paper awards at conferences and IEEE Young Author Award for his research excellence.
|Date||June 13, 2017 (Tue) 10:30 - 12:00|