Task-Oriented Learning of Word Embeddings for Semantic Relation Classification
This work addresses the problem of improving relation classification accuracy for natural language processing applications, representing an incremental advance by enhancing embeddings with task-specific features.
The authors tackled semantic relation classification by developing a novel method to learn word embeddings that incorporate relation-specific information, resulting in significant performance improvements over baseline embeddings and competitive results with state-of-the-art models using additional resources.
We present a novel learning method for word embeddings designed for relation classification. Our word embeddings are trained by predicting words between noun pairs using lexical relation-specific features on a large unlabeled corpus. This allows us to explicitly incorporate relation-specific information into the word embeddings. The learned word embeddings are then used to construct feature vectors for a relation classification model. On a well-established semantic relation classification task, our method significantly outperforms a baseline based on a previously introduced word embedding method, and compares favorably to previous state-of-the-art models that use syntactic information or manually constructed external resources.