Using Distributional Thesaurus Embedding for Co-hyponymy Detection
This work addresses a specific problem in natural language processing for researchers and practitioners, offering an incremental improvement in lexical relation detection.
The paper tackled the challenge of discriminating lexical relations among distributionally similar words by using network embedding of a distributional thesaurus to detect co-hyponymy relations, showing that this approach outperforms state-of-the-art models by huge margins in binary classification tasks.
Discriminating lexical relations among distributionally similar words has always been a challenge for natural language processing (NLP) community. In this paper, we investigate whether the network embedding of distributional thesaurus can be effectively utilized to detect co-hyponymy relations. By extensive experiments over three benchmark datasets, we show that the vector representation obtained by applying node2vec on distributional thesaurus outperforms the state-of-the-art models for binary classification of co-hyponymy vs. hypernymy, as well as co-hyponymy vs. meronymy, by huge margins.