Entity relation is an essential component of some famous knowledge bases,such as Freebase,Yago and Knowledge Graph,while the hyponymy plays an important role in entity relations that show the relationship between the ...Entity relation is an essential component of some famous knowledge bases,such as Freebase,Yago and Knowledge Graph,while the hyponymy plays an important role in entity relations that show the relationship between the more general terms(hypernyms)and the more specific instances of the terms(hyponyms).In this paper,we present a comprehensive scheme of open-domain Chinese entity hypernym hierarchical construction.Some of the most important unsupervised and heuristic approaches for building hierarchical structure are covered in sufficient detail along with reasonable analyses.We experimentally evaluate the proposed methods and compare them with other baselines.The result shows high precision of our method and the proposed scheme will be further improved with larger scale corpora.展开更多
Hypernym detection and discovery are fundamental tasks in natural language processing.The former task aims to identify all possible hypernyms of a given hyponym term,whereas the latter attempts to determine whether th...Hypernym detection and discovery are fundamental tasks in natural language processing.The former task aims to identify all possible hypernyms of a given hyponym term,whereas the latter attempts to determine whether the given two terms hold a hypernymy relation or not.Existing research on hypernym detection and discovery tasks projects a term into various semantic spaces with single mapping functions.Despite their success,these methods may not be adequate in capturing complex semantic relevance between hyponym/hypernymy pairs in two aspects.First,they may fall short in modeling the hierarchical structure in the hypernymy relations,which may help them learn better term representations.Second,the polysemy phenomenon that hypernyms may express distinct senses is understudied.In this paper,we propose a Multi-Projection Recurrent model(MPR)to simultaneously capture the hierarchical relationships between terms and deal with diverse senses caused by the polysemy phenomenon.Specifically,we build a multi-projection mapping block to deal with the polysemy phenomenon,which learns various word senses by multiple projections.Besides,we adopt a hierarchy-aware recurrent block with the recurrent operation followed by a multi-hop aggregation module to capture the hierarchical structure of hypernym relations.Experiments on 11 benchmark datasets in various task settings illustrate that our multi-projection recurrent model outperforms the baselines.The experimental analysis and case study demonstrate that our multi-projection module and the recurrent structure are effective for hypernym detection and discovery tasks.展开更多
基金supported by ZTE Industry-Academia-Research Cooperation Funds
文摘Entity relation is an essential component of some famous knowledge bases,such as Freebase,Yago and Knowledge Graph,while the hyponymy plays an important role in entity relations that show the relationship between the more general terms(hypernyms)and the more specific instances of the terms(hyponyms).In this paper,we present a comprehensive scheme of open-domain Chinese entity hypernym hierarchical construction.Some of the most important unsupervised and heuristic approaches for building hierarchical structure are covered in sufficient detail along with reasonable analyses.We experimentally evaluate the proposed methods and compare them with other baselines.The result shows high precision of our method and the proposed scheme will be further improved with larger scale corpora.
基金supported by the National Science and Technology Major Project of China(2022ZD0120202)the Natural Natural Science Foundation of China(Grant No.U23B2056).
文摘Hypernym detection and discovery are fundamental tasks in natural language processing.The former task aims to identify all possible hypernyms of a given hyponym term,whereas the latter attempts to determine whether the given two terms hold a hypernymy relation or not.Existing research on hypernym detection and discovery tasks projects a term into various semantic spaces with single mapping functions.Despite their success,these methods may not be adequate in capturing complex semantic relevance between hyponym/hypernymy pairs in two aspects.First,they may fall short in modeling the hierarchical structure in the hypernymy relations,which may help them learn better term representations.Second,the polysemy phenomenon that hypernyms may express distinct senses is understudied.In this paper,we propose a Multi-Projection Recurrent model(MPR)to simultaneously capture the hierarchical relationships between terms and deal with diverse senses caused by the polysemy phenomenon.Specifically,we build a multi-projection mapping block to deal with the polysemy phenomenon,which learns various word senses by multiple projections.Besides,we adopt a hierarchy-aware recurrent block with the recurrent operation followed by a multi-hop aggregation module to capture the hierarchical structure of hypernym relations.Experiments on 11 benchmark datasets in various task settings illustrate that our multi-projection recurrent model outperforms the baselines.The experimental analysis and case study demonstrate that our multi-projection module and the recurrent structure are effective for hypernym detection and discovery tasks.
