Modeling Relation Paths for Representation Learning of Knowledge Bases. Lin, Y., Liu, Z., Luan, H., Sun, M., Rao, S., & Liu, S. In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 705–714, Lisbon, Portugal, 2015. Association for Computational Linguistics. ![Modeling Relation Paths for Representation Learning of Knowledge Bases [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex Representation learning of knowledge bases aims to embed both entities and relations into a low-dimensional space. Most existing methods only consider direct relations in representation learning. We argue that multiple-step relation paths also contain rich inference patterns between entities, and propose a path-based representation learning model. This model considers relation paths as translations between entities for representation learning, and addresses two key challenges: (1) Since not all relation paths are reliable, we design a path-constraint resource allocation algorithm to measure the reliability of relation paths. (2) We represent relation paths via semantic composition of relation embeddings. Experimental results on real-world datasets show that, as compared with baselines, our model achieves significant and consistent improvements on knowledge base completion and relation extraction from text. The source code of this paper can be obtained from https://github.com/mrlyk423/ relation_extraction.
@inproceedings{lin_modeling_2015,
address = {Lisbon, Portugal},
title = {Modeling {Relation} {Paths} for {Representation} {Learning} of {Knowledge} {Bases}},
url = {http://aclweb.org/anthology/D15-1082},
doi = {10.18653/v1/D15-1082},
abstract = {Representation learning of knowledge bases aims to embed both entities and relations into a low-dimensional space. Most existing methods only consider direct relations in representation learning. We argue that multiple-step relation paths also contain rich inference patterns between entities, and propose a path-based representation learning model. This model considers relation paths as translations between entities for representation learning, and addresses two key challenges: (1) Since not all relation paths are reliable, we design a path-constraint resource allocation algorithm to measure the reliability of relation paths. (2) We represent relation paths via semantic composition of relation embeddings. Experimental results on real-world datasets show that, as compared with baselines, our model achieves significant and consistent improvements on knowledge base completion and relation extraction from text. The source code of this paper can be obtained from https://github.com/mrlyk423/ relation\_extraction.},
language = {en},
urldate = {2019-07-09},
booktitle = {Proceedings of the 2015 {Conference} on {Empirical} {Methods} in {Natural} {Language} {Processing}},
publisher = {Association for Computational Linguistics},
author = {Lin, Yankai and Liu, Zhiyuan and Luan, Huanbo and Sun, Maosong and Rao, Siwei and Liu, Song},
year = {2015},
pages = {705--714},
}
Downloads: 0
{"_id":"EfQzjKo97GgK7BS7s","bibbaseid":"lin-liu-luan-sun-rao-liu-modelingrelationpathsforrepresentationlearningofknowledgebases-2015","author_short":["Lin, Y.","Liu, Z.","Luan, H.","Sun, M.","Rao, S.","Liu, S."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","address":"Lisbon, Portugal","title":"Modeling Relation Paths for Representation Learning of Knowledge Bases","url":"http://aclweb.org/anthology/D15-1082","doi":"10.18653/v1/D15-1082","abstract":"Representation learning of knowledge bases aims to embed both entities and relations into a low-dimensional space. Most existing methods only consider direct relations in representation learning. We argue that multiple-step relation paths also contain rich inference patterns between entities, and propose a path-based representation learning model. This model considers relation paths as translations between entities for representation learning, and addresses two key challenges: (1) Since not all relation paths are reliable, we design a path-constraint resource allocation algorithm to measure the reliability of relation paths. (2) We represent relation paths via semantic composition of relation embeddings. Experimental results on real-world datasets show that, as compared with baselines, our model achieves significant and consistent improvements on knowledge base completion and relation extraction from text. The source code of this paper can be obtained from https://github.com/mrlyk423/ relation_extraction.","language":"en","urldate":"2019-07-09","booktitle":"Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing","publisher":"Association for Computational Linguistics","author":[{"propositions":[],"lastnames":["Lin"],"firstnames":["Yankai"],"suffixes":[]},{"propositions":[],"lastnames":["Liu"],"firstnames":["Zhiyuan"],"suffixes":[]},{"propositions":[],"lastnames":["Luan"],"firstnames":["Huanbo"],"suffixes":[]},{"propositions":[],"lastnames":["Sun"],"firstnames":["Maosong"],"suffixes":[]},{"propositions":[],"lastnames":["Rao"],"firstnames":["Siwei"],"suffixes":[]},{"propositions":[],"lastnames":["Liu"],"firstnames":["Song"],"suffixes":[]}],"year":"2015","pages":"705–714","bibtex":"@inproceedings{lin_modeling_2015,\n\taddress = {Lisbon, Portugal},\n\ttitle = {Modeling {Relation} {Paths} for {Representation} {Learning} of {Knowledge} {Bases}},\n\turl = {http://aclweb.org/anthology/D15-1082},\n\tdoi = {10.18653/v1/D15-1082},\n\tabstract = {Representation learning of knowledge bases aims to embed both entities and relations into a low-dimensional space. Most existing methods only consider direct relations in representation learning. We argue that multiple-step relation paths also contain rich inference patterns between entities, and propose a path-based representation learning model. This model considers relation paths as translations between entities for representation learning, and addresses two key challenges: (1) Since not all relation paths are reliable, we design a path-constraint resource allocation algorithm to measure the reliability of relation paths. (2) We represent relation paths via semantic composition of relation embeddings. Experimental results on real-world datasets show that, as compared with baselines, our model achieves significant and consistent improvements on knowledge base completion and relation extraction from text. The source code of this paper can be obtained from https://github.com/mrlyk423/ relation\\_extraction.},\n\tlanguage = {en},\n\turldate = {2019-07-09},\n\tbooktitle = {Proceedings of the 2015 {Conference} on {Empirical} {Methods} in {Natural} {Language} {Processing}},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Lin, Yankai and Liu, Zhiyuan and Luan, Huanbo and Sun, Maosong and Rao, Siwei and Liu, Song},\n\tyear = {2015},\n\tpages = {705--714},\n}\n\n","author_short":["Lin, Y.","Liu, Z.","Luan, H.","Sun, M.","Rao, S.","Liu, S."],"key":"lin_modeling_2015","id":"lin_modeling_2015","bibbaseid":"lin-liu-luan-sun-rao-liu-modelingrelationpathsforrepresentationlearningofknowledgebases-2015","role":"author","urls":{"Paper":"http://aclweb.org/anthology/D15-1082"},"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://api.zotero.org/groups/2244199/items?key=9pbdlqiip5aNXlOfkVSImlsv&format=bibtex&limit=100","dataSources":["mxP3KHaoaaFN8BHHH"],"keywords":[],"search_terms":["modeling","relation","paths","representation","learning","knowledge","bases","lin","liu","luan","sun","rao","liu"],"title":"Modeling Relation Paths for Representation Learning of Knowledge Bases","year":2015}