We explore methods to extract relations between named entities from free text in an unsupervised setting. In addition to standard feature extraction, we develop a novel method to re-weight word embeddings. We alleviate the problem of features sparsity using an individual feature reduction. Our approach exhibits a significant improvement by 5.8% over the state-of-the-art relation clustering scoring a F1-score of 0.416 on the NYT-FB dataset.
%0 Conference Paper
%1 elsahar2017unsupervised
%A Elsahar, Hady
%A Demidova, Elena
%A Gottschalk, Simon
%A Gravier, Christophe
%A Laforest, Frederique
%B Proceedings of the ESWC 2017 Satellite Events
%D 2017
%I Springer
%K alexandria data4urbanmobility gottschalk myown wdaqua
%R 10.1007/978-3-319-70407-4_3
%T Unsupervised Open Relation Extraction
%V Lecture Notes in Computer Science (LNCS), vol 10577.
%X We explore methods to extract relations between named entities from free text in an unsupervised setting. In addition to standard feature extraction, we develop a novel method to re-weight word embeddings. We alleviate the problem of features sparsity using an individual feature reduction. Our approach exhibits a significant improvement by 5.8% over the state-of-the-art relation clustering scoring a F1-score of 0.416 on the NYT-FB dataset.
@inproceedings{elsahar2017unsupervised,
abstract = {We explore methods to extract relations between named entities from free text in an unsupervised setting. In addition to standard feature extraction, we develop a novel method to re-weight word embeddings. We alleviate the problem of features sparsity using an individual feature reduction. Our approach exhibits a significant improvement by 5.8% over the state-of-the-art relation clustering scoring a F1-score of 0.416 on the NYT-FB dataset.},
added-at = {2017-05-21T10:23:19.000+0200},
author = {Elsahar, Hady and Demidova, Elena and Gottschalk, Simon and Gravier, Christophe and Laforest, Frederique},
biburl = {https://www.bibsonomy.org/bibtex/20af38519561d89a93971f8ecb8938c7f/demidova},
booktitle = {Proceedings of the ESWC 2017 Satellite Events},
doi = {10.1007/978-3-319-70407-4_3},
interhash = {84347f4d1dbaa3cbb4649be7c8355922},
intrahash = {0af38519561d89a93971f8ecb8938c7f},
keywords = {alexandria data4urbanmobility gottschalk myown wdaqua},
publisher = {Springer},
timestamp = {2018-10-23T09:52:15.000+0200},
title = {Unsupervised Open Relation Extraction},
volume = {Lecture Notes in Computer Science (LNCS), vol 10577.},
year = 2017
}