Publication Details
Combining Heterogeneous Models for Measuring Relational Similarity
Yih Wen-tau (FIT)
Meek Christopher (FIT)
Mikolov Tomáš, Ing., Ph.D.
Zweig Geoffrey
language modeling, heterogeneous models, recurrent neural networks
In this paper, we presented a system that combines heterogeneous models based on different information sources for measuring relational similarity.
In this work, we study the problem of measuring relational similarity between two word pairs (e.g., silverware:fork and clothing:shirt). Due to the large number of possible relations, we argue that it is important to combine multiple models based on heterogeneous information sources. Our overall system consists of two novel general-purpose relational similarity models and three specific word relation models. When evaluated in the setting of a recently proposed SemEval-2012 task, our approach outperforms the previous best system substantially, achieving a 54.1% relative increase in Spearmans rank correlation.
@inproceedings{BUT105978,
author="Alisa {Zhila} and Wen-tau {Yih} and Christopher {Meek} and Tomáš {Mikolov} and Geoffrey {Zweig}",
title="Combining Heterogeneous Models for Measuring Relational Similarity",
booktitle="Proceedings of NAACL-HLT 2013",
year="2013",
pages="1000--1009",
publisher="Association for Computational Linguistics",
address="Atlanata, Georgia",
isbn="978-1-937284-47-3",
url="http://www.aclweb.org/anthology/N/N13/N13-1120.pdf"
}