An Approach for Measuring Semantic Similarity between Words Using Multiple Information Sources

By Yuhua Li, Zuhair A. Bandar, David McLean. IEEE Transactions on Knowledge and Data Engineering, 2003

1,221 citations recorded by the source corpus.

Semantic similarity between words is becoming a generic problem for many applications of computational linguistics and artificial intelligence. This paper explores the determination of semantic similarity by a number of information sources, which consist of structural semantic information from a lexical taxonomy and information content from a corpus. To investigate how information sources could be used effectively, a variety of strategies for using various possible information sources are implemented. A new measure is then proposed which combines information sources nonlinearly. Experimental evaluation against a benchmark set of human similarity ratings demonstrates that the proposed measure significantly outperforms traditional similarity measures.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/cc8436d52b0ceb1db6b164ef808d8f5a2e827c07

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

UMBC_EBIQUITY-CORE: Semantic Textual Similarity Systems (2013, methodology): “Using the e − αD ( x,y ) to transform simple shortest path length has been demonstrated to be very effective according to (Li et al., 2003).”
Using ODP metadata to personalize search (2005, background): “We refer the reader to [16] for an in-depth view of the approach we took in this study.”
Evolution of Semantic Similarity—A Survey (2020, methodology): “…as t lcs , and the shortest path length between them denoted as min _ len ( t 1 , t 2 ) , path is measured as, and wup is measured as, Li et al.[49] proposed a measure that takes into account both the minimum path distance and depth. li is measured as, e βdepth ( t lcs ) − e − βdepth ( t lcs )…”
Semantic similarity methods in wordNet and their application to information retrieval on the web (2005, background): “[5] is among the best and it is also the fastest.”
Ontology-based semantic similarity: A new feature-based approach (2012, background): “Both Lin (1998) and Jiang and Conrath (1997) extended Resnik’s work by also considering the IC of each of the evaluated terms.”
Samsung Poland NLP Team at SemEval-2016 Task 1: Necessity for diversity; combining recursive autoencoders, WordNet and ensemble methods to measure semantic similarity. (2016, methodology): “The semantic distance D ( x, y ) is transformed to an award A using the formula e − αD ( x,y ) introduced by (Li et al., 2003), where α is set to 0.25, as this value seemed to yield the best results: where β is a positive number (5 by default) used to control the level of adjustment made by the…”