Computing Semantic Relatedness Using Word Frequency and Layout Information of Wikipedia
Authors | Patrick Chan Yoshinori Hijikata Shogo Nishida |
|---|---|
Publication date | 2013 |
DOI | 10.1145/2480362.2480424 |
Links | Original |
Computing Semantic Relatedness Using Word Frequency and Layout Information of Wikipedia - scientific work related to Wikipedia quality published in 2013, written by Patrick Chan, Yoshinori Hijikata and Shogo Nishida.
Overview
Computing the semantic relatedness between two words or phrases is an important problem for fields such as information retrieval and natural language processing. One state-of-the-art approach to solve the problem is Explicit Semantic Analysis (ESA). ESA uses the word frequency in Wikipedia articles to estimate the relevance, so the relevance of words with low frequency cannot always be well estimated. To improve the relevance estimate of the low frequency words, authors use not only word frequency but also layout information in Wikipedia articles. Empirical evaluation shows that on the low frequency words, method achieves better estimate of semantic relatedness over ESA.
Embed
Wikipedia Quality
Chan, Patrick; Hijikata, Yoshinori; Nishida, Shogo. (2013). "[[Computing Semantic Relatedness Using Word Frequency and Layout Information of Wikipedia]]".DOI: 10.1145/2480362.2480424.
English Wikipedia
{{cite journal |last1=Chan |first1=Patrick |last2=Hijikata |first2=Yoshinori |last3=Nishida |first3=Shogo |title=Computing Semantic Relatedness Using Word Frequency and Layout Information of Wikipedia |date=2013 |doi=10.1145/2480362.2480424 |url=https://wikipediaquality.com/wiki/Computing_Semantic_Relatedness_Using_Word_Frequency_and_Layout_Information_of_Wikipedia}}
HTML
Chan, Patrick; Hijikata, Yoshinori; Nishida, Shogo. (2013). "<a href="https://wikipediaquality.com/wiki/Computing_Semantic_Relatedness_Using_Word_Frequency_and_Layout_Information_of_Wikipedia">Computing Semantic Relatedness Using Word Frequency and Layout Information of Wikipedia</a>".DOI: 10.1145/2480362.2480424.
