Word sense distribution in a web corpus

Chen, Ping; Brown, David P.; Tran, Andrew; Ozoka, Noble; Ortiz, Rafael Seiz

doi:10.1109/coginf.2010.5599697

Cited by 1 publication

(2 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If we conduct a more detailed analysis word by word and compute the correlations of each test word, we can observe that there are three words ("newspaper", "production", and "procedure") with moderate correlations (between 0.80 and 0.86), four words ("energy", "church", "trial", and "element") with low correlations (between 0.51 and 0.79), and three ("particle", "function", and "relationship") without any correlation at all, since their values are lower than the significance level (0.35, p<.01) established for tests with 50 pairs. It should be noticed that these results seem to be not in accordance with those experiments reported in [6], where the web-based corpus is "quite similar" to SemCor in terms of sense distribution. In that work, the notion of "quite similar" must be considered as a non-technical and naive intuition, since no correlation measure was computed.…”

Section: Pearson Correlationcontrasting

confidence: 54%

See 1 more Smart Citation

Analyzing the Sense Distribution of Concordances Obtained by Web as Corpus Approach

Saralegi¹,

Gamallo

2013

Computational Linguistics and Intelligent Text Processing

View full text Add to dashboard Cite

Abstract. In corpus-based lexicography and natural language processing fields some authors have proposed using the Internet as a source of corpora for obtaining concordances of words. Most techniques implemented with this method are based on information retrieval-oriented web searchers. However, rankings of concordances obtained by these search engines are not built according to linguistic criteria but to topic similarity or navigational oriented criteria, such as page-rank. It follows that examples or concordances could not be linguistically representative, and so, linguistic knowledge mined by these methods might not be very useful. This work analyzes the linguistic representativeness of concordances obtained by different relevance criteria based web search engines (web, blog and news search engines). The analysis consists of comparing web concordances and SemCor (the reference) with regard to the distribution of word senses. Results showed that sense distributions in concordances obtained by web search engines are, in general, quite different from those obtained from the reference corpus. Among the search engines, those that were found to be the most similar to the reference were the informational oriented engines (news and blog search engines).

show abstract

Section: Pearson Correlationcontrasting

confidence: 54%

“…Chen et al [6] describe an experiment to manually evaluate concordances included in web documents retrieved from an SE for 10 test words. They annotated by hand about 1,000 to 3,000 instances for each test word.…”

Section: Related Workmentioning

confidence: 99%

Analyzing the Sense Distribution of Concordances Obtained by Web as Corpus Approach

Saralegi¹,

Gamallo

2013

Computational Linguistics and Intelligent Text Processing

View full text Add to dashboard Cite

show abstract

Word sense distribution in a web corpus

Cited by 1 publication

References 9 publications

Analyzing the Sense Distribution of Concordances Obtained by Web as Corpus Approach

Analyzing the Sense Distribution of Concordances Obtained by Web as Corpus Approach

Contact Info

Product

Resources

About