Human assessments of document similarity

Westerman, S. J.; Cribbin, Timothy; Collins, James F.

doi:10.1002/asi.21361

Cited by 2 publications

(3 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Given appropriate term selection and weighting, the BOW model can produce cognitively plausible results (Westerman et al, 2010;Lee et al, 2005). However, the validity of resultant similarities can be compromised by vocabulary mismatch, which is the tendency for different words to be used to express the same concepts (Furnas, Landauer, Gomez, & Dumais, 1987).As terms are deemed independent, the semantic relationship between documents using synonymous or semantically related terms will remain obscured, wrongly suppressing similarity measures.…”

Section: B Bow = a T · Amentioning

confidence: 99%

Section: Discovering Latent Semantic Structurementioning

confidence: 99%

See 1 more Smart Citation

Discovering latent topical structure by second-order similarity analysis

Cribbin

2011

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

This is the post-print of the Article - Copyright @ 2011 ASIS&TDocument similarity models are typically derived from a term-document vector space representation by comparing all vector-pairs using some similarity measure. Computing similarity directly from a ‘bag of words’ model can be problematic because term independence causes the relationships between synonymous and related terms and the contextual influences that determine the ‘sense’ of polysemous terms to be ignored. This paper compares two methods that potentially address these problems by modelling the higher-order relationships that lie latent within the original vector space. The first is latent semantic analysis (LSA), a dimension reduction method which is a well known means of addressing the vocabulary mismatch problem in information retrieval systems. The second is the lesser known, yet conceptually simple approach of second-order similarity (SOS) analysis, where similarity is measured in terms of profiles of first-order similarities as computed directly from the term-document space. Nearest neighbour tests show that SOS analysis produces similarity models that are consistently better than both first-order and LSA derived models at resolving both coarse and fine level semantic clusters. SOS analysis has been criticised for its cubic complexity. A second contribution is the novel application of vector truncation to reduce the run-time by a constant factor. Speed-ups of four to ten times are found to be easily achievable without losing the structural benefits associated with SOS analysis

show abstract

Section: B Bow = a T · Amentioning

confidence: 99%

Section: Discovering Latent Semantic Structurementioning

confidence: 99%

Discovering latent topical structure by second-order similarity analysis

Cribbin

2011

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

show abstract

“…None of the bag‐of‐words similarity measures approach this level. Furthermore, research has shown that estimates of interrater consistency based on partial document sets can be overoptimistic (Westerman, Cribbin, & Collins, ), which bolsters our method's performance.…”

Section: Evaluation Against Human Judgmentsmentioning

confidence: 99%

Learning a concept‐based document similarity measure

Huang

Milne

Frank

et al. 2012

J Am Soc Inf Sci Tec

View full text Add to dashboard Cite

Document similarity measures are crucial components of many text analysis tasks, including information retrieval, document classification, and document clustering.Conventional measures are brittle: they estimate the surface overlap between documents based on the words they mention and ignore deeper semantic connections.We propose a new measure that assesses similarity at both the lexical and semantic levels, and learns from human judgments how to combine them by using machine learning techniques. Experiments show that the new measure produces values for documents that are more consistent with people's judgments than people are with each other. We also use it to classify and cluster large document sets covering different genres and topics, and find that it improves both classification and clustering performance.2

show abstract

Human assessments of document similarity

Cited by 2 publications

References 22 publications

Discovering latent topical structure by second-order similarity analysis

Discovering latent topical structure by second-order similarity analysis

Learning a concept‐based document similarity measure

Contact Info

Product

Resources

About