LSA: First dimension and dimensional weighting

X. Hu; D Franceschetti; P. Penumatsa; AC Graesser; M Louwerse; D. S. McNamara; Tutoring Research Group

LSA: First dimension and dimensional weighting

X. Hu, D Franceschetti, P. Penumatsa, AC Graesser, M Louwerse, D. S. McNamara, Tutoring Research Group

Research output: Chapter in Book/Report/Conference proceeding › Chapter › Scientific › peer-review

Abstract

We report two discoveries concerning Latent Semantic Analysis (LSA). First, we observed the special properties of the first dimension of the LSA space. Second, we observed that dimensional weighting plays an important role in LSA analysis. Based on the first discovery, we examined the cosine matches without the first dimension. Based on the second discovery, we explored different dimensional weighting schemes. Based on these observations, we recommend a new algorithm for LSA cosine computation such that LSA becomes more sensitive to relevant similarities and differences.

Original language	English
Title of host publication	Proceedings of the 25h Annual Conference of the Cognitive Science Society
Editors	Richard Alterman, David Kirsh
Publisher	Cognitive Science Society
Pages	587-592
Publication status	Published - 2003
Externally published	Yes
Event	25th Annual Conference of the Cognitive Science Society - Boston, United States Duration: 31 Jul 2003 → 2 Aug 2003 Conference number: 25 http://csjarchive.cogsci.rpi.edu/proceedings/2003/mac/index.html

Conference

Conference	25th Annual Conference of the Cognitive Science Society
Abbreviated title	CogSci 2003
Country/Territory	United States
City	Boston
Period	31/07/03 → 2/08/03
Internet address	http://csjarchive.cogsci.rpi.edu/proceedings/2003/mac/index.html

Access to Document

http://csjarchive.cogsci.rpi.edu/proceedings/2003/pdfs/122.pdf

Cite this

Hu, X., Franceschetti, D., Penumatsa, P., Graesser, AC., Louwerse, M., McNamara, D. S., & Group, T. R. (2003). LSA: First dimension and dimensional weighting. In R. Alterman, & D. Kirsh (Eds.), Proceedings of the 25h Annual Conference of the Cognitive Science Society (pp. 587-592). Cognitive Science Society. http://csjarchive.cogsci.rpi.edu/proceedings/2003/pdfs/122.pdf

@inbook{5083ad8a110b442cb40fb614179ec41c,

title = "LSA: First dimension and dimensional weighting",

abstract = "We report two discoveries concerning Latent Semantic Analysis (LSA). First, we observed the special properties of the first dimension of the LSA space. Second, we observed that dimensional weighting plays an important role in LSA analysis. Based on the first discovery, we examined the cosine matches without the first dimension. Based on the second discovery, we explored different dimensional weighting schemes. Based on these observations, we recommend a new algorithm for LSA cosine computation such that LSA becomes more sensitive to relevant similarities and differences.",

author = "X. Hu and D Franceschetti and P. Penumatsa and AC Graesser and M Louwerse and McNamara, {D. S.} and Group, {Tutoring Research}",

year = "2003",

language = "English",

pages = "587--592",

editor = "Richard Alterman and David Kirsh",

booktitle = "Proceedings of the 25h Annual Conference of the Cognitive Science Society",

publisher = "Cognitive Science Society",

note = "25th Annual Conference of the Cognitive Science Society, CogSci 2003 ; Conference date: 31-07-2003 Through 02-08-2003",

url = "http://csjarchive.cogsci.rpi.edu/proceedings/2003/mac/index.html",

}

Hu, X, Franceschetti, D, Penumatsa, P, Graesser, AC, Louwerse, M, McNamara, DS & Group, TR 2003, LSA: First dimension and dimensional weighting. in R Alterman & D Kirsh (eds), Proceedings of the 25h Annual Conference of the Cognitive Science Society. Cognitive Science Society, pp. 587-592, 25th Annual Conference of the Cognitive Science Society, Boston, Massachusetts, United States, 31/07/03. <http://csjarchive.cogsci.rpi.edu/proceedings/2003/pdfs/122.pdf>

TY - CHAP

T1 - LSA

T2 - 25th Annual Conference of the Cognitive Science Society

AU - Hu, X.

AU - Franceschetti, D

AU - Penumatsa, P.

AU - Graesser, AC

AU - Louwerse, M

AU - McNamara, D. S.

AU - Group, Tutoring Research

N1 - Conference code: 25

PY - 2003

Y1 - 2003

N2 - We report two discoveries concerning Latent Semantic Analysis (LSA). First, we observed the special properties of the first dimension of the LSA space. Second, we observed that dimensional weighting plays an important role in LSA analysis. Based on the first discovery, we examined the cosine matches without the first dimension. Based on the second discovery, we explored different dimensional weighting schemes. Based on these observations, we recommend a new algorithm for LSA cosine computation such that LSA becomes more sensitive to relevant similarities and differences.

AB - We report two discoveries concerning Latent Semantic Analysis (LSA). First, we observed the special properties of the first dimension of the LSA space. Second, we observed that dimensional weighting plays an important role in LSA analysis. Based on the first discovery, we examined the cosine matches without the first dimension. Based on the second discovery, we explored different dimensional weighting schemes. Based on these observations, we recommend a new algorithm for LSA cosine computation such that LSA becomes more sensitive to relevant similarities and differences.

M3 - Chapter

SP - 587

EP - 592

BT - Proceedings of the 25h Annual Conference of the Cognitive Science Society

A2 - Alterman, Richard

A2 - Kirsh, David

PB - Cognitive Science Society

Y2 - 31 July 2003 through 2 August 2003

ER -

LSA: First dimension and dimensional weighting

Abstract

Conference

Access to Document

Fingerprint

Cite this