Examining the impact of stemming on clustering Turkish texts

Volkan Tunali; Turgay Tugay Bilgin

doi:10.1109/INISTA.2012.6246966

Examining the impact of stemming on clustering Turkish texts

Volkan Tunali^*, Turgay Tugay Bilgin

^*Corresponding author for this work

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

5 Citations (Scopus)

Abstract

Preprocessing is an important step in information retrieval and text mining. In this study, we examined the impact of stemming on clustering Turkish texts. We used two datasets compiled from web sites of Turkish news agencies, and performed extensive experiments. We empirically show that there is no significant evidence that stemming always improves the quality of clustering for texts in Turkish. However, when stemming is used, dimensionality of the document-term matrix dramatically decreases without inversely affecting the clustering performance. As a result, it is highly recommended to apply stemming for clustering Turkish texts.

Original language	English
Title of host publication	2012 International Symposium on Innovations in Intelligent Systems and Applications
Publisher	IEEE
Number of pages	4
ISBN (Electronic)	9781467314480
ISBN (Print)	9781467314466
DOIs	https://doi.org/10.1109/INISTA.2012.6246966
Publication status	Published - 23 Jul 2012
Externally published	Yes
Event	2012 International Symposium on Innovations in Intelligent Systems and Applications - Trabzon, Turkey Duration: 2 Jul 2012 → 4 Jul 2012

Conference

Conference	2012 International Symposium on Innovations in Intelligent Systems and Applications
Abbreviated title	INISTA
Country/Territory	Turkey
City	Trabzon
Period	2/07/12 → 4/07/12

Keywords

data mining
text mining
document clustering
preprocessing
stemming

Access to Document

10.1109/INISTA.2012.6246966Licence: Other

Cite this

@inproceedings{2eb55c9e321f4d11a7ce32e29f7e8908,

title = "Examining the impact of stemming on clustering Turkish texts",

abstract = "Preprocessing is an important step in information retrieval and text mining. In this study, we examined the impact of stemming on clustering Turkish texts. We used two datasets compiled from web sites of Turkish news agencies, and performed extensive experiments. We empirically show that there is no significant evidence that stemming always improves the quality of clustering for texts in Turkish. However, when stemming is used, dimensionality of the document-term matrix dramatically decreases without inversely affecting the clustering performance. As a result, it is highly recommended to apply stemming for clustering Turkish texts.",

keywords = "data mining, text mining, document clustering, preprocessing, stemming",

author = "Volkan Tunali and Bilgin, {Turgay Tugay}",

year = "2012",

month = jul,

day = "23",

doi = "10.1109/INISTA.2012.6246966",

language = "English",

isbn = "9781467314466",

booktitle = "2012 International Symposium on Innovations in Intelligent Systems and Applications",

publisher = "IEEE",

address = "United States",

note = "2012 International Symposium on Innovations in Intelligent Systems and Applications, INISTA ; Conference date: 02-07-2012 Through 04-07-2012",

}

TY - GEN

T1 - Examining the impact of stemming on clustering Turkish texts

AU - Tunali, Volkan

AU - Bilgin, Turgay Tugay

PY - 2012/7/23

Y1 - 2012/7/23

N2 - Preprocessing is an important step in information retrieval and text mining. In this study, we examined the impact of stemming on clustering Turkish texts. We used two datasets compiled from web sites of Turkish news agencies, and performed extensive experiments. We empirically show that there is no significant evidence that stemming always improves the quality of clustering for texts in Turkish. However, when stemming is used, dimensionality of the document-term matrix dramatically decreases without inversely affecting the clustering performance. As a result, it is highly recommended to apply stemming for clustering Turkish texts.

AB - Preprocessing is an important step in information retrieval and text mining. In this study, we examined the impact of stemming on clustering Turkish texts. We used two datasets compiled from web sites of Turkish news agencies, and performed extensive experiments. We empirically show that there is no significant evidence that stemming always improves the quality of clustering for texts in Turkish. However, when stemming is used, dimensionality of the document-term matrix dramatically decreases without inversely affecting the clustering performance. As a result, it is highly recommended to apply stemming for clustering Turkish texts.

KW - data mining

KW - text mining

KW - document clustering

KW - preprocessing

KW - stemming

U2 - 10.1109/INISTA.2012.6246966

DO - 10.1109/INISTA.2012.6246966

M3 - Conference contribution

SN - 9781467314466

BT - 2012 International Symposium on Innovations in Intelligent Systems and Applications

PB - IEEE

T2 - 2012 International Symposium on Innovations in Intelligent Systems and Applications

Y2 - 2 July 2012 through 4 July 2012

ER -

Examining the impact of stemming on clustering Turkish texts

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this