KNN algoritması ve R dili ile metin madenciliği kullanılarak bilimsel makale tasnifi

Deniz Kılınç; Emin Borandağ; Fatih Yücalar; Volkan Tunali; Macit Şimşek; Akın Özçift

doi:10.7240/mufbed.69674

KNN algoritması ve R dili ile metin madenciliği kullanılarak bilimsel makale tasnifi

Translated title of the contribution: Classification of scientific articles using text mining with KNN algorithm and R language

Deniz Kılınç^*, Emin Borandağ, Fatih Yücalar, Volkan Tunali, Macit Şimşek, Akın Özçift

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

Abstract

In order to perform analysis on text-based datasets, the techniques and methods in Text Mining (TM) which is a subdomain of Data Mining are used. In this study, it is aimed to evaluate the classification accuracy of academic articles which are produced in academic domain. In accordance with this purpose, the abstracts of the academic articles are obtained and a dataset is created from an academic knowledge sharing network named Research Gate by using self-developed software tools. The academic articles in the dataset fall into two categories as “Materials Science & Engineering” and “Social Sciences & Humanities”. KNN (k-nearest neighbors) classification algorithm is performed by utilizing R language and R Studio tools on the dataset. The experimental results show that the classification accuracy (ACC) of KNN is obtained as 96.67%.

Translated title of the contribution	Classification of scientific articles using text mining with KNN algorithm and R language
Original language	Turkish
Pages (from-to)	89-94
Number of pages	6
Journal	International Journal of Advances in Engineering and Pure Sciences
Volume	3
DOIs	https://doi.org/10.7240/mufbed.69674
Publication status	Published - 31 Dec 2016
Externally published	Yes

Keywords

text mining
R
R Studio
KNN
text classification

Access to Document

10.7240/mufbed.69674

https://dergipark.org.tr/en/pub/marufbd/issue/27585/290473

Cite this

@article{1d2d9969be254d52b7765a384087a7d4,

title = "KNN algoritması ve R dili ile metin madenciliği kullanılarak bilimsel makale tasnifi",

abstract = "In order to perform analysis on text-based datasets, the techniques and methods in Text Mining (TM) which is a subdomain of Data Mining are used. In this study, it is aimed to evaluate the classification accuracy of academic articles which are produced in academic domain. In accordance with this purpose, the abstracts of the academic articles are obtained and a dataset is created from an academic knowledge sharing network named Research Gate by using self-developed software tools. The academic articles in the dataset fall into two categories as “Materials Science & Engineering” and “Social Sciences & Humanities”. KNN (k-nearest neighbors) classification algorithm is performed by utilizing R language and R Studio tools on the dataset. The experimental results show that the classification accuracy (ACC) of KNN is obtained as 96.67%.",

keywords = "text mining, R, R Studio, KNN, text classification",

author = "Deniz Kılın{\c c} and Emin Borandağ and Fatih Y{\"u}calar and Volkan Tunali and Macit {\c S}im{\c s}ek and Akın {\"O}z{\c c}ift",

year = "2016",

month = dec,

day = "31",

doi = "10.7240/mufbed.69674",

language = "Turkish",

volume = "3",

pages = "89--94",

journal = "International Journal of Advances in Engineering and Pure Sciences",

issn = "2636-8277",

}

TY - JOUR

T1 - KNN algoritması ve R dili ile metin madenciliği kullanılarak bilimsel makale tasnifi

AU - Kılınç, Deniz

AU - Borandağ, Emin

AU - Yücalar, Fatih

AU - Tunali, Volkan

AU - Şimşek, Macit

AU - Özçift, Akın

PY - 2016/12/31

Y1 - 2016/12/31

N2 - In order to perform analysis on text-based datasets, the techniques and methods in Text Mining (TM) which is a subdomain of Data Mining are used. In this study, it is aimed to evaluate the classification accuracy of academic articles which are produced in academic domain. In accordance with this purpose, the abstracts of the academic articles are obtained and a dataset is created from an academic knowledge sharing network named Research Gate by using self-developed software tools. The academic articles in the dataset fall into two categories as “Materials Science & Engineering” and “Social Sciences & Humanities”. KNN (k-nearest neighbors) classification algorithm is performed by utilizing R language and R Studio tools on the dataset. The experimental results show that the classification accuracy (ACC) of KNN is obtained as 96.67%.

AB - In order to perform analysis on text-based datasets, the techniques and methods in Text Mining (TM) which is a subdomain of Data Mining are used. In this study, it is aimed to evaluate the classification accuracy of academic articles which are produced in academic domain. In accordance with this purpose, the abstracts of the academic articles are obtained and a dataset is created from an academic knowledge sharing network named Research Gate by using self-developed software tools. The academic articles in the dataset fall into two categories as “Materials Science & Engineering” and “Social Sciences & Humanities”. KNN (k-nearest neighbors) classification algorithm is performed by utilizing R language and R Studio tools on the dataset. The experimental results show that the classification accuracy (ACC) of KNN is obtained as 96.67%.

KW - text mining

KW - R

KW - R Studio

KW - KNN

KW - text classification

U2 - 10.7240/mufbed.69674

DO - 10.7240/mufbed.69674

M3 - Article

SN - 2636-8277

VL - 3

SP - 89

EP - 94

JO - International Journal of Advances in Engineering and Pure Sciences

JF - International Journal of Advances in Engineering and Pure Sciences

ER -

KNN algoritması ve R dili ile metin madenciliği kullanılarak bilimsel makale tasnifi

Abstract

Keywords

Access to Document

Fingerprint

Cite this