Publication: SOSYAL POLİTİKA ÜZERİNE YAZILMIŞ MAKALELERİN METİN MADENCİLİĞİYLE_x000D_
ANALİZİ
Loading...
Files
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Bu çalışmanın amacı, 2009-2019 yılları arasında Ulakbim ve Science Direct veri tabanlarında_x000D_
yer alan “sosyal politika”ya ilişkin makalelerin ilgi alanlarını ve bu makalelerde kullanılan anahtar_x000D_
kelimeler arasındaki birliktelik kurallarını incelemektir. Ulakbim veri tabanında 192 Türkçe ve Science_x000D_
Direct veri tabanında ise 481 İngilizce olmak üzere toplam 673 makale incelenmiştir. Yöntem olarak,_x000D_
Ulakbim ve Science Direct veri tabanları üzerinden indirilen makalelere ait anahtar kelimelerin_x000D_
kullanılma sıklığı incelenmiş daha sonra açık kaynak kodlu bir veri madenciliği yazılımı olan_x000D_
RapidMiner ile anahtar kelimeler üzerinde birliktelik kuralları analizi yapılmıştır. Çalışmanın bulguları_x000D_
incelendiğinde her iki veri tabanında yer alan makalelerde farklı ve benzer ilgi alanlarına rastlanmış_x000D_
ayrıca birliktelik kuralları bakımından aynı anahtar kelimeler üzerinde değerlendirme yapıldığında,_x000D_
konuların ele alınışının farklı boyutlara dayandığı görülmüştür. Türkiye’deki çalışmalarda devlet,_x000D_
refah, hizmet, yoksulluk ve eğitim, Science Direct veri tabanındaki çalışmalarda ise sağlık, yoksulluk,_x000D_
refah, bakım ve çocuk başlıkları ön plana çıkmıştır. Birliktelik kuralarında ise Ulakbim veri tabanındaki_x000D_
çalışmalarda refah devletinin diğer anahtar kelimelerle birlikte kullanımı yaygınken, Science Direct_x000D_
veri tabanındaki çalışmalarda birliktelik kuralları daha fazla çeşitlilik arz etmektedir. Çalışmanın en_x000D_
büyük kısıtı belirli bir zaman aralığıyla yetinmesi ve sadece iki veri tabanı üzerinden sosyal politikanın_x000D_
yönünün ve ilgi alanının tartışılmasıdır. Sosyal politika alanında benzer bir çalışmanın yapılmamış_x000D_
olması ve bu alana farklı bir bakış kazandıracak bir yöntemle konunun irdelenmesi, çalışmanın özgün_x000D_
yanını oluşturmaktadır.
The objective of this study is to investigate respective parts and examine association rules related_x000D_ to key words of articles contained in Ulakbim and Science Direct databases which are associated with_x000D_ ‘’social politics’’ between the years 2009-2019. Total of 673 articles were examined which were_x000D_ comprised of 192 articles in Turkish from the Ulakbim database and 481 articles in English from the_x000D_ Science Direct database. The methodology was to examine the frequency of the key words in the articles_x000D_ downloaded from Ulakbim and Science Direct databases and afterwards the association rule analysis_x000D_ was performed based on the key words by using an open source data mining software named_x000D_ RapidMiner. Upon examination of the results it can be concluded that the articles from the two_x000D_ databases under consideration contained both different and similar parts, whereas the result of_x000D_ association rule analysis made on same key words showed that handling of the subjects in the articles_x000D_ was based on different views. The Turkish studies emphasized government, welfare, care, poverty and_x000D_ education while studies in Science Direct database concentrated on health, poverty, welfare, care and_x000D_ children. The biggest restriction of the study is that it examines only a certain time period and only two_x000D_ databases which include articles arguing and discussing social politics. However, the unique part is_x000D_ that no other study was ever made in the area of social politics and that the study itself was conducted_x000D_ by using a methodology which brought a new perspective to the subject.
The objective of this study is to investigate respective parts and examine association rules related_x000D_ to key words of articles contained in Ulakbim and Science Direct databases which are associated with_x000D_ ‘’social politics’’ between the years 2009-2019. Total of 673 articles were examined which were_x000D_ comprised of 192 articles in Turkish from the Ulakbim database and 481 articles in English from the_x000D_ Science Direct database. The methodology was to examine the frequency of the key words in the articles_x000D_ downloaded from Ulakbim and Science Direct databases and afterwards the association rule analysis_x000D_ was performed based on the key words by using an open source data mining software named_x000D_ RapidMiner. Upon examination of the results it can be concluded that the articles from the two_x000D_ databases under consideration contained both different and similar parts, whereas the result of_x000D_ association rule analysis made on same key words showed that handling of the subjects in the articles_x000D_ was based on different views. The Turkish studies emphasized government, welfare, care, poverty and_x000D_ education while studies in Science Direct database concentrated on health, poverty, welfare, care and_x000D_ children. The biggest restriction of the study is that it examines only a certain time period and only two_x000D_ databases which include articles arguing and discussing social politics. However, the unique part is_x000D_ that no other study was ever made in the area of social politics and that the study itself was conducted_x000D_ by using a methodology which brought a new perspective to the subject.
