An Approach for Variable Selection in Classification with Entropy-Based Grey Relational Analysis

Serkan Akogul

Makale detayları

An Approach for Variable Selection in Classification with Entropy-Based Grey Relational Analysis

Yazarlar:

Serkan Akogul

Yayıncı:

Süleyman Demirel University

Ana Başlıklar:

Strateji | Yönetim ve Örgütsel Davranış (Diğer)

Dergi:

Süleyman Demirel Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi Vol.29 Issue 1, pp.1-12

Belge Tipi:

Makale

Yayın Tarihi:

4/2024

Belge Dili:

Türkçe

Document ID (Org No.):

2001316975

pISSN:

1301-0603

Anahtar kelimeler:

Özet:

 Nowadays, there has been a rapid increase in data size due to developments in information and measurement technologies. Due to the high dimensionality, many analyses, such as classification and clustering, require the use of data reduction methods before analysis. Variable selection plays an important role in interpreting the classification and clustering results of the data set and obtaining successful results with fewer variables. In this study, the variable selection problem to be used in classification is considered a multi-criteria decision-making MCDM problem. The decision matrix was created by taking variables as alternatives and filtering-based variable selection algorithms as criteria. This decision matrix was analyzed using the Grey Relational Analysis GRA method, which is one of the MCDM methods. Thus, a new approach for ranking and selecting variables has been proposed. In addition, entropy, which is an objective method, is used for weighting the criteria. The proposed method is applied to the Wine and Ionosphere datasets, which are frequently used in the classification literature, and compared with the classification success of filtering-based variable selection algorithms. As a result, successful results were obtained with fewer variables using the proposed approach.

 Nowadays, there has been a rapid increase in data size due to developments in information and measurement technologies. Due to the high dimensionality, many analyses, such as classification and clustering, require the use of data reduction methods before analysis. Variable selection plays an important role in interpreting the classification and clustering results of the data set and obtaining successful results with fewer variables. In this study, the variable selection problem to be used in classification is considered a multi-criteria decision-making MCDM problem. The decision matrix was created by taking variables as alternatives and filtering-based variable selection algorithms as criteria. This decision matrix was analyzed using the Grey Relational Analysis GRA method, which is one of the MCDM methods. Thus, a new approach for ranking and selecting variables has been proposed. In addition, entropy, which is an objective method, is used for weighting the criteria. The proposed method is applied to the Wine and Ionosphere datasets, which are frequently used in the classification literature, and compared with the classification success of filtering-based variable selection algorithms. As a result, successful results were obtained with fewer variables using the proposed approach.

 Nowadays, there has been a rapid increase in data size due to developments in information and measurement technologies. Due to the high dimensionality, many analyses, such as classification and clustering, require the use of data reduction methods before analysis. Variable selection plays an important role in interpreting the classification and clustering results of the data set and obtaining successful results with fewer variables. In this study, the variable selection problem to be used in classification is considered a multi-criteria decision-making MCDM problem. The decision matrix was created by taking variables as alternatives and filtering-based variable selection algorithms as criteria. This decision matrix was analyzed using the Grey Relational Analysis GRA method, which is one of the MCDM methods. Thus, a new approach for ranking and selecting variables has been proposed. In addition, entropy, which is an objective method, is used for weighting the criteria. The proposed method is applied to the Wine and Ionosphere datasets, which are frequently used in the classification literature, and compared with the classification success of filtering-based variable selection algorithms. As a result, successful results were obtained with fewer variables using the proposed approach.

 Günümüzde bilgi ve ölçüm teknolojilerindeki gelişmelere bağlı olarak veri boyutlarında da hızlı bir artış olmuştur. Yüksek boyutluluk nedeniyle özellikle sınıflandırma ve kümeleme gibi birçok analiz için, analiz öncesi veri indirgeme yöntemlerinin kullanılması gerekmektedir. Veri setinin sınıflandırma ve kümeleme sonuçlarının yorumlanması ve daha az değişken ile başarılı sonuçlar elde edilmesi için değişken seçimi önemli rol oynamaktadır. Bu çalışmada, sınıflandırmada kullanılacak değişken seçimi problemi çok kriterli karar verme ÇKKV problemi olarak düşünülmüştür. Alternatif olarak değişkenler, kriterler ise filtreleme tabanlı değişken seçim algoritmaları alınarak karar matrisi oluşturulmuş. Bu karar matrisi, ÇKKV yöntemlerinden birisi olan Gri İlişkisel Analiz GİA yöntemi kullanarak analiz edilmiştir. Böylelikle değişkenlerin sıralanması ve seçimi için yeni bir yaklaşım ortaya konulmuştur. Ayrıca kriterlerin ağırlıklandırılmasında objektif bir yöntem olan Entropi den yararlanılmıştır. Önerilen yöntem sınıflandırma literatüründe sıkça kullanılan Wine Şarap ve Ionosphere İyonosfer veri setlerinde uygulanmış olup, filtreleme tabanlı değişken seçim algoritmalarının sınıflandırma başarılarıyla karşılaştırılmıştır. Sonuç olarak önerilen yaklaşım kullanılarak az değişken ile başarılı sonuçlar elde edilmiştir.

Makale detayları

An Approach for Variable Selection in Classification with Entropy-Based Grey Relational Analysis

Yazarlar:

Serkan Akogul

Yayıncı:

Süleyman Demirel University

Ana Başlıklar:

Strateji | Yönetim ve Örgütsel Davranış (Diğer)

Dergi:

Süleyman Demirel Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi Vol.29 Issue 1, pp.1-12

Belge Tipi:

Makale

Yayın Tarihi:

4/2024

Belge Dili:

Türkçe

Document ID (Org No.):

2001316975

pISSN:

1301-0603

Entropy

Filter-based variable selection

Grey Relational Analysis (GRA)

Multi-criteria decision making (MCDM)Çok kriterli karar verme (ÇKKV)

Entropi

Filtreleme tabanlı değişken seçimi

Gri İlişkisel Analiz (GİA).

Sayı

Makale detayları

Yazarlar: Serkan Akogul

Yayıncı: Süleyman Demirel University

Ana Başlıklar: Strateji | Yönetim ve Örgütsel Davranış (Diğer)

Dergi: Süleyman Demirel Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi Vol.29 Issue 1, pp.1-12

Belge Tipi: Makale

Yayın Tarihi: 4/2024

Belge Dili: Türkçe

Document ID (Org No.): 2001316975

pISSN: 1301-0603

Sayı

Yazarlar:

Serkan Akogul

Yayıncı:

Süleyman Demirel University

Ana Başlıklar:

Strateji | Yönetim ve Örgütsel Davranış (Diğer)

Dergi:

Süleyman Demirel Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi Vol.29 Issue 1, pp.1-12

Belge Tipi:

Makale

Yayın Tarihi:

4/2024

Belge Dili:

Türkçe

Document ID (Org No.):

2001316975

pISSN:

1301-0603