The Effectiveness of Feature Selection Methods for Imbalanced Text Classification

Hande Tiryaki; Alper Kürşat Uysal

Makale detayları

The Effectiveness of Feature Selection Methods for Imbalanced Text Classification

Yazarlar:

Hande Tiryaki | Alper Kürşat Uysal

Yayıncı:

Afyon Kocatepe University

Ana Başlıklar:

Bilgisayar Yazılımı

Dergi:

Afyon Kocatepe Üniversitesi Fen Ve Mühendislik Bilimleri Dergisi Vol.23 Issue 2, pp.370-379

Belge Tipi:

Makale

Yayın Tarihi:

3/2023

Belge Dili:

Türkçe

Document ID (Org No.):

2001288779

pISSN:

2149-3367

Anahtar kelimeler:

Özet:

 The distribution of text data across classes is often imbalanced. This situation has a negative impact on the performance of classifiers in the text classification process. Many studies have been performed on imbalanced text classification. The feature selection stage, which is one of the important stages of the text classification process, is also critical in the imbalanced text classification problem. The effect of feature selection methods on the classification of imbalanced texts has been thoroughly investigated in this study. In this direction, many experiments were carried out with three different classifiers and nine different feature selection methods on two different data sets. In addition, the success of feature selection methods has been observed employing different number of features. Nine different feature selection methods called NDM, DFSS, PFS, POISSON, CHI2, IG, GINI, DFS and MDFS were evaluated. Experimental results obtained with Support Vector Machines SVM, Decision Tree DTREE, and Naïve Bayes MNB classifiers. On the Reuters-21578 dataset, DFS and CHI2 feature selection methods obtained approximately 80 as the highest Macro-F1 score. On the SPAM SMS dataset, DFS feature selection method obtained 95 and CHI2 feature selection method obtained 94 as the highest Macro-F1 score. It is seen that feature selection methods DFS and CHI2 are more successful than the others for imbalanced text classification.

 The distribution of text data across classes is often imbalanced. This situation has a negative impact on the performance of classifiers in the text classification process. Many studies have been performed on imbalanced text classification. The feature selection stage, which is one of the important stages of the text classification process, is also critical in the imbalanced text classification problem. The effect of feature selection methods on the classification of imbalanced texts has been thoroughly investigated in this study. In this direction, many experiments were carried out with three different classifiers and nine different feature selection methods on two different data sets. In addition, the success of feature selection methods has been observed employing different number of features. Nine different feature selection methods called NDM, DFSS, PFS, POISSON, CHI2, IG, GINI, DFS and MDFS were evaluated. Experimental results obtained with Support Vector Machines SVM, Decision Tree DTREE, and Naïve Bayes MNB classifiers. On the Reuters-21578 dataset, DFS and CHI2 feature selection methods obtained approximately 80 as the highest Macro-F1 score. On the SPAM SMS dataset, DFS feature selection method obtained 95 and CHI2 feature selection method obtained 94 as the highest Macro-F1 score. It is seen that feature selection methods DFS and CHI2 are more successful than the others for imbalanced text classification.

 The distribution of text data across classes is often imbalanced. This situation has a negative impact on the performance of classifiers in the text classification process. Many studies have been performed on imbalanced text classification. The feature selection stage, which is one of the important stages of the text classification process, is also critical in the imbalanced text classification problem. The effect of feature selection methods on the classification of imbalanced texts has been thoroughly investigated in this study. In this direction, many experiments were carried out with three different classifiers and nine different feature selection methods on two different data sets. In addition, the success of feature selection methods has been observed employing different number of features. Nine different feature selection methods called NDM, DFSS, PFS, POISSON, CHI2, IG, GINI, DFS and MDFS were evaluated. Experimental results obtained with Support Vector Machines SVM, Decision Tree DTREE, and Naïve Bayes MNB classifiers. On the Reuters-21578 dataset, DFS and CHI2 feature selection methods obtained approximately 80 as the highest Macro-F1 score. On the SPAM SMS dataset, DFS feature selection method obtained 95 and CHI2 feature selection method obtained 94 as the highest Macro-F1 score. It is seen that feature selection methods DFS and CHI2 are more successful than the others for imbalanced text classification.

 Metin verilerinin sınıflar arasında dağılımı genellikle eşit değildir. Bu durum, metin sınıflandırma işleminde sınıflandırıcıların performansına olumsuz yansımaktadır. Dengesiz metin sınıflandırma ile ilgili birçok çalışma yapılmıştır. Metin sınıflandırma işleminin önemli aşamalarından olan öznitelik seçim aşaması, dengesiz metin probleminde de kritik öneme sahiptir. Öznitelik seçme metotlarının dengesiz metinlerin sınıflandırılması üzerindeki etkisi bu çalışmada etraflıca araştırılmıştır. Bu doğrultuda, iki farklı veri seti üzerinde üç farklı sınıflandırıcı ve dokuz farklı öznitelik seçim metodu ile birçok deney yapılmıştır. Ayrıca öznitelik seçim yöntemlerinin başarıları farklı öznitelik sayılarında da gözlemlenmiştir. NDM, DFSS, PFS, POISSON, CHI2, IG, GINI, DFS ve MDFS olarak adlandırılan 9 farklı öznitelik seçim metodu değerlendirilmiştir. Destek Vektör Makinesi SVM, Karar Ağacı DTREE ve Basit Bayes MNB sınıflandırıcıları ile deneysel sonuçlar elde edilmiştir. Reuters-21578 veri setinde DFS ve CHI2 öznitelik seçim yöntemleri Makro-F1 değerlendirme metriği üzerinden yaklaşık en yüksek 80 değerini alırken, SPAM SMS veri setinde, DFS öznitelik seçim yöntemi en yüksek skor olarak 95 ve CHI2 öznitelik seçim yöntemi 94 değerlerini almıştır. Öznitelik seçme metotlarından DFS ve CHI2nin dengesiz metin sınıflandırmada daha başarılı olduğu görülmektedir.

Makale detayları

The Effectiveness of Feature Selection Methods for Imbalanced Text Classification

Yazarlar:

Hande Tiryaki | Alper Kürşat Uysal

Yayıncı:

Afyon Kocatepe University

Ana Başlıklar:

Bilgisayar Yazılımı

Dergi:

Afyon Kocatepe Üniversitesi Fen Ve Mühendislik Bilimleri Dergisi Vol.23 Issue 2, pp.370-379

Belge Tipi:

Makale

Yayın Tarihi:

3/2023

Belge Dili:

Türkçe

Document ID (Org No.):

2001288779

pISSN:

2149-3367

Imbalanced text classification

dimension reduction

feature selection

term selectiondengesiz metin sınıflandırma

boyut indirgeme

öznitelik seçimi

terim seçimi

Sayı

Makale detayları

Yazarlar: Hande Tiryaki | Alper Kürşat Uysal

Yayıncı: Afyon Kocatepe University

Ana Başlıklar: Bilgisayar Yazılımı

Dergi: Afyon Kocatepe Üniversitesi Fen Ve Mühendislik Bilimleri Dergisi Vol.23 Issue 2, pp.370-379

Belge Tipi: Makale

Yayın Tarihi: 3/2023

Belge Dili: Türkçe

Document ID (Org No.): 2001288779

pISSN: 2149-3367

Sayı

Yazarlar:

Hande Tiryaki | Alper Kürşat Uysal

Yayıncı:

Afyon Kocatepe University

Ana Başlıklar:

Bilgisayar Yazılımı

Dergi:

Afyon Kocatepe Üniversitesi Fen Ve Mühendislik Bilimleri Dergisi Vol.23 Issue 2, pp.370-379

Belge Tipi:

Makale

Yayın Tarihi:

3/2023

Belge Dili:

Türkçe

Document ID (Org No.):

2001288779

pISSN:

2149-3367