Hybrid feature selection based on enhanced genetic algorithm for text categorization

Ghareb, Abdullah Saeed; Bakar, Azuraliza Abu; Hamdan, Abdul Razak

doi:10.1016/j.eswa.2015.12.004

Cited by 193 publications

(80 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A number of works can be traced in recent years addressing the problem of text classification through feature selection. Feature selection algorithms such as chisquare, information gain, and mutual information (Yang and Pedersen., 1997) though seem to be powerful techniques for text data, a number of novel feature selection algorithms based on genetic algorithm (Bharti and Singh., 2016;Ghareb et al, 2016), ant colony optimization (Dadaneh et al, 2016;Moradi and Gholampour., 2016;Uysal., 2016;Meena et al, 2012), Bayesian principle Zhang et al, 2016;Feng et al, 2012;Fenga et al, 2015;Sarkar et al, 2014), clustering of features (Bharti and Singh., 2015), global information gain (Shang et al, 2013), adaptive keyword (Tasci and Gungor., 2013), global ranking (Pinheiro et al, 2012;Pinheiro et al, 2015) are proposed.…”

Section: Related Workmentioning

confidence: 99%

Cluster Based Symbolic Representation for Skewed Text Categorization

Raju

Suhil

Guru

et al. 2017

Communications in Computer and Information Science

View full text Add to dashboard Cite

Abstract. In this work, a problem associated with imbalanced text corpora is addressed. A method of converting an imbalanced text corpus into a balanced one is presented. The presented method employs a clustering algorithm for conversion. Initially to avoid curse of dimensionality, an effective representation scheme based on term class relevancy measure is adapted, which drastically reduces the dimension to the number of classes in the corpus. Subsequently, the samples of larger sized classes are grouped into a number of subclasses of smaller sizes to make the entire corpus balanced. Each subclass is then given a single symbolic vector representation by the use of interval valued features. This symbolic representation in addition to being compact helps in reducing the space requirement and also the classification time. The proposed model has been empirically demonstrated for its superiority on bench marking datasets viz., Reuters 21578 and TDT2. Further, it has been compared against several other existing contemporary models including model based on support vector machine. The comparative analysis indicates that the proposed modeloutperforms the other existing models.

show abstract

Section: Related Workmentioning

confidence: 99%

Cluster Based Symbolic Representation for Skewed Text Categorization

Raju

Suhil

Guru

et al. 2017

Communications in Computer and Information Science

View full text Add to dashboard Cite

show abstract

“…al. [1] proposed Hybrid feature selection based on enhanced genetic algorithm for text categorization. This method uses a hybrid search technique that mixes the advantages of filter feature selection methods with an improved GA (EGA) in a wrapper procedure to handle the high dimensionality of the feature space and enhance categorization performance concurrently.…”

Section: Hybrid Feature Selection Based On Enhanced Genetic Algorithmmentioning

confidence: 99%

“…The data mining [18,3] techniques are mainly used for the extraction of logical patterns from structured database. Text mining [14,1] is similar to data mining, except that data mining tools [15] are designed to handle structured data from databases, but text mining can work with unstructured or semi-structured data sets such as emails, full-text documents. Text mining [8,16] is a knowledge discovery technique that provides computational intelligence.…”

Section: Related Workmentioning

confidence: 99%

Survey of Text Mining Techniques, Challenges and their Applications

Sailaja¹,

Padmasree²,

Mangathayaru³

2016

IJCA

View full text Add to dashboard Cite

In our everyday life communication interaction among people leading to mutual learning and sharing of valuable knowledge, such as chat, messaging, comments, and posts on board etc. Also, social networking websites, search engines sharing huge data texts in websites. The text is nothing but the combination of characters. Therefore, analyzing and extracting information patterns from such data sets are more complex. Several methods have been proposed for analyzing such texts and extracting information.In this paper, we present different text mining techniques to discover various textual patterns from the different sources. This topic is also deals with the areas such as information retrieval, machine learning, statistics, computational data sciences and advanced data mining. We also discuss future challenges of this area using different techniques, particularly rough set based text mining techniques, improvements and research directions in this paper.

show abstract

“…Sebuah penelitian lain menerapkan metode hybrid feature selection berbasis peningkatan GA untuk pengkategorian teks. Pada penelitian ini, teknik pencarian hybrid dikombinasikan menjadi metode filter feature selection tingkat tinggi Enhanced GA (EGA) untuk mengantisipasi dimensi tinggi dari feature space dan meningkatkan kinerja kategori teks secara simultan [16].…”

unclassified

Feature Selection Klasifikasi Kategori Cerita Pendek Menggunakan Naïve Bayes dan Algoritme Genetika

Somantri¹,

Khambali²

2017

Jurnal Nasional Teknik Elektro dan Teknologi Informasi (JNTETI)

View full text Add to dashboard Cite

Abstract-Classification of short stories category based on age of the reader is still difficult. Therefore, a decision support system to classify the short stories category is needed. Naïve Bayes is one of methods suitable for short stories classification. However, Naïve Bayes has flaws in accuracy level, and needs to be optimized. In this paper, Genetic algorithm is proposed to increase the level of accuracy. In this case, genetic algorithm is used for feature selection. The results show an increase in the level of accuracy produced. The accuracy increases from 78,59% to 84,29%. In conclusion, the application of genetic algorithm on Naïve Bayes in classifying the online short stories category can improve the accuracy.Intisari-Penetapan klasifikasi kategori cerpen yang sesuai dengan keinginan, yaitu adanya kesesuaian jenis cerpen yang dibaca dengan usia pembaca, masih sulit dilakukan. Hal ini menjadikan perlu adanya pendukung keputusan untuk dapat mengklasifikasikan kategori cerpen sehingga jenis cerpen tersebut sesuai dengan kategori pembaca. Naïve Bayes merupakan metode yang sesuai untuk digunakan dalam klasifikasi kategori cerpen. Naïve Bayes masih mempunyai kekurangan dalam tingkat akurasi yang dihasilkan, sehingga perlu adanya optimasi. Algoritme genetika sebagai algoritme optimasi merupakan metode yang diusulkan untuk meningkatkan tingkat akurasi yang digunakan. Dalam hal ini, algoritme genetika digunakan untuk feature selection. Hasil yang diperoleh memperlihatkan adanya sebuah peningkatan akurasi, yaitu dari 78,59% menjadi 84,29%. Dengan demikian, dapat disimpulkan bahwa penerapan algoritme genetika pada naïve bayes untuk klasifikasi kategori cerpen online dapat meningkatkan akurasi.Kata Kunci-klasifikasi, kategori cerpen, Naive Bayes, algoritme genetika.I. PENDAHULUAN Indonesia merupakan negara yang memiliki berbagai macam budaya dan kultur yang beragam, seperti halnya dalam penciptaan karya sastra, sehingga tidak heran apabila setiap daerah di Indonesia mempunyai karya sastra yang beragam. Cerpen adalah sebuah karya sastra yang sangat disukai oleh semua orang, terutama di Indonesia. Institusi pendidikan pun pada proses pembelajaran yang dilaksanakan di sekolah memberikan materi yang terkait dengan cerita pendek. Hal inilah yang memberikan motivasi kepada setiap orang untuk dapat membuat sebuah cerpen sesuai dengan keinginan dan inspirasi yang muncul dari penulis. Saat ini penulisan cerpen masih didominasi oleh para remaja dan anak-anak, sehingga tidak heran apabila cerpen yang muncul sekarang ini sudah banyak dari lintas generasi baik itu ditulis lewat buku, majalah, blog pribadi, maupun media online serta situs website yang menyediakan cerpen online yang menyajikan kumpulan cerpen yang dikirim oleh penulisnya.Sebuah karya sastra mempunyai dua unsur pokok di dalamnya. Yang pertama adalah unsur intrinsik, yaitu unsur dalam sastra yang ikut memengaruhi terciptanya karya sastra yang terdiri atas tema dan amanat, alur cerita, penokohan atau perwatakan, latar, sudut pandang, dialog, dan gaya bahasa. Sedangk...

show abstract

Hybrid feature selection based on enhanced genetic algorithm for text categorization

Cited by 193 publications

References 27 publications

Cluster Based Symbolic Representation for Skewed Text Categorization

Cluster Based Symbolic Representation for Skewed Text Categorization

Survey of Text Mining Techniques, Challenges and their Applications

Feature Selection Klasifikasi Kategori Cerita Pendek Menggunakan Naïve Bayes dan Algoritme Genetika

Contact Info

Product

Resources

About