LDA-Based Clustering Algorithm and Its Application to an Unsupervised Feature Extraction

Li, Cheng‐Hsuan; Kuo, Bor‐Chen; Lin, Chin‐Teng

doi:10.1109/tfuzz.2010.2089631

Cited by 59 publications

(27 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In the section, an unsupervised LDA using the concept of membership values is described [3]. Let and the within scatter matrix UFLDA w S as follows: represents the total mean.…”

Section: Unsupervised Version Of Linear Discriminant Analysis (Uflda)mentioning

confidence: 99%

“…Let and the within scatter matrix UFLDA w S as follows: represents the total mean. It is easy to observe that the scatter matrices of LDA is a special case of the scatter matrices of UFLDA [3], and UFLDA outperforms principal component analysis (PCA) and independent component analysis (ICA).…”

Section: Unsupervised Version Of Linear Discriminant Analysis (Uflda)mentioning

confidence: 99%

“…Hence, we want to develop a both spectral and spatial information based LDA. An unsupervised version of linear discriminant analysis (UFLDA) based on the membership values is proposed [3].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Hyperspectral image classification using spectral and spatial information based linear discriminant analysis

Chu

Kuo

et al. 2011

2011 IEEE International Geoscience and Remote Sensing Symposium

Self Cite

View full text Add to dashboard Cite

Feature extraction plays an essential role in Hyperspectral image classification. Linear discriminant analysis (LDA) is a commonly used feature extraction (FE) method to resolve the Hughes phenomenon for classification. The Hughes phenomenon (also called the curse of dimensionality) is often encountered in classification when the dimensionality of the space grows and the size of the training set is fixed, especially in the small sampling size problem. Recent studies show that the spatial information can greatly improve the classification performance. Hence, for hyperspectral image classification, it is not only necessary to use the available spectral information but also to exploit the spatial information. In this paper, spatial information is acquired by the concept of the Markov random field (MRF), and this spatial information is used to form the membership values of every pixel in the hyperspectral image. The experimental results on two hyperspectral images, the Washington DC Mall and the Indian Pine Site, show that the proposed method can yield a better classification performance than LDA in the small sampling size problem.

show abstract

“…In the section, an unsupervised LDA using the concept of membership values is described [3]. Let and the within scatter matrix UFLDA w S as follows: represents the total mean.…”

Section: Unsupervised Version Of Linear Discriminant Analysis (Uflda)mentioning

confidence: 99%

Section: Unsupervised Version Of Linear Discriminant Analysis (Uflda)mentioning

confidence: 99%

See 1 more Smart Citation

Hyperspectral image classification using spectral and spatial information based linear discriminant analysis

Chu

Kuo

et al. 2011

2011 IEEE International Geoscience and Remote Sensing Symposium

Self Cite

View full text Add to dashboard Cite

show abstract

“…Research shows that only a few existing clustering algorithms are able to achieve good adaptivity for reliable separation of nucleus and cytoplasm (Mohapatra et al, 2012a;Mohapatra et al, 2012b;Mohapatra et al, 2014). Therefore, the existing methods like k-means (Fatma and Sharma, 2014), FCS1 (Li et al, 2011), FCS2 (Wu et al, 2005), FCM (Bezdek et al, 1984), LDA (Li et al, 2011) are not reliable because of the limitations of the clustering algorithms (Kuo and Landgrebe, 2004). This paper aims to overcome the above mentioned challenges.…”

Section: Introductionmentioning

confidence: 99%

Detection of acute lymphoblastic leukemia using microscopic images of blood

Bhukya

2017

Int. j. adv. appl. sci.

View full text Add to dashboard Cite

“…Penyelesaian dicoba dilakukan dengan beberapa cara. Beberapa peneliti melakukan pemilihan fitur [12] dan pengekstrakan fitur [13] untuk peningkatan kinerja klasifikasi ini. Selain itu, beberapa penelitian melakukan penerapan meta-algoritme untuk peningkatan kinerja ini.…”

unclassified

Meta-Algoritme Adaptive Boosting untuk Meningkatkan Kinerja Metode Klasifikasi pada Prestasi Belajar Mahasiswa

Yamasari¹,

Nugroho²,

Suyatno³

et al. 2017

Jurnal Nasional Teknik Elektro dan Teknologi Informasi (JNTETI)

View full text Add to dashboard Cite

Abstract-Determining the right class on student achievement is important in an evaluation process, because placing students in the right class helps lecturer in reflecting the successfullness of learning process. This problem relates to the performance of classification method which is measured by the classifier metrics. High performance is indicated by the optimality of these classifier's metrics. Besides, meta-algorithm adaptive boosting has been proven to be able to improve the performance of classifier in various fields. Therefore, this paper employs adaptive boosting to reduce the number of incorrect student placement in a class. The experimental results of implementing adaptive boosting in classifying student achievement shows that there is an increase of performance of all classification metrics, i.e., Kappa, Precision, Recall, F-Measure, ROC, and MAE. In terms of accuracy, J-48 is able to rise about 3.09%, which means this method reduces three misclassified students. Additionally, decisionStump increases 12.37% of accuracy. This also means this method is able to decrease 12 misclassified students. Finally, Simple Cart reaches the highest accuracy of about 23.71%, while the number of misclassified students is reduced to 24 students. However, there is no improvement in Random Forest method by using this adaptive boosting. Intisari-Penetapan kelas yang tepat pada prestasi belajar mahasiswa adalah suatu hal yang penting dalam proses evaluasi.Hal ini disebabkan keberadaan mahasiswa pada kelas yang tepat dapat membantu dosen untuk merefleksikan berhasil tidaknya proses belajar mengajar yang telah dilakukan. Masalah ketepatan kelas ini berkaitan dengan kinerja metode klasifikasi yang diukur dengan ukuran-ukuran classifier. Kinerja yang tinggi diindikasikan dengan optimalnya ukuran-ukuran classifier tersebut. Di sisi lain, meta-algoritme adaptive boosting telah terbukti mampu meningkatkan kinerja classifier pada berbagai bidang. Oleh karena itu, makalah ini menerapkan adaptive boosting untuk meningkatkan kinerja classifier agar keberadaan mahasiswa pada kelas yang tidak sesuai bisa direduksi. Hasil uji coba penerapan adaptive boosting pada klasifikasi prestasi belajar mahasiswa ini menunjukkan peningkatan kinerja pada semua ukuran classifier, Penggunaan sistem e-learning dan sistem evaluasi ini akan menghasilkan tumpukan data pada tempat penyimpanan. Tumpukan data ini dapat digunakan untuk pengoptimalan pada proses-proses tersebut dengan penerapan data mining. Hal ini dapat dilakukan karena penerapan ini menghasilkan pengetahuan yang sangat berguna. Pengetahuan ini diperoleh dari beberapa aktivitas dari data mining, antara lain klasifikasi, klaster, dan analisis asosiasi. Beberapa penelitian tentang klasifikasi antara lain prediksi mahasiswa nonaktif dengan menggunakan beberapa metode klasifikasi [6], klasifikasi untuk memprediksi prestasi belajar siswa dengan menggunakan Naïve Bayesian [7], dan klasifikasi dengan decision tree untuk mengetahui hubungan kinerja akademik dan bidang minat [8]. Untuk penelitian pen...

show abstract

LDA-Based Clustering Algorithm and Its Application to an Unsupervised Feature Extraction

Cited by 59 publications

References 35 publications

Hyperspectral image classification using spectral and spatial information based linear discriminant analysis

Hyperspectral image classification using spectral and spatial information based linear discriminant analysis

Detection of acute lymphoblastic leukemia using microscopic images of blood

Meta-Algoritme Adaptive Boosting untuk Meningkatkan Kinerja Metode Klasifikasi pada Prestasi Belajar Mahasiswa

Contact Info

Product

Resources

About