Automatic kernel clustering with a Multi-Elitist Particle Swarm Optimization Algorithm

Das, Swagatam; Abraham, Ajith; Konar, Amit

doi:10.1016/j.patrec.2007.12.002

Cited by 174 publications

(64 citation statements)

References 46 publications

(50 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Following the study of Iam-On et al (2010), it is interesting observe the performance of this improved framework with the problem of microarray data analysis, with which link-based approach has been successful. Also, new clustering algorithms, e.g., that of Das et al (2008), can be used to form a more accurate cluster ensemble.…”

Section: Discussionmentioning

confidence: 99%

A new link-based method to ensemble clustering and cancer microarray data analysis

Iam-On

Boongoen²,

Kongkotchawan³

2014

IJCI

View full text Add to dashboard Cite

Ensemble clustering or cluster ensembles have been shown to be better than any standard clustering algorithm at improving accuracy. This meta-learning formalism helps users to overcome the dilemma of selecting an appropriate technique and the parameters for that technique, given a set of data. It has proven effective for many problem domains, especially microarray data analysis. Among different state-of-the-art methods, the link-based approach (LCE) recently introduced by Iam-On et al. (2011) provides a highly accurate clustering. This paper presents the improvement of LCE with a new link-based metric being developed and engaged. Additional information that is already available in a network is included in the similarity assessment. As such, this refinement can increase the quality of the measures, hence the resulting cluster decision. The performance of this improved LCE is evaluated on synthetic and UCI benchmark datasets, in comparison with the original and several well-known cluster ensemble techniques. In addition, the application of improved LCE to microarray data analysis is also empirically assessed. The findings suggest that the new model can improve the accuracy of LCE and performs better than the others investigated in this study.

show abstract

Section: Discussionmentioning

confidence: 99%

A new link-based method to ensemble clustering and cancer microarray data analysis

Iam-On

Boongoen²,

Kongkotchawan³

2014

IJCI

View full text Add to dashboard Cite

show abstract

“…Some relevant studies that have explored the problem of clustering using various approaches include evolutionary algorithms such as evolutionary programing [9], particle swarm optimization [10][11][12], ant colony algorithms [13,14], artificial bee colony [15], simulated annealing [16,17] and tabu search [18]. Conversely, there have been many attempts to use GAs to solve clustering applications [7,[19][20][21][22][23][24][25][26][27].…”

Section: Literature Reviewmentioning

confidence: 99%

A Selection Process for Genetic Algorithm Using Clustering Analysis

et al. 2017

View full text Add to dashboard Cite

This article presents a newly proposed selection process for genetic algorithms on a class of unconstrained optimization problems. The k-means genetic algorithm selection process (KGA) is composed of four essential stages: clustering, membership phase, fitness scaling and selection. Inspired from the hypothesis that clustering the population helps to preserve a selection pressure throughout the evolution of the population, a membership probability index is assigned to each individual following the clustering phase. Fitness scaling converts the membership scores in a range suitable for the selection function which selects the parents of the next generation. Two versions of the KGA process are presented: using a fixed number of clusters K (KGA f ) and via an optimal partitioning K opt (KGA o ) determined by two different internal validity indices. The performance of each method is tested on seven benchmark problems.

show abstract

“…Salah satu metode hasil penelitian tersebut adalah menggabungkan swarm intelligence pada kernel clustering. Sebuah penelitian mengusulkan suatu metode gabungan tersebut yaitu Automatic Kernel Clustering with Multi-Elitist Particle Swarm Optimization (AKC-MEPSO) [8]. Penelitian ini menyajikan suatu pengelompokan data dari data set yang kompleks tanpa terlebih dahulu harus memiliki pengetahuan awal mengenai cluster-nya.…”

Section: Pendahuluanunclassified

“…Pengujian dilakukan dengan melakukan koleksi data kemudian dilakukan normalisasi data, pengujian algoritme, evaluasi efisiensi, dan efektivitas pembentukan cluster. Untuk mendapatkan hasil performa yang bagus dari algoritme, dilakukan uji akurasi (8). Hal ini dilakukan dengan menghitung jumlah cluster yang benar dibanding total keseluruhan cluster yang terbentuk.…”

Section: Kernel Gaussianunclassified

Pengelompokan Data Menggunakan Pattern Reduction Enhanced Ant Colony Optimization dan Kernel Clustering

Hidayat¹,

Fatichah

Ginardi

2016

Jurnal Nasional Teknik Elektro dan Teknologi Informasi (JNTETI)

View full text Add to dashboard Cite

Abstract-One method of optimization that can be used for clustering is Ant Colony Optimization (ACO). This method is good in data clustering, but has disadvantage in terms of time and quality or solution convergence. In this study, ACO-based Pattern Reduction Enhanced Ant Colony Optimization (PREACO) method with a gaussian kernel function is proposed. First, it sets up initial solution. Second, the magnitude of pheromone is calculated to find the centroid randomly. With the initialized solution, the weight of the solution is calculated and the center of cluster is revised. The solution will be evaluated through a gaussian kernel functions. Function 'pattern enhanced reduction' is useful to ensure maximum value of pheromone update. Those steps will be conducted repeatedly until the best solution is chosen. Tests are performed on multiple datasets, with three test scenarios. The first test is carried out to get the right combination of parameters. Second, the error rate measurement and similarity data using Sum of Squared Errors is done. Third, level of accuracy of the methods ACO, ACO with the kernel, PREACO, and PREACO with the kernel is compared. The test results show that the proposed method has a higher accuracy rate of 99.8% for synthetic data, 93.8% for wine data than other methods. But it has a lower accuracy by 88.7% compared to the ACO. Intisari- I. PENDAHULUANCluster merupakan pengelompokan dari banyak data dengan berdasarkan pada kesamaan atau perbedaannya. Kesamaan maupun perbedaan itu dilihat dari ciri yang dipunyai data tersebut. Misalnya saja terdapat sebuah data termasuk ke dalam Cluster tertentu dikarenakan memiliki ciri yang sama dengan Cluster itu. Pengklusteran sebuah data sesuai cluster masing-masing dilakukan dengan metode tertentu. Oleh karena itu melakukan cluster terdapat dua jenis yaitu soft clustering dan hard clustering. Maksudnya, sebuah data bisa termasuk ke dalam beberapa cluster. Hal ini disebut dengan soft clustering. Sebaliknya, sebuah data harus dimasukkan ke dalam satu cluster tertentu saja, disebut dengan hard clustering.adalah satu dari beragam metode clustering. Metode tersebut dapat dengan baik melakukan clustering, tetapi masih memiliki permasalahan yaitu pada penentuan jumlah cluster di awal dan juga masih memasukkan sebuah data ke dalam beberapa cluster yang berbeda. Oleh karena itu akan menjadi masalah ketika sebuah data masuk dalam beberapa cluster yang berbeda. Itu artinya data tersebut tidak terdefinisi dengan baik pengklusterannya. Oleh sebab itu, penentuan jumlah cluster di awal diperlukan. Penelitian tentang permasalahan tersebut masih terbuka untuk diteliti. Salah satu metode yang berusaha menyelesaikan permasalahan itu adalah kernel clustering, untuk mendefinisikan dengan baik jumlah cluster, sehingga pada akhirnya sebuah data akan dimasukkan ke dalam sebuah cluster dengan tepat.Cluster analysis atau kernel clustering saat ini menjadi rujukan dalam melakukan pembelajaran tak terawasi pada proses pencarian cluster. Berangkat dari kelemahan ini, telah dilakukan p...

show abstract

Automatic kernel clustering with a Multi-Elitist Particle Swarm Optimization Algorithm

Cited by 174 publications

References 46 publications

A new link-based method to ensemble clustering and cancer microarray data analysis

A new link-based method to ensemble clustering and cancer microarray data analysis

A Selection Process for Genetic Algorithm Using Clustering Analysis

Pengelompokan Data Menggunakan Pattern Reduction Enhanced Ant Colony Optimization dan Kernel Clustering

Contact Info

Product

Resources

About