A pillar algorithm for K-means optimization by distance maximization for initial centroid designation

Barakbah, Ali Ridho; Kiyoki, Yasushi

doi:10.1109/cidm.2009.4938630

Cited by 48 publications

(31 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…al. in [16] proposed a pillar algorithm to solve the initial centroid designation problem, by considering the pixel maximization i.e., select maximum pixel value for centroid. To improve the performance of segmentation and detecting the accurate tumor cell from the MR image, pillar algorithm will be merged with k-means clustering, which is proposed in [17] and also proposed feature extraction with approximate reasoning to calculate the area of tumor based on the number of white pixels in the segmented MR brain image.…”

Section: Relevant Workmentioning

confidence: 99%

Quantitative Analysis of Metastasis Brain Tumor and its Area Estimation in MR Images

Vidyasagar¹,

Bhujangarao²,

Madhu³

2016

IJCA

View full text Add to dashboard Cite

Metastasis brain tumor develops multiple tumors at asymmetrical location of the human brain. MRI Imaging is one of the prudent mechanisms to extract the tumor regions and to map the brain for diagnosing. For the better diagnosis, one must detect the tumor accurately and need to calculate the area and volume of the tumor exactly. Here in this letter, we proposed a novel resolution enhancement technique to improve the quality of MR brain image and optimized hybrid clustering with region split and merge algorithm to detect the tumor cells from the original MR images and to estimate the tumors from different locations. Simulation results show that the proposed algorithm has performed superior to conventional clustering algorithms such as Fuzzy C-means (FCM), K-Means and even optimized pillar algorithm.

show abstract

Section: Relevant Workmentioning

confidence: 99%

Quantitative Analysis of Metastasis Brain Tumor and its Area Estimation in MR Images

Vidyasagar¹,

Bhujangarao²,

Madhu³

2016

IJCA

View full text Add to dashboard Cite

show abstract

“…Finally, K-means is applied to obtain final clusters. Barakbah et al [17] propose a Pillar based approach which generates the position of initial centers by using the farthest accumulated distance between all clusters centers. Celebi [18] investigates the performance of k-means as a colour quantization with different initialization scheme.…”

Section: Literature Reviewmentioning

confidence: 99%

Feature Selection with Efficient Initialization of Clusters Centers for High Dimensional Data Clustering

Rajput

Singh

Bhattacharya

2011

2011 International Conference on Communication Systems and Network Technologies

View full text Add to dashboard Cite

Most of the traditional data clustering algorithms suffer from two main problems (i) the curse of dimensionality and (ii) random initialization of clusters centers which leads to local optimum clustering. In this paper, we propose a technique for selecting most relevant dimensions of data set and efficient initialization of clusters centers. Our proposed technique uses the median absolute deviation (MAD) for selection of relevant dimensions of data set and then it uses most frequent value (MODE) of selected dimensions to determine the initial clusters centers. Finally these initial clusters centers are used in the k-means algorithm for optimum clustering. Empirical results show that the algorithm produces comparatively efficient results. The quality measures also validate good quality of the obtained results.

show abstract

“…Support Vector Decomposition (SVD) for dimension reduction and concept extraction is performed as an initialization followed by a clustering task using K-Means, optimized by centroid initialization namely Pillar Algorithm [15]. A method to measure the term importance from the risk opinion corpus is performed using the widely use TF-IDF, combined with positive-negative polarity measurement using the SentiWordNet 3.0 library [8].…”

Section: Introductionmentioning

confidence: 99%

Cluster Analysis for SME Risk Analysis Documents Based on Pillar K-Means

2016

View full text Add to dashboard Cite

show abstract

A pillar algorithm for K-means optimization by distance maximization for initial centroid designation

Cited by 48 publications

References 9 publications

Quantitative Analysis of Metastasis Brain Tumor and its Area Estimation in MR Images

Quantitative Analysis of Metastasis Brain Tumor and its Area Estimation in MR Images

Feature Selection with Efficient Initialization of Clusters Centers for High Dimensional Data Clustering

Cluster Analysis for SME Risk Analysis Documents Based on Pillar K-Means

Contact Info

Product

Resources

About