對K均值分群估計潛在群體程序作平行運算

Parallel Computing for K-means Clustering on Estimating Underlying Latent Classes

林吟玲(Lin Yin-Ling)

指導教授 ： 黃冠華

國立交通大學/理學院/統計學研究所/碩士(2010年)

https://doi.org/10.6842/NCTU.2010.00273

摘要

本論文主要目的是對k-means分群方法估計潛在群體的計算過程作平行運算，透過OpenMP與MPI平行運算，將updated k-means與non-updated k-means兩種不同k-means分群方法作平行，使得程式計算時間縮短，並且在個人電腦、國家高速電腦中心與Amazon EC2三種不同電腦環境上運作，觀測他們的平行效率。除此之外，利用乳癌的微陣列為例，作更詳細的說明。在乳癌資料的例子中，兩種k-means分群方法都達到縮短運算時間的效果！

關鍵字

OpenMP ； MPI ； 平行運算

並列摘要

The main purpose of the study is to perform parallel computing for k-means clustering on estimating the underlying latent class process. OpenMP and MPI parallel computing make computing time shorter for updated and non-updated k-means clustering method. We compare the parallel efficiency of OpenMP and MPI in the personal computers, the national center for high-performance computing and the Amazon EC2 environment. Besides, the breast cancer microarray data are used for illustration. The results display that parallel computing can reduce the computation time in all three computing environments.

並列關鍵字

OpenMP ； MPI ； Parallel Cmputing

參考文獻

Amazon web services

Bandeen-Roche, K., Miglioretti, D. L., Zeger, S. L., & Rathouz, P.J. (1997). Latent variable regression for multiple outcomes. Journal of the American Statistical Association, 92 , 1375-1386.

Goodman, L. A. (1974). Exploratory latent structure analysis using both identifiable and unidentifiable models. Biometrika, 61 , 215-231.

Huang, G. H., & Bandeen-Roche, K. (2004). Building an identifiable latent class model with covariate effects on underlying and measured variables. Psychometrika, 69 , 5-32.

Kraj, P., Sharma, A., Garge, N., Podolsky, R., & McIndoe, R. A. (2008). ParaKMeans: Implementation of a parallelized K-means algorithm suitable for general laboratory use. BMC Bioinformatics, 9, 200.