運用免疫演算法於大數據社群網路最佳群集偵測之研究

近來社會網路分析(social network analysis, SNA)的方法與技術已廣泛的被應用在不同領域問題上作為分析與制訂決策的有效工具，然而對於分析者要怎麼獲得隱藏在網路複雜資料中的社會網路概念，以做出精確的詮釋和分析是相當困難的，特別是對於在大數據的領域問題，對於問題的分析與決策的制訂更是嚴苛的考驗。對於社會網路分析方法對分析者而言，必須經由分析者主觀的認知方式對同一社會網路以不同觀點進行資料不同的子群集組合來推論重要的概念，但此社會網路若存在著子群集的組合非唯一時，表示所探討的潛在子群集組成可能有多重最佳組合，此對分析者而言是一項極具困難的挑戰且極易造成誤判。因此本研究將擴展過去文獻上的研究常受較限於大數據資料之分析和較少以大數據為基礎的社會網路中偵測最適化的子群集，且把偵測子群集的問題融入於具多重最佳解特性的免疫演算法中，而將之轉變為最佳化問題之求解，據此以提升SAN的分析精準與細膩度。

關鍵字

大數據；社會網路分析；子群集偵測；免疫演算法

並列摘要

Recently, the methods and the technology of social network analysis (SNA) have been applied extensively in the different fields, which is as an effective tool, to analyzing and decision-making. However, how to get the concepts of Social network hidden in the complex web is quite difficult for analyzers to make the accurate interpretation and analysis. Especially for the problems in the field of Big Data, analyzing and decision-making are more severe tasks. As far as analyzers are concerned, SNA is supposed to being deduced to become the vital concept, which is based on the analyzers’ subjective cognition in the same Social Network, in the diverse subgroups with the different information and in the different viewpoints. But if the community Social Network is not sole combination, the potential community may have multiple best compositions. To find the multiple combinations of communities is a big challenge for the analyzers, and erroneous judgment may happen easily. Therefore, this study will expand the previous research in literature, which is often limited by the analysis of the Big Data and is seldom based on the Big Data to detect the modest community. Also, we put the problems of community detection into immune algorithm with the character of multiple optimal solutions. It will become the answer to optimization problem. According to the numerical results, it indicates the analysis precision of the proposed approach is better than current net nodes in literature.

並列關鍵字

Big Data ； Social Network Analysis ； Subgroup Identification ； Immune Algorithm

參考文獻

1.Cross, R., and Parker, A (2004). The hidden power of social networks: understanding how work really gets done in organizations, Harvard Business School Publishing, Boston, M.A., USA.

2.Bajaj, A. and Russell, R. (2010). “AWSM: Allocation of workflows utilizing social network metrics,” Decision Support Systems, 50, pp. 191-202.

3.Chen, H., Chung, W., Xu, J., Wang, G., Chau, M., and Qin, Y. (2004). “Crime data mining: a general framework and some examples,” IEEE Computer, 37(4), pp. 50-56.

4.Zhongming Ma a,Olivia R.L., Sheng and Gautam Pant. (2009).“Discovering company revenue relations from news: A network approach”Decision Support Systems 47,pp.408–414

5.Freeman, L.C. (1979). “Centrality in social networks: I. Conceptual clarification,” Social Networks, 1. pp. 215-239.

國際替代計量

運用免疫演算法於大數據社群網路最佳群集偵測之研究

未授權

主題瀏覽