碩士 / 國立中興大學 / 資訊科學與工程學系所 / 107 / Since many successful results in recent years show that it is possible to find out valuable information from daily data. Human can make more appropriate decisions by these valuable information. To develop a “good” data analysis system for data deluge has become a popular research topic. How to analyze such data has been a promising research in data mining. Data clustering is a representative research topic because its solution can be used to classify the unknown data without prior knowledge. Several recent studies attempted to use metaheuristic algorithms to solve clustering problems, and most of them provide a high-quality result than traditional clustering algorithms. In this paper, we present a high performance clustering algorithm based on coral reefs optimization with substrate layers (CRO-SL). To reduce the computation time of the proposed algorithm, we also have implemented it on Apache Spark. In experimental results, we compare the proposed algorithm with k-means algorithm, genetic k-means algorithm (GKA), particle swarm optimization (PSO), and simple CRO algorithm in terms of the sum of squared errors (SSE). The simulation results show that the proposed algorithm can reduce the computation time significantly and also can provide a better clustering result than the other clustering algorithms.
| Identifer | oai:union.ndltd.org:TW/107NCHU5394010 |
| Date | January 2018 |
| Creators | Yi-Chung Wang, 王翊仲 |
| Contributors | 蔡崇煒 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 57 |
Page generated in 0.0017 seconds