A.1D-C: A novel fast automatic heuristic to handle large-scale one-dimensional clustering |
| |
| Affiliation: | 1. Department of Computational Intelligence, Faculty of Computer Science and Management Wroclaw, University of Science and Technology, Wybrzeze Wyspianskiego 27, 50-370 Wroclaw, Poland;2. Department of Systems and Computer Networks, Faculty of Electronics, Wroclaw University of Science and Technology, Wybrzeze Wyspianskiego 27, 50-370 Wroclaw, Poland;1. School of Data and Computer Science, Sun Yat-sen University, Guangzhou, China;2. School of Computer Science and Engineering, South China University of Technology, Guangzhou, China;3. Guilin University of Electronic Technology, Guilin, China;4. School of Engineering, University of Glasgow, Glasgow, UK;1. Department of Mathematics, Faculty of Arts and Sciences, Eastern Mediterranean University, Famagusta, North Cyprus, Mersin-10, Turkey;2. Department of Computer Science and Information Technology, Faculty of Electrical Engineering and Information Technology, University of Oradea, Oradea, Romania;1. School of Computer Science, Laboratory of Cognitive Modeling and Algorithms, Shanghai Key Laboratory of Data Science, Fudan University, Shanghai, China;2. School of Computer Science and Technology, China University of Mining and Technology, Xuzhou, China;1. Faculty of Computing, Universiti Teknologi Malaysia, 81310 Skudai, Johor, Malaysia;2. Department of Computer Engineering, Hashtgerd Branch, Islamic Azad University, Alborz, Iran |
| |
| Abstract: | The one-dimensional clustering aims to group real-values of an input array into identified number of clusters. Some of the current algorithms, such as the k-means, need the number of clusters in advance, and use a goal function based on minimizing the sum of squared Euclidean distances to the mean of each group. This paper shows why this goal function is not efficient, even for one-dimensional case, then proposes an O (n × log n) efficient algorithm for the one-dimensional clustering purposes. The proposed algorithm can automatically detect the number of clusters. The performance of the proposed algorithm is approved across several experiments. In addition, results of experiments show why the goal function used in some current algorithms like the k-means is not suitable for the one-dimensional clustering. |
| |
| Keywords: | A 1D-C One-dimensional Number of clusters Array of integers New heuristic |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
