臺灣博碩士論文加值系統

Kohonen的自我組織映射圖在可視化領域中是一個知名的非監督學習演算法。對於自組織映射圖而言如何在現實世界的數據中找出有用的資訊和知識是非常重要的。近年來，自我組織映射圖已經有各式各樣的改進和延伸。例如，延伸自我組織映射圖能夠有效的處理種類型數據和混合型數據。批次延伸自我組織映射圖是批次版本的延伸自我組織映射圖演算法，其執行速度比原先的演算法更快。然而，如果現實世界的數據是高維度並且具有較大的數據量，許多的神經元將會在批次延伸自我組織映射圖訓練過程中動態產生。作為結果而言，訓練時間由於增加的神經元而增加計算量。因此，在此研究中，我們提出了批次的固定大小的映射圖，擴展自我組織映射圖以便提高性能。實驗結果表明，在訓練效率方面所提出的方法性能優於先前的方法。

The Kohonen’s Self-Organizing Map (SOM) is a well-known unsupervised learning algorithm in the visualization field. For the Self-Organizing Map (SOM), how to find out useful information or knowledge in real-world data is very important. In recent years, the SOM has had many variants that are improved and extended to the original model. For instance, the Generalized Self-Organizing Map (GenSOM) is able to process categorical data and mixed-type data as well. The Batch Generalized Self-Organizing Map (BatchGSOM) is batch and dynamic growth version of the GenSOM algorithm, which runs faster than the previous algorithms. However, if the real-world data is high-dimensional and has a large amount of data, many of neurons will be dynamically generated during the training process of BatchGSOM. As a result, the training time will increase due to the large amount of calculation. Therefore, in this study we propose a fixed-sized map version of batch, extended SOM so as to improve the performance. Experimental results indicate that the proposed approach outperforms the previous approaches in terms of training efficiency.

摘要 i
Abstract ii
Table of Contents iii
List of Tables iv
List of Figures v
1. Introduction 1
1.1 Motivation 1
1.2 Objective 3
2. Literature review 4
2.1 Self-organizing map 4
2.2 Distance hierarchy of GenSOM 5
2.3 K-means Clustering 6
2.4 Batch Map algorithm 7
3. Methodology 9
3.1 Initial phase 10
3.2 Training phase 11
3.2.1 Find Neighborhood and Compute Contribution 11
3.2.2 Find the Maximum Branch 11
3.2.3 Find the Centroid that represents weights updating 13
3.3 The example of training process 14
4. Experiments 17
4.1 Datasets 17
4.2 Stu3a9g360 Dataset 18
4.3 Drink9mix v3 Dataset 19
4.4 Drink9mix v4 Dataset 20
4.5 Drink9mix v6 Dataset 21
4.6 Drink8G2C2N Dataset 23
4.7 Result of Projection 24
4.8 Evaluations 26
5. Conclusion and Future work 30
References 31

Alahakoon, D., Halgamuge, S., & Srinivasan, B. (2000). Dynamic self-organizing maps with controlled growth for knowledge discovery. Neural Networks, IEEE Transactions on, 11(3), 601-614.
Długosz, R., Talaśka, T., Pedrycz, W., & Wojtyna, R. (2010). Realization of the conscience mechanism in CMOS implementation of winner-takes-all self-organizing neural networks. IEEE Transactions on Neural Networks, 21(6), 961-971.
Forti & Foresti, G.L. (2006). Growing hierarchical tree SOM: an unsupervised neural network with dynamic topology. Neural Networks (19), 1568–1580.
Hsu, A. L., & Halgamuge, S. K. (2003). Enhancement of topology preservation and hierarchical dynamic self-organising maps for data visualisation. International Journal of Approximate Reasoning, 32(2), 259-279.
Hsu, C.-C. (2006). Generalizing Self-Organizing Map. IEEE Transactions on Neural Networks, 17(2), 294-304.
Hsu, C.-C., Halgamuge, S.K. (2008). Class structure visualization with semi-supervised growing self-organizing maps. Neurocomputing (71), 3124–3130.
Hsu, C.-C., & Su, Y.-W. Categorical K-means Clustering Algorithm (R1). unpublished manuscript
Huang, J. Z., Ng, M. K., Rong, H., & Li, Z. (2005). Automated variable weighting in k-means type clustering. IEEE Transactions on Pattern Analysis and Machine Intelligence 27(5), 657 - 668.
Kim, S. W., & To, T. V. (2013). A Self-Growing and Self-Organizing Batch Map with Automatic Stopping Condition. International Conference on Knowledge and Smart Technology, 21-26.
Kohonen, T. (1998).The self-organizing map. Neurocomputing. (21), 1-6.
Kohonen, T. (2002). How to make large self-organizing maps for nonvectorial data. Neural Networks (15), 945-952.
Kohonen, T. (2013). Essentials of the self-organizing map. Neural Networks (37), 52-65.
Kohonen, T., Hynninen, J., Kangas, J., & Laaksonen, J. (1996). Som pak: The self-organizing map program package. Report A31, Helsinki University of Technology, Laboratory of Computer and Information Science,.
Kohonen, T., Oja, E., Simula, O., Visa, A., & Kangas, K. (1996). Engineering Applications of the Self-Organizing Map. Proceedings of the IEEE 84(10), 1358-1384.
Koskela, M., Laaksonen, J., & Oja, E. (2004). Entropy-based measures for clustering and SOM topology preservation applied to content-based image indexing and retrieval. Paper presented at the Pattern Recognition, 2004. ICPR 2004. Proceedings of the 17th International Conference on.
MAQUEEN, J. B. (1967). Some methods for classification and analysis of multivariate observations. MacQueen, J. B., Proceedings of Fifth Symposium on Math, Statistics, and Probability, 281-297.
Rauber A., Merkl D., & Dittenbach M. (2002). The growing hierarchical self-organizing map: exploratory analysis of high-dimensional data. IEEE Transactions on Neural Networks (13), 1331–1341.
Vesanto, J. (1999). SOM-based data visualization methods. Intelligent Data Analysis, 3, 111-126.
Vesanto, J., Himberg, J., Alhoniemi, E., & Parhankangas, J. (1999). Self-organizing map for data mining in Matlab. Simulation News Europe 25(54).