請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/50346完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 胡明哲 | |
| dc.contributor.author | Wen-Ning Liang | en |
| dc.contributor.author | 梁文寧 | zh_TW |
| dc.date.accessioned | 2021-06-15T12:37:09Z | - |
| dc.date.available | 2016-09-13 | |
| dc.date.copyright | 2016-09-13 | |
| dc.date.issued | 2016 | |
| dc.date.submitted | 2016-07-29 | |
| dc.identifier.citation | [1]. 邱國益，2008，「流形學習之應用的概觀研究」，逢甲大學應用數學學系碩士班碩士論文。
[2]. Pearson, K. (1901). LIII. On lines and planes of closest fit to systems of points in space. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 2(11), 559-572. [3]. Eckart, C., & Young, G. (1936). The approximation of one matrix by another of lower rank. Psychometrika, 1(3), 211-218. [4]. LJP, P. E., & Van Den, H. H. (2007). Dimensionality Reduction: A Comparative Review. Tech. Rrep. [5]. Turk, M., & Pentland, A. (1991). Eigenfaces for recognition. Journal of cognitive neuroscience, 3(1), 71-86. [6]. Tenenbaum, J. B., De Silva, V., & Langford, J. C. (2000). A global geometric framework for nonlinear dimensionality reduction. science, 290(5500), 2319-2323. [7]. Diamantaras, K. I., & Kung, S. Y. (1996). Principal component neural networks: theory and applications. John Wiley & Sons, Inc.. [8]. Scholkopf, B., Smola, A., & Muller, K. R. (1997). Kernel principal component analysis. In Artificial Neural Networks—ICANN'97 (pp. 583-588). Springer Berlin Heidelberg. [9]. Burges, C. J. (1998). A tutorial on support vector machines for pattern recognition. Data mining and knowledge discovery, 2(2), 121-167. [10]. Saul, L. K., & Roweis, S. T. (2000). An introduction to locally linear embedding. unpublished. Available at: http://www. cs. toronto. edu/~ roweis/lle/publications. html. [11]. Roweis, S. T., & Saul, L. K. (2000). Nonlinear dimensionality reduction by locally linear embedding. Science, 290(5500), 2323-2326. [12]. Gastner, M. T., & Newman, M. E. (2004). Diffusion-based method for producing density-equalizing maps. Proceedings of the National Academy of Sciences of the United States of America, 101(20), 7499-7504. [13]. Seung, H. S., & Lee, D. D. (2000). The manifold ways of perception. Science, 290(5500), 2268-2269. [14]. Cayton, L. (2005). Algorithms for manifold learning. Univ. of California at San Diego Tech. Rep, 1-17. [15]. Vellido, A., Garcia, D. L., & Nebot, A. (2013). Cartogram visualization for nonlinear manifold learning models. Data Mining and Knowledge Discovery, 27(1), 22-54. [16]. Patwari, N., Hero III, A. O., & Pacholski, A. (2005, August). Manifold learning visualization of network traffic data. In Proceedings of the 2005 ACM SIGCOMM workshop on Mining network data (pp. 191-196). ACM. [17]. Burges, C. J. (2010). Dimension reduction: A guided tour. Now Publishers Inc. [18]. Kruskal, J. B. (1964). Multidimensional scaling by optimizing goodness of fit to a nonmetric hypothesis. Psychometrika, 29(1), 1-27. [19]. Kruskal, J. B., & Wish, M. (1978). Multidimensional scaling (Vol. 11). Sage. | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/50346 | - |
| dc.description.abstract | 本研究主要目的是要呈現一個新的流形分析降維方法，透過分析得到原始地圖的每一點之間的距離矩陣後，誤差最小法是透過最小化變形地圖的與原始地圖的距離矩陣誤差，在低維空間去建構新的變形座標矩陣，若是誤差趨近於零，新的座標矩陣就能完整保留原始座標矩陣的距離特性。
本研究使用了台北市做案例分析，由於台北市本身豐富的多樣化地理環境與密集的城市建築，點與點之間路徑的距離不可能皆為直線距離，透過原始座標系統與交通路網建構出一個台北市的距離矩陣，透過MDS與誤差最小化方法分別作出新的變形變量圖，而這個變量圖就是以距離為主要變數將台北市的每個座標點重新分配位置，希望能在維持台北市地圖結構與維持距離屬性之間取得平衡。 | zh_TW |
| dc.description.abstract | The research presents the new method for manifold analysis. Error Minimizing method is a dimensional reduction method that find the new node coordinate by seeking minimal error. the error is equal to the square of new distance over the square of original distance. The new distance will be accordant with original distance when error is close to 0, therefore the new node coordinate in new 2-D space maintain the characteristic.
As a research case, Taipei city has diverse geographical environment and the great dense of buildings, therefore the real path between two nodes can not be a straight line. The research is committed to develop a cartogram which the distance between two nodes will change according to the real distance constructed by traffic route for automobile. obtain the balance of that maintain the structure of the map and maintain the characteristic of distance. | en |
| dc.description.provenance | Made available in DSpace on 2021-06-15T12:37:09Z (GMT). No. of bitstreams: 1 ntu-105-R03622027-1.pdf: 4371193 bytes, checksum: 6c4aa45b424c3efe4e7c07223ff7e08a (MD5) Previous issue date: 2016 | en |
| dc.description.tableofcontents | I. Introduction 7
1.1 General Background Information 7 II. Literature Reviews 8 III. Research Motivation and Purpose 9 3.1 Motivation and Purpose 9 3.2 Research Framework 10 3.2.1 Research Area 10 3.2.2 Area Characteristic 13 IV. Methodology 14 4.1 Manifold learning 14 4.1.1 Steps 16 4.2 Isomap 16 4.2.1 the shortest path and geodesic distance 17 4.2.2 Choose neighborhood point 17 4.3 Multidimensional scaling (MDS) 18 4.3.1 Principle of MDS (Multidimensional scaling) 18 4.4 Error Minimize Method 21 4.4.1 Steps of Border Contracted Fix Method (only lower half) 21 4.4.2 Steps of Border Fix Method 23 4.4.3 Steps of Center of District Fix Method 25 V. Case study 27 5.1 MDS (Multidimensional scaling) 27 5.1.1 Result 27 5.2 Border Contracted Fix Method (only lower half) 29 5.2.1 Result 29 5.3 Borden Fix Method 30 5.3.1 Result 30 5.4 Center of District Fix Method 41 5.4.1 result 41 VI. Discussion 53 6.1 Comparison of four method 53 6.2 Error of each method 55 VII. Future develop 56 7.1 Suggestion for methods 56 7.1.1 MDS 56 7.1.2 Contracted Border Fix method 57 7.1.3 Border Fix method 57 7.1.4 Center of District Fix method 57 7.2 Application of transformed map 58 7.3 Flexible development 58 VIII. Reference 59 | |
| dc.language.iso | en | |
| dc.title | 誤差最小降維模型於流形之分析 | zh_TW |
| dc.title | Error Minimizing Dimensional Reduction based on Manifold Analysis | en |
| dc.type | Thesis | |
| dc.date.schoolyear | 104-2 | |
| dc.description.degree | 碩士 | |
| dc.contributor.oralexamcommittee | 童慶斌,溫在弘,許少瑜,余化龍 | |
| dc.subject.keyword | 流形分析,誤差最小法,降維, | zh_TW |
| dc.subject.keyword | Manifold Learning,Error Minimizing,Dimensional Reduction, | en |
| dc.relation.page | 59 | |
| dc.identifier.doi | 10.6342/NTU201601542 | |
| dc.rights.note | 有償授權 | |
| dc.date.accepted | 2016-07-30 | |
| dc.contributor.author-college | 生物資源暨農學院 | zh_TW |
| dc.contributor.author-dept | 生物環境系統工程學研究所 | zh_TW |
| 顯示於系所單位： | 生物環境系統工程學系 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-105-1.pdf 目前未授權公開取用 | 4.27 MB | Adobe PDF |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。