本研究以纖毛蟲為研究對象，針對餵食時其移動軌跡，以推估、判別出不同環境下的移動行為特徵。主要是研究四種不同食物濃度條件的資料，先討論單一指標之間的差異；再討論綜合多種指標來判斷、區分其差異性；最後整合資料，並透過不同的資料分析技術尋找出移動行為的特徵，從資料中尋找出適合的資訊與知識。
本研究所嘗試的分析技術之中，以決策樹的結果最適合做為預測，並也具有一定的可信度。若以生物能量上的意義來看，分析結果則與最適捕食的理論(optimal foraging theory)相似，並從不同食物(能量)條件之下的結果得知，纖毛蟲的移動行為相似於最適捕食的理論。
不論是利用統計或資料探勘的分析，不同食物濃度之間的差異，都是將低食物濃度與極高食物濃度的資料劃分在一起，而中食物濃度與高食物濃度則較為相近，且在分析過程中所產生的決策樹分類法則，可提供後續研究區分環境資料時的依據。

It is a research of the move trajectory of the ciliates while feeding the food, in order to estimate, differentiate from the movement behavior under different environments. First, discuss the differently distinguish with the single indicator. Second, discuss with integrate four kinds of indicator whether can distinguish differently. Finally, combine the indicator data and through different analysis technology look out the features of movement behavior, expect to be able to look out suitable information and knowledge from the indicator data.
After deal with analytical technology, the result of decision tree is most suitable for predicted and have credibilities. If according to energy of biological, the analysis result is similar to optimal foraging theory. And learn from result under different condition, the movement behavior of the ciliates similar to the optimal foraging theory.
In the matter of the result of analysis technology, data of the density of low food similar to data of the density of extremely high food. Besides, data of medium food and high food are analogous. The rule of decision tree can distinguish the density of different food, and can offer follow-up study to distinguish the environmental conditions. Those models are evaluated by predicting accuracies, and rules extracted from decision tree models are also of great help to prediction as well.

圖目錄 VII
表目錄 IX
第一章 緒論 1
1.1 研究動機與目的 2
1.2 資料說明 3
1.3 研究流程 3
第二章 文獻回顧 5
2.1 文獻指標 5
2.1.1 布朗運動理論 5
2.1.2 碎形理論 7
2.1.3 NGDR 9
2.2 假設檢定 10
2.3 資料分析技術 12
2.3.1 分類分析 12
2.3.2 關連法則分析 14
2.3.3 分群分析 15
第三章 研究方法 17
3.1 布朗運動理論 17
3.2 碎形理論 18
3.3 NGDR 23
3.4 轉彎程度指標 23
3.5 假設檢定 25
3.6 分類分析 27
3.7 關連法則分析 31
3.8 分群分析 34
第四章 結果分析 38
4.1 指標計算結果 38
4.2 假設檢定結果 48
4.3 資料探勘結果 57
4.4 分析結果總結 64
第五章 結論與建議 67
5.1 研究結論 67
5.2 未來研究建議 69
參考文獻 70
附錄 74

[1] Agrawal R. and Srikant R. (1994), Fast algorithms for mining associations rules, In Proc. VLDB Conference, pp.487-499.
[2] Alados C. L., Escos J. M. and Emlen J. M. (1996), Fractal structure of sequential behaviour patterns: an indicator of stress, Animal behavior, vol.51, pp.437-443.
[3] Alvarez M. C. and Fuiman L. A. (2003), Exposure to atrazine at environmentally realistic levels affects survival potential of a marine fish larvae, University of Texas at Austin, Marine Science Institute, Port Aransas, TX.
[4] Alfaro A. C., Jeffs A. G. and Creese R. G. (2004), Bottom-drifting algal/ mussel spat associations along a sandy coastal region in northern New Zealand, Aquaculture, vol.241, pp.269-290.
[5] Antony B., Brian D. H., David C. W., Martyn G. F. and Philip P. (2004), Biological data mining with neural networks: implementation and application of a flexible decision tree extraction algorithm to genomic problem domains, Neurocomputing, vol.57, pp.275-293.
[6] Andre W. V. and Thomas K. (2006), Plankton motility patterns and encounter rates, Oecologia, vol.48, pp.538-546.
[7] Buskey E. J. (1984), Swimming pattern as an indicator of the roles of copepod sensory systems in the recognition of food, Marine Biology, vol.79, pp.165-175.
[8] Bourke P. (1993), Fractal Dimension Calculator User Manual, New York：Macintosh Application.
[9] Bovill C. (1996), Fractal Geometry in Architecture and Design, School of Architecture, University of Maryland, Boston, pp.1-8.
[10] Cheng Y. C., Lee P. J. and Lee T. Y. (1999), Self-similarity dimensions of the Taiwan Island landscape, Computer and Geosciences, vol.25, pp.1043-1050.
[11] Chae Y. M., Kim H. S., Tark K. C., Park H. J. and Ho S. H. (2003), Analysis of healthcare quality indicator using data mining and decision support system, Expert Systems with Applications, vol.24, pp.167-172.
[12] Cox B., Kislinger T. and Emili A. (2005), Integrating gene and protein expression data: pattern analysis and profile mining, Methods, vol.35, pp.303-314.
[13] Chang Y. C., Lai P. C. and Lee M. T. (2007), An integrated approach for operational knowledge acquisition of refuse incinerators, Expert Systems with Applications, vol.33, pp.413-419.
[14] Erlandsson J. and Kostylev V. (1995), Trail following, speed and fractal dimension of movement in a marine prosobranch, Littorina littorea, during a mating and a non-mating season, Marine Biology, vol.122, pp.87-94.
[15] Gupta H. M. and Campanha J. R. (2002), Tsallis statistics andgrad ually truncated Levy flight-distribution of an economical index, Physica A, vol.309, pp. 381-387.
[16] Gee L., Abbott J., Hart A., Conway S. P., Etherington C. and Webb A. K. (2005), Associations between clinical variables and quality of life in adults with cystic fibrosis, Journal of Cystic Fibrosis, vol.4, pp.59-66.
[17] Haslett J. R. (1994), Community structure and the fractal dimensions of mountain habitats, Biological theory, vol.167, pp.407-411.
[18] Jourdam L., Dhaenens C., Talbi E. G. and Gallina S. (2002), A data mining approach to discover genetic and environmental factors involved in multifactorial diseases, Knoeledge-Based Systems, vol.15, pp.235-242.
[19] Mandelbrot B. B. (1967), How Long is the Coast of Britain? Statistical Self-Similarity and Fractal Dimension, Science, vol.155, pp.636-638.
[20] Maddouri M. and Elloumi M. (2002), A data mining approach based on machine learning techniques to classify biological sequences, Knowledge-Based Systems, vol.15, pp.217-223.
[21] Mullins G. L., Aldridge R. J. and Siveter D. J. (2004), Microplankton associations, biofacies and palaeoenvironment of the type lower Ludlow Series, Silurian, Review of Palaeobotany and Palynology, vol.130, pp.163-194.
[22] Malone J., McGarry K. and Bowerman C. (2006), Automated trend analysis of proteomics data using an intelligent data mining architecture, Expert Systems with Applications, vol.30, pp.24-33.
[23] Mezzasalma S. A. (2007), The special theory of Brownian relativity: Equivalence principle for dynamic and static random paths and uncertainty relation for diffusion, Journal of Colloid and Interface Science, vol.307, pp.386-397.
[24] Olsen M. G. and Adrian R. J. (2000), Brownian motion and correlation in particle image velocimetry, Optics and Laser Technology, vol.32, pp. 621-627.
[25] Pan F., Wang B., Hu X. and Perrizo W. (2004), Comprehensive vertical sample-based KNN/LSVM classification for gene expression analysis, Journal of Biomedical Informatics, vol.37, pp.240-248.
[26] Pasternak Z., Blasius B., Abelson A. and Achituv Y. (2005), Host-finding behaviour and navigation capabilities of symbiotic zooxanthellae, Coral Reefs, vol.25, pp.201-207.
[27] Sommer U., Meusel B. and Stielau C. (1999), An experimental analysis of the importance of body-size in the seastar-mussel predatorprey relationship, Acta Oecologica, vol.20, pp.81-86.
[28] Seuront L., Hwang J. S., Tseng L. C., Schmitt F. G., Souissi S. and Wong C. K. (2004), Individual variability in the swimming behavior of the sub-tropical copepod Oncaea venusta (Copepoda: Poecilostomatoida), Marine Ecology Progress Series, vol.283, pp.199-217.
[29] Su F., Zhou C., Lyne V., Du Y. and Shi W. (2004), A data-mining approach to determine the spatio-temporal relationship between environmental factors and fish distribution, Ecological Modelling, vol.174, pp.421-431.
[30] Sara M. L. and Richard G. V. (2004), Behavioral Responses of Newly Hatched Zebrafish (Danio rerio) to Amino Acid Chemostimulants, Chemosensitivity of Recently Hatched Zebrafish, Senses 29, pp.93-100.
[31] Schmitt F. G., Seuront L., Hwang J. S., Souissi S. and Tseng L. C. (2006), Scaling of swimming sequences in copepod behavior: Data analysis and simulation, Physical A, vol.364, pp.287-296.
[32] Ueda T., Koya S. and Maruyama Y. K. (1999), Dynamic patterns in the locomotion and feeding behaviors by the placozoan Trichoplax adhaerence, BioSystems, vol.54, pp.65-70.
[33] Yang J. S., Chae S. B., Jung W. S. and Moon H. T. (2006), Microscopic spin model for the dynamics of the return distribution of the Korean stock market index, Physica A, vol.363, pp.377-382.
[34] Yen J. (1988), Directionality and swimming speeds in predator-prey and male-female interactions of Euchaeta rimana, a subtropical marine copepod, Bulletin of marine science, No.3, vol.43, pp.395-403.
[35] Zhang X., Johnson S. N., Crawford J. W., Gregory P. J. and Young I. M. (2007), A general random walk model for the leptokurtic distribution of organism movement-Theory and application, ecological modeling, vol.200, pp.79-88.
[36] 李松、方金釧(1990)，中國海洋浮游橈足類幼體，海洋出版社，北京。
[37] 張志三(1993)，漫談分形，湖南教育出版社，北京。
[38] 黃將修與Dahms H. U.(2006)，高速攝影光學科技在浮游動物行為與生態研究上之應用，海洋及水下科技，第16卷第2期，第36-39頁。
[39] 李灝銘(1996)，以聚類分析解析北臺灣酸沈降特性之初步研究，國立中央大學環境工程研究所，桃園，http://aerosol.atm.ncu.edu.tw/information/paper/84/2.pdf。
[40] 王素芬(1998)，地理資訊系統和碎形維度於森林地景空間變化上之應用，國立台灣大學森林學研究所碩士論文，台北。
[41] 簡淑君(2000)，台南七股潟湖區及附近海域浮游性仔稚魚種類組成及其與環境因子相關性之研究，國立中山大學海洋資源學系研究所碩士論文，高雄，第1-6頁。
[42] 黃祥豪(2002)，高屏海域浮游橈足類之時空分佈及攝食研究，國立中山大學海洋資源學系研究所碩士論文，高雄，第1-6頁。
[43] 胡英(1999)，物理化學，高等教育出版社，上冊第四版。
[44] 陳銘宗(2002)，組合式關連法則應用於缺值問題之研究，朝陽科技大學資訊管理學研究所碩士論文，台中，第6-21頁。
[45] 李育嘉(1985)，漫談布朗運動，數學傳播，第9卷第3期，第22-28頁。
[46] 游文淮(2003)，以離散元素法模擬膠粒粒間作用與布朗運動之研究，國立台灣大學土木工程學研究所碩士論文，台北，第17-22頁。
[47] 林光彥、施博偉、黃仁佐和張仁宗(2004)，布朗運動量測之隨機分析與波茲曼常數估測，國立中山大學國立成功大學機械工程學系，中國機械工程學會第21屆全國學術研討會論文集，高雄，第1-5頁。
[48] 林炤映(2004)，以水質自動監測系統與統計方法分析日月潭水庫之水質變化趨勢，大葉大學環境工程學系碩士班碩士論文，台北，第38-50頁。
[49] 蔣安仁(2004)，開刀房利用最佳化之研究，國立中山大學醫務管理研究所碩士論文，高雄，第20-38頁。
[50] 桂德君(2004)，環境因素對菊花心種龍鬚菜外表形態碎形維度的影響，國立中山大學海洋生物研究所碩士論文，高雄，第13-15頁。
[51] 林伯航(2004)，應用碎形分析河川棲地分佈之時空特性，國立中央大學土木工程研究所碩士論文，桃園，第20-27頁。
[52] 賴亦德與陳俊宏(2004)，探討光潤金線蛭(Whitmania laevis)捕食有口蓋淡水螺類之偏好，國立台灣大學生命科學系，特有生物研究，第6卷，第2期，第67-78頁。
[53] 梁鈞泰(2005)，隨機行走於布朗運動間，物理雙月刊，國家圖書館，第27卷第3期，第461-464頁。
[54] 陳亭羽與賀千盈(2006)，應用決策樹探討適用於電子行銷市場之區隔基礎，管理科學研究，vol.3，No.1，第1-25頁。
[55] 陳煦森(2006)，台灣西南近岸海域底棲蝦類群聚之研究，國立中山大學海洋生物科技暨資源學系碩士論文，高雄，第8-12頁。
[56] 謝菱春(2006)，玉山國家公園之高山植群分類研究，國立中山大學生物科學系研究所碩士論文，高雄，第13-18頁。
[57] 楊智凱(2006)，台灣產羊耳蒜屬植物之分類研究，國立中山大學生物科學系研究所碩士論文，高雄，第12-15頁。