混合式粗略集合論於水質分析之應用

A Hybrid Rough Set Model for Water Quality Analysis

李鳳娟(Fong Chuan Lee)

指導教授 ： 白炳豐

國立暨南國際大學/管理學院/資訊管理學系/碩士(2009年)

摘要

日常用水的安全與否關係著世界上所有人的健康狀況，所以水質分析在近年來一直大力推行與實踐。在所有的資料探勘的方法中，粗略集合論 (Rough Set Theory, RST)運用的領域甚廣，但並未在水質分析進行深入探討。並且在RST中，重要屬性的擷取在資料更為龐大時非常耗時，故本研究建立混和式粗略集合論結合離散化方法及運用統計方法來進行屬性擷取來進行台灣水質與台灣環境因素的關連性。混和式粗略集合論是以環境因素為條件屬性及水質分析為其分類標準來進行模型檢定。結果顯示出RST與多元邏輯斯迴歸可以使水質分析方便且準確並且能歸納出規則供水質分析專家運用。因此，混和式粗略集合論一個值得深入且對水質分析有幫助的一套方法模型。

關鍵字

水質分析 ； 多元邏輯斯迴歸 ； 粗略集合論

並列摘要

Water quality analyses have started to put many efforts these days. In our everyday life, clean, fresh water for drinking, cooking, washing, sewage disposal, and agriculture is vital to healthy human life. Rough set theory (RST) is a novel technique in data mining and has been successfully employed in many fields. However, the application of RST has not been widely investigated in water quality analysis. Furthermore, the generation of reducts for RST models is very time-consuming when the problem size increases. Therefore, the aim of this investigation is to develop a hybrid model combining dimensionality reduction method, statistical method for attribute extraction and RST to analyze relation between water qualities and environmental factors in Taiwan. This study used environmental condition factors and the degree of water pollution to examine the feasibility of the proposed model. Empirical results indicated that the model combined with multinomial logistic regression and RST could analyze the water quality efficiently and accurately, and provide decision rules for staff of water quality management. Thus, the proposed model is a promising and helpful scheme in analyzing water quality.

並列關鍵字

Water Quality ； Multinomial Logistic Regression ； Rough Set Theory

參考文獻

1. Abou-Elela, S.I., Nasr, F.A., Ibrahim, H.S., Badr, N.M., Askalany, A.R.M., 2008. Pollution prevention pays off in a board paper mill. Journal of Cleaner Production 16, 330-334.

2. Aguilera, P.A., Frensich, A.G., Torres, J.A., Castro, H., Vidal, J.L., Canton, M., 2001. Application of the Kohonen neural network in coastal water management: methodological development for the assessment and prediction of water quality. Water Research 35, 4053-4062.

3. Arms, K., 1994. Environmental science: 2nd Ed. Orlando, Florida: Saunders College Publishing.

4. Bello, R., Nowe, A., Caballero, Y., Goméz, Y., Vrancx, P., 2005. A model based on Ant Colony System and Rough Set Theory to Feature Selection. Proceedings of the 2005 conference of Genetic and Evolutionary Conference, Washington DC, USA, 275-276.

5. Bello, R., Gomez, Y., Nowe, A., García, M., 2007. Two step particle swarm optimization to solve the feature selection problem. In: Proc 7th International Conference on Intelligent Systems Design and Applications (ISDA 2007).