劉氏轉換法支撐向量機應用於皮膚病的分類

Application of Liu-Transformation and Support Vector Machine in the Dermatology Disease Classification

廖健雄(Chien-Hsiung Liao)

指導教授 ： 劉湘川

亞洲大學/資訊學院/生物資訊學系碩士班/碩士(2007年)

摘要

在台灣溫暖潮濕的氣候型態下，皮膚病是人們常會發生的疾病，發病的原因很多，不同的皮膚病會有不同的症狀。本研究主要針對六種皮膚疾病作為研究主題，將皮膚病資料經由特徵轉換，再利用支撐向量機(SVM)分類，建構一個新的皮膚醫學疾病的預測模型。
支撐向量機是一個被普遍採用的分類器，在原始資料分類前做轉換，可以得到更好的效能，但即使這樣做了，得到的結果並不是很令人滿意。在這篇文章裡，考量兩種轉換法：一個是眾所周知的NWFE轉換法，另一個是先前我們提議採用的新轉換法—劉氏轉換法(Liu-Transformation)。實驗採用一個真實資料做5-fold、10-fold和Leave-one-out的交叉比對正確率，用來評估支撐向量機，在不使用任何轉換法、使用NWFE轉換法及使用劉氏轉換法在分類上的效果。實驗結果顯示支撐向量機配合被推薦的劉氏轉換演算法有最好的效能。
藉由此研究，將這些人工智慧的分類技術，提供醫師或病人作為診斷時的輔助參考，避免不必要的醫療資源浪費，以提升醫療服務品質。

關鍵字

劉氏轉換法 ； 無參數加權特徵萃取 ； 支撐向量機

並列摘要

Dermatological disease is common in places that are humid, damp, and hot like Taiwan, various reasons can cause skin diseases with which the symptoms are varied. This study has mainly focusing on six major types of skin diseases which the actual data of the skin condition were feature transformed and were then classified by using SVM to build a new prediction model for dermatology.
The support vector machine (SVM) classifier is a popular and appealing classifier. It could be improved by taking some transformation about the original data before classification even sometimes its performance is not good. In this paper, two transformations are considered. One is the well known transformation, NWFE-Transformation and the other is a novel transformation, Liu-Transformation proposed by our previous work. For evaluating the performances of the SVM without any transformation, the SVM with the NWFE-Transformation and the SVM with the Liu-Transformation, a real data experiment by using 5-fold 10-fold and Leave-one-out Cross-Validation accuracy is conducted. Experimental result shows that the SVM with the proposed Liu-Transformation algorithm has the best performance.
The study is conducted in the hope that the AI classification technology can serve as useful references for physicians and patients in the diagnose of the skin diseases to avoid unnecessary medical expenses and to enhance health care quality.

並列關鍵字

Liu–Transformation ； Nonparametric Weighted Feature　Extraction, NWFE ； Support Vector Machine, SVM

參考文獻

13.Hsiang-Chuan Liu (2008). "A novel nonparametric weighted feature extraction transformation algorithm based on the outmost points". Journal of Taichung University(JNTCU), 22(1),1-7.

2.Cortes, C. and V. Vapnik (1995). "Support-vector networks". Machine Learning 20, 273–297.

4.S. R. Gunn (1998). "Support Vector machines for classification and regression". Technical Report, University of Southampton.

8.B. C. Kuo & K. Y. Chang (2007). "Feature Extractions for Small Sample Size Classification Problem". IEEE Trans. on Geoscience and Remote Sensing, 45(3), 756-764.

10.B. C. Kuo & D. A. Landgrebe (2004). "Nonparametric Weighted Feature Extraction for Classification". IEEE Trans. on Geoscience and Remote Sensing, 42(5), 1096-1105.