利用基因演算法之Fuzzy ID3方法

Abstract

許多關於圖形辨識、機器學習和專家系統的解決方法，智慧診斷系統最常被使用於。在這個領域中一項最重要的發展是ID3方法，它是個受歡迎且有效的方法。ID3對符號屬性的資料，產生一個決策樹來做分類辨識，但不需要大量的計算過程。根據ID3方法的精神，模糊ID3方法依照資料各特徵向量重要性及所定義之模糊函數來產生決策樹。FID3可被延伸應用至處理含有連續數特徵屬性的資料，而不只是可處理符號屬性的資料而已。 在本篇論文中，我們提出一個以基因演算法為基礎的Fuzzy ID3理論方法，來建構一個模糊分類系統，其中特徵向量所定義模糊集有最佳的參數。接著我們提出一個刪簡的方式來使我們所得到的模糊規則庫更精簡有效率，最後我們利用到一些有名資料來驗證本方法的有效性。

Different approaches from pattern recognition, machine learning, and expert systems have been used in intelligent diagnostic systems. One of the most significant developments in this domain is the ID3 algorithm, which is a popular and efficient method of making a decision tree for classification from symbolic data without much computation. The fuzzy decision tree rooted from ID3 algorithm is similar to that of ID3 algorithm. Fuzzy ID3 algorithm is extended to apply to a data set containing continuous attribute values instead of symbolic attribute and generates a fuzzy decision tree using fuzzy sets. In this thesis, we proposed a genetic algorithm based Fuzzy ID3 algorithm to construct a classification system with a set of best tuned fuzzy membership functions. Next, we formulated a pruning method for our algorithm to obtain a more efficient rule base. Finally, we validate our new FID3 scheme via some famous data sets.