頁籤選單縮合
題 名 | A Tree-Net Classifier for Non-Linear Boundary Problems=樹型網路分類器於非線性界限問題之處理 |
---|---|
作 者 | 王景弘; 洪宗貝; 曾憲雄; | 書刊名 | Proceedings of the National Science Council : Part A, Physical Science and Engineering |
卷 期 | 22:4 1998.07[民87.07] |
頁 次 | 頁509-516 |
分類號 | 310.153 |
關鍵詞 | 樹型網路分類器; 非線性界限; Back-propagation; Decision tree; ID3; Neural network; Non-linear boundary; Tree net; |
語 文 | 英文(English) |
中文摘要 | 於本篇論文中,我們提出一個混合型學習演算法,它結合傳統樹型學習法及 Back-propagation類神經網路學習法,去建立一個樹型網路分類器。在整個學習過程中,我 們不需要預先設定學習網路的架構,網路最後的架構會隨著學習樹的成長自動建立。最後, 我們將此學習演算法應用於腦瘤疾病診斷領域,並和其它三種學習法比較,於準確度比較, 我們發現本演算法高於傳統樹型學習法及Back-propagation學習法,於學習速度比較,我們 發現本演算法快於Back-propagation學習法。 |
英文摘要 | In this paper, we propose a tree net classifier (TNC) that integrates the learning strategies of decision trees and modified back-propagation neural networks. The TNC uses a tree-pruning mechanism to avoid overfitting problems. It has small tree sizes anti low error rates, especially for complex classification. Furthermore, users need not lay out the structure of a tree net in advance; the structure is automatically constructed by the tree-growing process Finally, results of experiments on diagnosing brain tumors are given to compare the proposed algorithm with three other learning methods in terms of accuracy, complexity of the knowledge structure, and learning speed. The experimental results show that the proposed classifier has very high accuracy rates and learns much faster than the back-propagation neural network. |
本系統中英文摘要資訊取自各篇刊載內容。