類神經網路已經被驗證可以應用於自動化的文件分類。然而，相較於其他的分類方法，要建構一個能夠自動分類的類神經網路，必須耗費龐大的空間以及冗長的訓練時間，這樣的限制導致類神經網路實用性大幅的降低。對此，本論文針對人類的神經系統與學習模式進行研究，並結合現有的類神經網路理論，提出了一個以人性為基礎的類神經網路概念，命名為人性化類神經網路。此網路概念主要包含了一個以擁有多種不同能力為導向而建立的類神經網路架構，與其在進行運作時可自動進行學習的非監督式學習方法。
本論文使用Reuters-21578進行文件分類之實證分析，觀察所提方法之成效。實驗結果顯示人性化類神經網路可以有效的將文件進行分類，其最佳的F1-measure可達92.5%；而所提的非監督式學習法則可有效且迅速的提升分類的準確性；且隨著架構的擴充，訓練與測試的時間是以線性成長，相較於傳統類神經網路的非線性成長，人性化類神經網路具有更良好的擴充性與實用性。

Artificial neural network (ANN) has been applied to text classification. However, comparing with other classification methods, ANN needs enormous memory space and training time to build the model. This makes ANN infeasible in practical applications. In this paper, we try to adopt the ideas of human neural system and learning style, and combine with the existing models of ANN. We propose a humanized neural network architecture that is based on human intelligence. In this architecture, a neural network with multiple abilities to solve numerous problems is presented, and with its unsupervised learning algorithm.
In our experiment, Reuters-21578 was used as the dataset to show the effect of the proposed architecture on text classification. The experiment result showed that HNN can effectively classify texts with the best F1-measure of 92.5%. It also showed the learning algorithm can enhance the accuracy effectively and efficiently. With the expansion of the architecture, the training time and test time shows a linear growth. Comparing with ANN, HNN has better scalability and practicality.

中文摘要 i
英文摘要 ii
誌謝 iii
目錄 iv
表目錄 vii
圖目錄 viii
第一章　緒論 1
1.1　研究背景與動機 1
1.2　研究目的 1
1.3　論文架構 2
第二章　文獻探討 3
2.1　類神經網路 3
2.1.1　類神經網路概念 3
2.1.2　基本網路架構 5
2.1.3　倒傳遞學習法 6
2.1.4　類神經網路特徵 6
2.1.5　類神經網路應用於文件分類 7
2.2　文件分類演算法 8
2.2.1　樸素貝葉斯分類器 8
2.2.2　支援向量機 9
2.2.3　近鄰演算法 9
第三章　研究方法與架構 10
3.1　研究方法介紹 10
3.1.1　方法流程圖 10
3.1.2　人性化類神經網路之子網路建構 11
3.1.3　人性化類神經網路架構合併 11
3.1.4　自主性非監督式學習 12
3.2　人性化類神經網路 13
3.2.1　人性化類神經網路概念 13
3.2.2　人性化類神經網路架構 14
3.2.3　人性化類神經網路特徵 16
3.3　自主性非監督式學習法 17
3.3.1　自主性非監督式學習法概念 18
3.3.2　自主性非監督式學習法流程 18
3.3.3　自主性非監督式學習法特徵 19
第四章　實驗架構與流程 20
4.1　使用資料集介紹 20
4.2　實驗設計 21
4.3　系統架構介紹 21
4.3.1　資料前處理 22
4.3.2　建構人性化類神經網路 23
4.3.2.1　文件特徵識別 24
4.3.2.2　特徵節點建構 25
4.3.2.3　關係連結分析 26
4.3.2.4　文件相似度估計 27
4.3.2.5　非監督式學習 27
第五章　實驗結果與分析 29
5.1　實驗使用平台 29
5.2　比較方法介紹 29
5.3　驗證方法介紹 29
5.4　實驗結果分析與討論 32
5.4.1　最大10類別分類正確率比較 32
5.4.2　90類別分類正確率比較 34
5.4.3　最大10類別訓練時間與測試時間比較 36
5.4.4　90類別訓練時間與測試時間比較 39
第六章　結論與建議 42
6.1　研究結論 42
6.2　後續研究建議 42
參考文獻 43

[1]C. M. Bishop, “Neural Networks for Pattern Recognition”, Clarendon Press, Oxford, 1995.
[2]C. Apté, F. Damerau, S. M. Weiss, “Automated Learning of Decision Rules for Text Categorization”, ACM Transactions on Information Systems (TOIS), Vol. 12 No. 3, pp. 233-251, July 1994.
[3]G. Cybenko, “Approximations by Superpositions of a Sigmoidal Function”, Math. Control Signals Systems, Vol. 2, pp. 303-314, 1989.
[4]A. Cochocki, R. Unbehauan, “Neural Networks for Optimization and Signal Processing”, John Wiley and Sons Inc., 1993.
[5]J. Herrero, A. Valencia, J. Dopazo, “A Hierarchical Unsupervised Growing Neural Network for Clustering Gene Expression Patterns”, Oxford University Press Bioinformatics, Vol. 17 No. 2, pp. 126-136, 2001.
[6]J. J. Hopfield, “Neural Networks and Physical Systems with Emergent Collective Computational Abilities”, Proceedings of the National Academy of Sciences of the USA, Vol. 79 No. 8, pp. 2554–2558, April 1982.
[7]D. E. Rumelhart, G. E. Hinton, J. L. McClelland, “Learning Representations by Back-Propagating Errors”, Nature, Vol. 323, pp. 533-536, 1986.
[8]P. J. Werbos, “Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences”, PhD Thesis, Harvard University, 1974.
[9]H. M. Lee, C. M. Chen, C. W. Hwang, “A Neural Network Document Classifier with Linguistic Feature Selection”, Proceeding of the International Conference on Industrial and Engineering Applications of Artificial Intelligence and Expert System, Vol. 13, pp. 555-560, June 2000.
[10]Z. Q. Liu, Y. Zhang, “A Competitive Neural Network Approach to Web-Page Categorization”, International Journal of Uncertainty, Fuzziness and Knowledge Based Systems, Vol. 9, No. 6, pp. 731-741, 2001.
[11]M. E. Ruiz, P. Srinivasan, “Hierarchical Text Categorization Using Neural Network”, Information Retrieval, Vol.5, pp. 87-118, 2002.
[12]A. Selamat, S. Omatu, “Web page feature selection and classification using neural networks”, Information Sciences, Vol. 158, pp. 69-88, 2003.
[13]M. L. Zhang, Z. H. Zhou, “Multi-Label Neural Networks with Applications to Functional Genomics and Text Categorization”, IEEE Transactions on Knowledge and Data Engineering, Vol. 18, pp. 1338-1351, October 2006.
[14]A. F. De Souza, F. Pedroni, E. Oliveira, P. M. Ciarelli, W. F. Henrique, L. Veronese, C. Badue, “Automated multi-label text categorization with VG-RAM weightless neural networks”, Neurocomputing, Vol. 72, pp. 2209-2217, 2009.
[15]M. Ghiassi, M. Olschimke, B. Moon, P. Arnaudo, “Automated text classification using a dynamic artificial neural network model”, Expert Systems with Applications, Vol. 39, pp. 10967-10976, 2012.
[16]B. E. Boser, I. M. Guyon, V. N. Vapnik, “A Training Algorithm for Optimal Margin Classifiers”, 5th Annual ACM Workshop on COLT, pp. 144-152, 1992.
[17]D. D. Lewis, “Reuters-21578 Text Categorization Test Collection Distribution 1.0”, AT&T Labs, September 1997.
[18]G. Holmes, A. Donkin, L. H. Witten, “Weka: A Machine Learning Workbench”, Proc Second Australia and New Zealand Conference on Intelligent Information Systems, pp. 357-361, November 1994.