臺灣博碩士論文加值系統

文學作品中往往蘊含豐富的訊息, 有些含意, 即使是人腦這樣熟練而精細的自然語言系統, 都不一定能分辨出來. 本篇論文旨在闡述一個corpus-based的方法, 針對一位文學家的作品全集所提供的前後文資訊, 將出現過的字集在Self-Organizing Map中群集分堆(cluster). 我們再利用分堆的資訊將每個字做語意編碼, 進而提供"語意檢索"的功能. 讓使用者輸入文字來尋找文學作品中密切相關的段落. 一般搜尋系統多半是利用關鍵字的統計來決定相似度, 或使用拆字的方法來提昇效果, 但本系統所使用的語意檢索可以找出語意相關的搜尋結果, 找到的段落中所包含的文字與使用者輸入的文字甚至可能沒有交集, 但是兩者蘊含了最相近的語意結構.

The information in literary works is rich. The task to read between the lines is challenging even for the most sophisticated system such as human brains. The main idea of this paper is to illustrate the design of a corpus-based method to find semantic structures in the complete works of Mark Twain (Samuel L. Clemens, 1835-1910), who was famous for his extraordinary sense of humor as well as social concern. Self-organizing Map (SOM) is applied to investigate the relations of vocabulary within the context of Mark Twain’s works. SOM make words cluster according to their context vectors, and are encoded based on their unique contextual structures. We call this practice "semantic encoding". Based on semantic encoding, a system for "semantic search" is further designed to search for passages in Mark Twain’s corpus which are closely related to user’s input. The passages found may have nothing in common with the input wording, however, they share similar semantic meaning.

章節目錄1
圖表目錄4
摘要5
ABSTRACT6
第一章 簡介7
第二章: 指標與意含9
學習自然語言的過程9
自然語言處理10
條列式窮舉法與造規則推理法10
窮舉法與造句文法的缺點11
類神經網路14
句型的學習14
系統的功能與結構16
概念點(concept node)16
角色點(conceptual role node)16
順序點(sequence position node)16
連接點(binding node)16
Elman提出的實驗介紹21
類神經網路的推廣能力25
推廣與不推廣26
第三章: 馬克吐溫作品輪廓29
第四章: 前置處理31
(A) 去標點符號31
(B) 去功能字33
(C) 還原字形33
第五章: SOM34
脈絡指標34
絕對頻率35
相對頻率35
模範指標36
SOM ALGORITHM37
學習方程式37
模範指標的選取38
norm38
entropy41
第六章: 系統改良43
虛擬模範指標43
分堆結果45
第七章: 文學作品語意編碼47
語意編碼47
編碼函式49
語意編碼的哲學意義50
第八章: 語意檢索方法與介面52
檢索流程52
比較相似度52
查詢介面53
第九章: 總結57
個人化語意搜尋引擎57
語言學習系統的自動化57
文學作品導讀58
參考文獻59

1.Guido Deboeck, Teuvo Kohnen (eds). Visual Exploration in Finance: with Self-Organizing Maps. Springer-Verlag London 1998.
2.Krista Lagus, Timo Honkela, Samuel Kaski, Teuvo Kohonen. WEBSOM for Textual Data Mining. Artificial Intelligence Review 13: 345-364, 1999.
3.Teuvo Kohonen (ed). Self-Organizing Maps. Springer-Verlag Berlin Heidelberg 1995.
4.Jeff Elman. Generalization, simple recurrent networks, and the emergence of structure. Preceedings of the 20th Annual Conference of the Cognitive Science Society. Mahway, NJ: Laurence ErlBaum Associates. 1998.
5.Jeffrey L. Elman. Finding Structure in Time. Cognitive Science, 14, 179-211, 1990.
6.R.F. Hadley, V.C. Cardei. Language acquisition from sparse input without error feedback. Neural Networks 12, 217-235, 1999.
7.M.F. Porter. An Algorithm for suffix stripping. Program. Vol 14, no. 3. pp 130-137, july 1980
8.W. B. Frakes. Stemming Algorithm. Data Structures and Algorithms. Prentice-Hall. 1992
9.Frakes, W. B., "Stemming Algorithms", in Frakes, William B. and Baeza-Yates, Ricardo (Eds.), Information Retrieval: Data Structures and Algorithms, Englewood Cliffs, NJ: Prentice-Hall, 1992. pp. 131-160.
10.Tony McEnery, Michael Oakes. Authorship Identification and Computational Stylometry. Handbook of Natural Language Processing. 545-562. Marcel Dekker, Inc. 2000
11.Ido Dagan. Contextual Word Similarity. NLP Handbook 459-475. Marcel Dekker, Inc. 2000
12.Dieter Merkl. Text Data Mining. Handbook of Natural Language Processing. 889-903. Marcel Dekker, Inc. 2000
13.P.N. Johnson-Laird, Paolo Legrenzi, Vittorio Girotto, Maria S. Legrenzi. Illusions in Reasoning About Consistency. SCIENCE vol. 288, 21 APRIL 2000.
14.Noam Chomsky. New Horizons in the Study of Language and Mind. Cambrage University Press UK. 2000.
15. H.B. Barlow, T.P. Kaushal, G.J. Mitchison. Finding Minumum Entropy Codes. Neural Computation 1, 412-423. MIT, 1989.