以語意為基礎之佛典知識結構在計算機中的表達－以心經為例

Abstract

由於古漢語在語言結構上的特質，以及佛典譯法的不同，使得古漢文佛典在資訊檢索上很難達到最佳化的效果，本文以《心經》為例，嘗試利用詞彙分析與結構化知識表達的方法，透過漢文佛典的知識結構－科判為中心軸，一方面探討計算機中佛典多版本呈現問題，另一方面藉由詞彙語意聯繫以窺探佛典知識全貌及佛教教理知識體系。

本研究之目的為：提供使用者一個善用資訊工具以整合個人之知識建構的方法，使之能將佛典內容進行結構化的知識表達，建立個人化的佛典知識庫，並利用知識節點的關聯性，使之能透過語意網絡的架構，令使用者得以概括綜覽佛典完整知識架構及佛教教理知識體系。

Because classical Chinese and modern Chinese have different syntax, and because there are different ways of translating sutras into classical Chinese, it is difficult to effectively search for Chinese expressions (classical or modern) in classical Chinese translations of sutras.

This thesis demonstrates one way to solve this problem, by analyzing sentences into phrases and structuring the phrases, using the Heart Sutra as an example.

Ke-Pan is used as the central idea to connect the different classical Chinese translations of the sutra, and at the same time to give an overview of the sutra’s structure. The purpose is to let users use existing software tools to structure the information in the sutra, so a user can establish his or her own Knowledge-based systems for the sutra and can also use points of connection to give an overview of the ideas in the sutra.