以多媒體劇情輸入方式之中文成語檢索系統之研究

The study of chinese idiom search system using multimedia scenarios as input

張宏茂(Hung-Mao Chang)

指導教授 ： 陳登吉

國立交通大學/資訊學院/資訊學院資訊學程/碩士(2013年)

https://doi.org/10.6842/NCTU.2013.00355

摘要

隨著資訊檢索技術的進步及廣泛應用在各領域上，在語文學習上的應用已經由紙本形式，歷經單機軟體查詢方式，逐漸演進成網路查詢的型態，除了能快速的取得各種新的資訊及解釋外，更能實現無所不在的學習環境。然而，隨著行動裝置快速的普及，傳統使用鍵盤輸入的查詢方式，在行動裝置上顯得相對的不便。傳統的文字成語檢索系統，使用者需先牢記成語的正確寫法，才能順利進行查詢。雖然部分研究提出關聯式檢索的技術，但是對於難以使用文字表達的情境則顯得幫助有限。因此，本研究擬提出視覺化劇情成語檢索系統，使用圖像式操作介面，透過圖片及劇情編輯為搜尋關鍵進行成語檢索，提供一種更直覺、更自然的輸入方式，來解決成語檢索的困難。
利用圖像易於拖拉放的操作特性，輸入視覺化元件進行成語的檢索，讓使用者能輕鬆的使用手機及平板電腦等各式可攜帶手持裝置進行操作，解決成語學習於行動裝置上的缺陷。除了利用圖像進行簡單的檢索功能外，也能讓使用者使用拖拉圖像的方式來進行劇情的編輯，並將編輯的劇情自動的轉換成檢索的關鍵字詞。透過劇情編輯的操作，更能幫助學生從中了解成語的意境及內涵，利用做中學的理論特性，幫助並加強學生對於成語的了解與記憶。
本研究也設計了學習成效、認知負荷及軟體使用性的相關實驗，結果顯示，應用視覺化劇情成語檢索系統於國小學生的成語自我學習上，與使用傳統文字檢索有相同的學習成效。於認知負荷的分析上，我們發現，透過視覺化劇情成語檢索系統能幫助減輕學生學習的壓力，並能幫助成語的記憶。在使用者的軟體使用性的評估上，結果呈現著使用較便利、願意經常性使用及介面設計便於閱讀等正面的回饋。

關鍵字

多媒體 ； 視覺化劇情 ； 劇情重用 ； 圖像提示 ； 圖像提示 ； 成語查詢 ； 成語檢索 ； 同義詞

並列摘要

With the advances in information retrieval technology and widely used in learning fields, in addition to quickly obtaining a variety of new knwoledge and explanations, the better to achieve ubiquitous learning environment, it has been reported by researchers in the language learning applications, through stand-alone software query, and gradually evolved into a network query patterns. However, it is relatively inconvenience on mobile devices while using the traditional way by the keypad to enter a query with the mobile devices. The users must remember correct wording of idioms in order to successfully carry out inquiries in traditional text idiom search system. Although some studies propose associative search techniques, it may sometimes limited to the difficultly to use appropriate words to express the situation. Therefore, this study intends to make idiom search system of using visual scenario as input. By using graphical user interface, through pictures and story editor as the search key to proceed idiom search for providing a more intuitive, more natural input methods, to solve the idiom search difficulties.
Using the benefit of graphical dragging, pulling, setting free easily, inputting visual components idiom search, allowing users to easily use mobile phones and tablet computers and other kinds of portable handheld devices in operation to solve learning idioms flaws on mobile ones. Not only using simple image search function, but also allowing users to use ways of dragging and dropping graphics to proceed the editor of scenario and scenario of editor automatically for being converted into searchable keywords. Through the operation of scenario editing will help the students learn about the mood and meaning of idioms, assist and enhance students' understanding and memory for idioms, using theoretical characteristics by learning in progress.
The study also designed related experiments of learning achievement, cognitive load and software usability for future investigation. The results showed that the learning achievement of using visual scenario search system for elementary school students on idiom self-learning is the same as the traditional text search one. In the analysis of cognitive load, we found it can help alleviate the pressure of students' learning through idiom search system of visual scenario, and can help memorize idiom for them well. In the user's software usability evaluation whose results show such positive feedbacks as using more conveniently, using more often willingly and reading more easily for interface design.

並列關鍵字

Multimedia ； Visual Scenario ； Reusable Component ； Image Prompting ； Information Retrieval ； Idiom Search System ； Idiom Retrieval System ； Synonyms

參考文獻

[3] 張正霖，「以成語涵義為基礎之中文成語檢索系統」，國立交通大學，碩士論文，民國98年。

[2] Zhang Hui, et al. "ERP Correlates of Compositionality in Chinese Idiom Comprehension", Journal of Neurolinguistics, Vol. 26, No.1, pp. 89-112, 2013.

[4] Rommers, Joost, Ton Dijkstra, and Marcel Bastiaansen. "Context-Dependent Semantic Processing in the Human Brain: Evidence From Idiom Comprehension.", Journal of Cognitive Neuroscience, Vol.25, No.5, pp. 762-776, 2013.

[6] Ran Jie, Qi Li-juan. "The Research of Semantic Retrieval Model Based on Idiom Literary Quotation Ontology", National Conference on Information Technology and Computer Science, pp. 587-589, 2012.

[10] 張瀚文，「The Description and Analysis of Images」，大學圖書館，第三卷第一期，104~115頁，民國88年1月。