以研究者初始想法為基礎之問卷內容自動生成模式

An Automatic Questionnaire Generation Model Based on Researcher’s Initial Ideas

許暐羚(Wei-Ling Syu)

指導教授 ： 侯建良

國立清華大學/工學院/工業工程與工程管理學系/碩士(2014年)

https://doi.org/10.6843/NTHU.2014.00287

摘要

當研究者欲實施問卷調查前，其乃需先釐清研究議題與問卷調查實施之目的，以產生與問卷調查目的、研究議題相關之初始想法（即欲釐清問題、原始構想問題），並依此些初始想法蒐集其設計問卷內容時之相關資訊，以參考所得之相關資訊設計問卷內容。然而，在研究者蒐集其初始想法相關之資料時，其往往需花費許多時間篩選與其初始想法相關之資料，並耗費精力理解此些相關資料之細節，以依據理解後之結果設計問卷內容。此外，研究者可能因考量不周而未合理地設計問卷內容，導致後續之問卷分析不盡能完整且合理，甚至造成研究者實施問卷調查之效度降低。因此，為解決上述問題，本研究乃提出一套「以研究者初始想法為基礎之問卷內容自動生成」模式，其可先解析研究者初始想法的質化與量化特性，並依解析結果發展一套問卷內容自動生成方法論，此方法乃可將研究者之初始想法予以結構化，並可將結構化之研究者初始想法搭配基因演算法，以自動地篩選問卷內容相關資料，之後再從相關資料中自動地擷取可用素材，以生成初步問卷內容；最後再將初步問卷內容後製，以提升研究者實施問卷調查之效度，並將所生成之問卷內容提供予研究者，以提升研究者設計問卷內容時之效率。

關鍵字

問卷生成 ； 機率分配 ； 基因演算法

並列摘要

As a researcher wants to design a questionnaire, he/she has to clarify the purposes of the survey to figure out the initial ideas about the questionnaire design. After that, the researcher has to search related information for questionnaire design based on their initial ideas, which is usually cost and time consuming. This research aims at developing a model for automatic questionnaire generation. Before developing the model, this study analyzes the statistic characteristics of components of initial ideas. On the basis of the analysis results, this research develops a methodology for automatic generation of questionnaires. By using the methodology, the initial ideas of a researcher can be converted into structured components and the genetic algorithm can be applied to extract questionnaire related data from the Internet based on the structured components. Useful materials from the related data can be extracted to generate a preliminary questionnaire. After that, the questions from the preliminary questionnaire can be sorted redesigned. By utilizing the model, efficiency and effectiveness for questionnaire design can be enhanced.

並列關鍵字

questionnaire design ； probability distribution ； genetic algorithm

參考文獻

1. Adigüzel, F., and Wedel, M., 2008, "Split questionnaire design for massive surveys," Journal of Marketing Research, Vol. 45, No. 5, pp. 608-617.

4. Bagheri, A., Saraee, M. and de Jong, F., 2013, "Care more about customers: Unsupervised domain-independent aspect detection for sentiment analysis of customer reviews," Knowledge-Based Systems, Vol. 52, pp. 201-213.

5. Balakrishnan, V. and Zhang, X., 2014, "Implicit user behaviours to improve post-retrieval document relevancy," Computers in Human Behavior, Vol. 33, pp. 104-112.

6. Blake, C., 2010, "Beyond genes, proteins, and abstracts: Identifying scientific claims from full-text biomedical articles," Journal of Biomedical Informatics, Vol. 43, No. 2, pp. 173-189.

8. Bouras, C. and Tsogkas, V., 2010, "Noun retrieval effect on text summarization and delivery of personalized news articles to the user’s desktop," Data & Knowledge Engineering, Vol. 69, No. 7, pp. 664-677.