應用文字探勘之自動化新聞文本分析以探討社會對新聞事件之反應

Automatic Content Analysis Using Text Mining to Investigate How News Events Trigger the Response of Society

戴瑜廷(Tai, Yu Ting)

指導教授 ： 林福仁

國立清華大學/科技管理學院/服務科學研究所/碩士(2015年)

摘要

近年來食品安全問題層出不窮，接連爆發塑化劑、毒澱粉、假油一連串的事件。然而，相關報導的數量龐大，一般民眾難以有效閱讀完所有資訊；再者，一連串的事件都與食品安全相關，社會是否從過去事件中學習到經驗，並在類似事件發生時做出不同的因應也是值得探討的議題。
但新聞閱聽者難以從非結構化的訊息中了解事件之間社會反應的差異，因此本研究的目的在於自動化分析同一主題的多個事件，探討社會對新聞事件的反應。
本研究旨在提出一個自動化的文本分析系統，自動分析隸屬同一主題的多個新聞事件。首先，本研究透過分群技術(Clustering)，以事件發展階段及利害關係人二維向度，呈現各利害關係人在事件各階段的言論內容。再者，系統將透過摘要技術(summarization)萃取事件發展重點以提供單一事件發展的新聞摘要。最後，以焦點訪談法(ORID)衡量系統的有效性，並同時探索讀者對於事件的反應。
藉由本研究提出的自動化文本分析系統，一般民眾可以更快速及有效的了解新聞事件的發展，回顧事件發生當下的感受、想法與行動。

關鍵字

新聞摘要 ； 文字探勘 ； 文本分析 ； 食品安全 ； 焦點訪談 ； 社會學習

並列摘要

In recent years, the crisis of food safety events continued happened in interval. There are three main food safety events, in sequence, “Plasticizer”, “Poison starch” and “Fake oil”. However, the related news reports are too enormous to be digested efficiently by the readers. In addition, it’s interested to know if similar events happen again, would they learn something from the past experiences and responds in a different way.
This study aimed to propose a system that can automatic analyze the related news belonging to the same topic. First, this study presents the opinions of each stakeholder on each period of the news development by clustering. Second, this system extracts the important content of news reports using summarization and provides the summarization of each news event to readers. Finally, this study combines the system with Focused Conversation Method (ORID) to evaluate the effective of the system and to explore the response of readers to the news events.
With the facility of the system that we proposed, the readers can understand the development of news event efficiently and recall their feeling, thought, and reaction for the news events at the moment that the event happened.

並列關鍵字

Summarization ； Text Mining ； Automatic Content Analysis ； Food safety ； Focused Conversation Method ； ORID ； Social Learning

參考文獻

Alguliev, R. M., Aliguliyev, R. M., & Mehdiyev, C. A. (2011). Sentence selection for generic document summarization using an adaptive differential evolution algorithm. Swarm and Evolutionary Computation, 1(4), 213-222.

Baptiste, N. (1995). Professional development Always growing and learning: The ORID—A technique to enhance communication. Early Childhood Education Journal, 22(4), 39-40.

Berghel, H. (1997). Cyberspace 2000: Dealing with information overload. Communications of the ACM, 40(2), 19-24.

Chang, Y. H., Chang, C. Y., & Tseng, Y. H. (2010). Trends of science education research: An automatic content analysis. Journal of Science Education and Technology, 19(4), 315-331.

Feldman, R., & Sanger, J. (2007). The text mining handbook: advanced approaches in analyzing unstructured data. Cambridge University Press.