透過您的圖書館登入
IP:3.129.23.30
  • 學位論文

新聞時間性特徵分析——以COVID-19疫情新聞為例

A Temporal Analysis of COVID-19 News

指導教授 : 謝吉隆
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


時間是新聞敘事中最基本的敘事元素,除了構成新聞事件的起承轉合,讓讀者更容易理解新聞資訊,同時反映新聞價值,當新聞發布時間與事件時間越相近,就容易被視為重要的的報導。而在數位新聞時代,新聞工作者為在更競爭的報導環境中提供即時報導,不僅考量新聞的時間相近性,也考慮新聞對讀者來說的新鮮程度,並在新聞報導加入更多來自過去、未來等不同時間層次的資訊。而此新聞與時間的關係變化,影響著公共時間的構成,並都將影響讀者對新聞事件的認知。本研究以COVID-19疫情新聞為研究對象,以瞭解新聞工作者在數位時代面臨即時報導壓力時,如何運用時間資訊進行報導,並討論不同的敘事方式反映出哪些時間特性,並影響疫情新聞敘事。 為分析新聞中的時間特徵,本研究嘗試以文字探勘方法輔助內容分析,從文本中區分出新聞中所引用的不同時間層次資訊,並透過主題建模方法,以次要新聞主題探索在關鍵疫情新聞事件前後時間性特徵的變化,藉此討論新聞工作者運用時間資訊的策略與其影響。 本研究發現,在疫情新聞報導中,新聞工作者的確以較豐富的時間層次進行敘事,反映出數位新聞時代連結時間跨度較大資訊的特徵,從研究結果可得包含過去、當下、未來共11層次時間資訊。整體而言,新聞媒體仍專注在當下的報導,其次為引用來自過去的資訊,而在運用有關未來的資訊時,新聞工作者的態度較為謹慎,且在引述時間跨度較大的資訊時,多仰賴特定人士作為預測性消息來源,對比過往研究有所差異。而在本土感染重大事件發生後,新聞工作者引用更多「過去/未來二日內」或「當日」時間層次的資訊,且從次要新聞主題的變化可見,「確診足跡與匡列隔離」、「人流管制」等主題的新聞增加。本研究所採研究方法為利用中文進行文字探勘法,雖未臻完善仍有可改進之處,其中以系統性方式檢視新聞時間資訊與針對繁體中文文本處理的經驗,仍可供未來研究參考使用。

並列摘要


Time is the most basic element in news narrative, which not only constitutes the beginning and end of news events but also makes it easier for readers to understand news information with news value. As providing real-time reports become more competitive in the digital news era, journalists have to consider the time proximity and the freshness of news at the same time. Research has shown that journalists refer to more information from the past, future, and other temporal layers. From this point of view, the role of time in the news has changed, which also affected the constitution of public time and people's relation to events that happened in the past. This study aims to understand how journalists use time as a narrative resource and deal with the pressure of highly competitive real-time reporting in the digital era with the use case of the COVID-19 outbreak news reports. The ways journalists refer to temporal information would also discuss in this article. To analyze the temporality of news, this study attempted to use content analysis with a text mining approach, distinguish different temporal layers of information used in the news, and figure out the differences in reporting strategies and time-related words usages before and after the key epidemic news events with topic modeling and secondary news topics. This study found that journalists used a richer temporal layer in the news coverage on reporting the epidemic with a larger time span in the digital news era. There are 11 temporal layers found in this study. In general, the news media still focus on reporting on the present, followed by quoting information from the past, and journalists are more cautious in using information about the future. After key epidemic news events, journalists cited more information at the "past/next two days" or "today" time level, and the change in secondary news topics showed an increase in the number of news stories on topics such as "Confirmation of footprint and quarantine" and "crowd control". The research method used in this study is a textual analysis method in Chinese. Although it is not perfect and is still room for improvement, future research could refer to the experience of examining news time information in a systematic way and traditional Chinese textual processing used in this study.

並列關鍵字

news time temporality temporal layers topic modeling

參考文獻


中研院詞知識庫小組(1998年8月)。〈中央研究院平衡語料庫的內容與說明〉。上網日期:2022年4月11日,取自http://godel.iis.sinica.edu.tw/CKIP/tr/9804_2013.pdf
《CKIP 詞類標記》(無日期)。上網日期:2022年4月11日,取自https://ckip.iis.sinica.edu.tw/CKIP/paper/poslist.pdf
林婷嫻(2018)。〈斷開中文的鎖鍊!自然語言處理 (NLP)〉。《研之有物》。上網日期:2022年4月11日,取自https://research.sinica.edu.tw/nlp-natural-language-processing-chinese-knowledge-information/
許中川、陳景揆(2001)。〈探勘中文新聞文件〉。《資訊管理學報》,7(2),103-122。https://doi.org/10.6382/JIM.200101.0103
陳世榮(2015)。〈社會科學研究中的文字探勘應用:以文意為基礎的文件分類及其問題〉。《人文及社會科學集刊》,27(4),683-718。

延伸閱讀