請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/68217
標題: | 整合社群網路訊息之災害資訊分析 Disaster Information Analysis of Integrating Social Media Message |
作者: | You-Rui Liu 劉又瑞 |
指導教授: | 徐百輝(Pai-Hui Hsu) |
關鍵字: | 災害,災害資訊,社群網路,文本分析,機器學習, Disaster,Disaster information,Social media,Text analysis,Machine learning, |
出版年 : | 2020 |
學位: | 碩士 |
摘要: | 近年來社群網路的蓬勃發展,已經使其成為各種資訊蒐集的主要來源,並常被用來當作資訊傳遞的主要媒介。一般使用者除了可被動接收社群媒體的資訊外,也可以主動地發佈、傳遞訊息,成為資訊的提供者。當災害發生時,對於災情資訊的傳遞與救災資訊的更新,社群網路便成為一個十分重要且方便的工具。在國內、Facebook(臉書)及PTT可說是大家最常用的社群網路,其優勢為使用者不僅可以留言,也可以上傳照片,這個特點在災害防救的應用上是十分重要的關鍵。若能透過網路即時傳達災害資訊,相關搜救單位就能迅速掌握災情並因地制宜,作出更有效率的救災應變措施。在台灣經常發生的災害,多為地震、水災等,而當災害發生時,第一時間的黃金搜救時間就成為了關鍵時刻,而對於救災工作的進行,首要了解的便是哪裡發生災害?災害的嚴重與否?故最重要的就是能即時取得災害資訊,災害資訊的即時取得就須仰賴第一時間在災害現場的民眾即時回報,讓救災單位能把握時間進行救援,讓災害不再擴大。故此研究主要探討以社群網路蒐集災害資訊之應用,並針對災害資訊進行分析,期望能透過社群網路的即時性,快速蒐集災害資訊,判斷出災害發生的位置,即時作出處理,並針對災害資訊,進一步作文本關聯性分析,期能從中判別出可能衍生的相關災害,另外亦針對分析成果,以機器學習方法進行成果評估。 In recent years, the social network has flourished. Many sources of information are transmitted through the social network. In addition to being a tool for people to connect, it plays an important role in disaster response. In the past, most users usually passively obtained information through the Internet. With the rise of social media, they can actively deliver messages and become information providers. In the event of disasters, using the social network to actively deliver information becomes a very beneficial tool. With the development of technology, as long as the disaster information is transmitted through the Internet, search and rescue units can immediately grasp the situation of the disaster and make a more effective disaster response. There are many kinds of disasters. The disasters that occur frequently in Taiwan are mostly earthquakes, floods, etc. The most important thing is to get disaster information in real-time. The immediate acquisition of disaster information depends on the immediate return of people at the disaster site so that the search and rescue units can grasp the time to rescue and let the disaster not expand. This research focuses on the application of social network to collect disaster information and analyzes disaster information. Through the immediacy of the social network, it can quickly collect disaster information and immediately deal with it. For the disaster information, further textual relevance analysis and machine learning can be used to identify relevant disasters that may be derived. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/68217 |
DOI: | 10.6342/NTU202003775 |
全文授權: | 有償授權 |
顯示於系所單位: | 土木工程學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
U0001-1708202015403000.pdf 目前未授權公開取用 | 3.07 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。