Study on Knowledge Management for Climate Change Adaptation with Text Mining Techniques - A Case study of Water Resources

• Main Authors: Jung Huang, 黃蓉
• Other Authors: 童慶斌
• Format: Others
• Language: zh-TW
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/hds568

碩士 === 國立臺灣大學 === 生物環境系統工程學研究所 === 105 === Text mining becomes popular in recent years. It helps people saving many time and efforts when they mining useful information from huge amount of text. As a result, there are many researches and applications about text mining. Another issue getting people much attention is climate change. To reduce the risk of climate change, it is necessary to develop adaptation action plan, which mining useful information is essential. However, there are few researches applying text mining to climate change study. The pourpose of this study is to strengthen climate change adaptation knowledge management by using DIKW pyramid, text minging and the six-step decision support tool developed by the Taiwan integrated Research Program on Climate Change Adaptation Technology (TaiCCAT). In addition, this study focuses on water resources. The textual data are download from Scopus that is an online abstract and citation database. There are 1788 articles that title, abstract or keywords has climate change, adaptation and water resources. Their publication year are from 1982 to 2017, and about 80% of them published by journals. Morever, the most of articles are from the journal of “Climatic Change”. There are two results after using text mining. First result is term frequency. These terms are classified into sereral categories, including water usage, event, climate change risk assessment, other field, hydrology, and solving strategy. These categories can build keyword databases to help users quickly and precisely know the most popular terms of the group when they need to identify key issues. Another result is the cluster analysis to help user narrow down the number of the articles when they need to identify adaptation options.