A Data Processing Model for Big Data Based on Hadoop Approach：An Example of Air Pollution

• Main Authors: Zeng,YAN-JHIH, 曾彥志
• Other Authors: CHEN,JYI-TA
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/2x632j

碩士 === 南臺科技大學 === 資訊管理系 === 104 === In recent years, the Big Data research has an explosion growth. The Institute for Information Industry said that software market trends by " mobile applications, big data, social media, cloud computing," driven by the integration of emerging technologies, such as innovative service and look forward to the future. In an era of Big Data, from large and messy data, identifies effective hidden information to improve organizational efficiency of operation is very important. This paper plan the concept of Hadoop-based distributed computing technologies by integrating with HDFS, MapReduce and other related approaches, using SQL server analysis services(SSAS) data mining analysis function to construct the mining model of air pollution. The final results under data mining procedure produce easy to understand information, reports, assessment data. For example, the continental air mass southward and haze in China result in concentrations of PM10 and PM2.5 standard, the results derived analysis suggested that there the heart, respiratory and cardiovascular diseases in adults and children, should reduce physical exertion, especially for outdoor activities. Those data complete the goal of fast processing about Big Data and get valuable data for business and related uses to prevent the problems of air pollution.