Design and Implementation of Real-time Heterogeneous Data Analyzer over Cloud Computing Platform

• Main Authors: Yong-Ting Wu, 吳勇霆
• Other Authors: Jenq-Shiou Leu
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/sz8h45

碩士 === 國立臺灣科技大學 === 電子工程系 === 104 === Nowadays people can easily connect to the Internet and produce data through mobile applications anywhere and anytime. They can also upload their status words, pictures, or videos to the network as they wish, thus data volumes will grow rapidly. One of the data generators is social network, and many people join social networks all over the world. In the social network, there are thousands of people producing data at the same time, and huge of data will be generated in seconds. In addition, the social network users are increasing recently, and data is growing faster and faster than before. To make good use of those data, we will build a real-time data analyzer in our study, and collect tweet data flying in the social network to process and analyze for finding out the hot topics around the world. We will also conduct the sentiment analysis to get people’s sentiment for each issue, and show the positive and negative sentiment distributions. On the other hands, we will fetch web news content data to check the trend consistency with social network. In our system, we use the cloud computing platform to process data in real-time and quickly get the analysis results. In accordance with our experiment result and evaluation, we can find the topic trends between social network and web news are consistent after the real-time data processing and analysis.