Study On The Big-data-based Social Media Field Monitoring Technology

Social media is an important form of Internet communication, the user through Facebook, Twitter, Sina Weibo and other online media platforms to share ideas, opinions and attitudes, forming a huge social impact. Demonstrations through social networks to launch games, color revolution and political struggles, social activities is becoming difficult to find on the Internet, resulting in a potential social and governance risks, requires the use of specific techniques in social media monitoring and management.Information on social networks to produce fast, a huge amount of data, and content distribution range of breakthrough time and space limitations of traditional media, public opinion, the existing network monitoring system can not meet the massive data collection, storage and analysis needs. Based on these problems, public opinion monitoring system based on distributed big data technology designed social media, under the Hadoop framework to achieve a collection of text data, storage and analysis, and the use of a regional monitoring network monitoring technology.On the basis of the full study public opinion monitoring system based on the design of the structural model with Hadoop-based monitoring system, the model in a hierarchical manner using different technology to solve big data corresponding control problems. Achieve a storage model and massive data collection system meets MapReduce programming model; utilization string matching algorithm, segmentation techniques and word frequency algorithms, data analysis model of distributed environments.Completion of the entire monitoring system with large data functionality and data analysis capabilities, can be used as basis for achieving different monitoring objectives.