A Platform Software Of Data Mining Based On Hadoop On Traffic And User Behavior Of Wireless Network

With the emerging of big data has brought tremendous changes to various industries, people have come to realize the importance of big data. Big data has penetrated into daily life and played an important role in determining the future direction of the industry, as a disrup-tive technology revolution after cloud computing and IOT (Internet of Things). By making decisions relying from outside causation to internal correlations of data is the greatest contri-bution of big data for humans. With the rapid development of mobile Internet technology and the data acquisition technology, the network data collected from operators and equipment manufacturers through network gateway can has a more complex structure with more dimen-sions and more detailed. Furthermore, the development of mobile network are also gradually influence the user’s behaviors, and with the high-speed development and widely used of LTE, users can have different traffic model and business model compare to traditional2G/3G net-works. So, the traditional data analysis platform cannot be available to today’s wireless net-work considered on complexity of data and data capacity.In this research, at first, we collected a huge amount of network data of different protocol, including CDMA2000, WCDMA, LTE and so on; Because of the volume of data, we use several kinds of data warehouses to store the original data in response to different scenarios;Second, as the network data has much fields and complicated construction, this paper propose a data segment method to deal with the different protocol of networks.Third, after analyzing the characteristics and internal law of the network data, the re-search did much work on detailed classification analysis and mining of the data, then this pa-per propose several method to predict the traffic and communication resource. At last, the research designed a platform software of data mining based on Hadoop on traffic and user behavior of wireless network, aiming to analyze4cities’current network data with Hadoop’s big data processing ability, and show operating strategies of what net planning can do.