The mass data mining research based on the information platform of Internet of Things

摘要

This paper analyzes the key technologies of the distributed data, and proposes the solution of the mass data processing based on the Information platform of Internet of Things. The solution uses Hadoop as the open source framework to realize the distributed computing system, and uses Mahout as the data mining algorithm library to realize the parallelization of k-means clustering algorithm. This will achieve the high efficiency and the large expansibility through the mass data processing. The mass data source used in this project is from the intelligent agricultural Information service platform. The system takes the deployment test on the mass sensing information, and it optimizes and parallelizes the K-means algorithm to realize the mass data processing based on the Information platform of Internet of Things. It can make the statistical analysis, and provide fine management and other services. the algorithm is used to improve the efficiency and the accuracy of the platform with the supercomputing and the reliable storage ability.