Simulation and Analysis of Classification Optimization model of Temperature Sensing Big Data in Intelligent Building

The temperature sensor network in intelligent building classified collection of big data processing has the problem of big data redundancy interference, which results in unable to determine the fixed filter thresholds. This paper proposed Chaos differential disturbance based fuzzy C-means clustering model for big temperature sensing data classification tasks. It requires to analyze temperature sensor in the intelligent building big distributed structure model of data in a database storage system, the big data information flow feature fusion and time series analysis. Based on traditional fuzzy c-means clustering processing, we introduced chaos disturbance to avoid the classification into local convergence and local optimum, and therefore improve the performance of data clustering. The testing results show that our proposed classification method effectively reduces the error rate for classification tasks of temperature data in intelligent building and have achieved the best performance among the existing algorithms.