Classification Technology and Application of Big Data based on Mathematical Model 

 Kai Lv

Jiaxing Vocational Technical College, Jiaxing, 314036, China 

Abstract: Under the condition of large-scale data input and output, it is necessary to classify massive data to improve the ability of data access and scheduling. A classification technique of big data based on mathematical model is proposed. The mathematical model of probability and statistics of big data classification is constructed, the probability density functional of classification objective function is carried out in infinite dimensional vector space, and the confidence interval is constructed in the geometric neighborhood of data clustering center. The stable error approximation analysis is carried out in singular semi-positive infinite-dimensional vector space, the Taylor series expansion is performed at the stable point of big data classification, the mathematical model of big data classification is constructed by binomial Poisson model, and the mathematical model of big data classification is constructed in online subspace. Big data classification is realized in invariant convex combination model by stability control of equilibrium point. The simulation results show that this method has good accuracy and low error rate in big data classification.  

Keywords:Mathematical model; Big data classification; Probability and statistics; Subspace