Weak Association Mining Algorithm for Long Distance Wireless Hybrid Transmission Data in Cloud Computing

Abstract

Long distance wireless hybrid transmission data is vulnerable to noise, resulting in low data mining accuracy, large mining error and poor mining effect. Therefore, a weak association mining algorithm for remote wireless hybrid transmission data under cloud computing is proposed. The moving average method is used to eliminate noise data, and the attribute values of continuous data are divided into discrete regions, make it form a unified conversion code for data conversion. The Bayesian estimation method is used for static fusion to eliminate the uncertain data with noise. The rough membership function is constructed to distinguish the truth value, complete data preprocessing. According to the principle of relationship matching between data, data feature decomposition is realized. The non sequential Monte Carlo simulation sampling method is adopted to build the data loss probability evaluation model and integrate the data association rules. In the background of cloud computing, permission item sets are generated, and the rationality of association rules is judged by the minimum support. The dynamic programming principle is used to build the mining model, and the improved DTW algorithm is used to read out and analyze the structured, semi-structured and unstructured data to obtain the weak association mining results of mixed data transmission. The experimental results show that the algorithm can completely mine data sets, and the mining error is less than 0.10, with good mining results.