Authors: Jinming Zhang
Addresses: Department of Information Technology, Yantai Vocational College, Yantai, Shandong, 264670, China
Abstract: Aiming at the problems of low data recall rate, poor data mining accuracy and poor redundant data interference ability of traditional data mining technology, an adaptive recommendation algorithm for big data mining based on Hadoop platform is proposed. In this algorithm, a distributed storage structure model of big data is constructed on Hadoop cloud platform; the statistical regression analysis method is adopted to construct a big data similarity mathematical model; the autocorrelation matching detection method is adopted to extract correlation features of big data and Bäcklund transform is adopted to decompose time-frequency variation characteristics of the correlation features of big data to design an adaptive recommendation model for big data mining based on Hadoop platform to optimise the performance of big data mining on Hadoop platform. The experimental results show that the accuracy of data mining increases with the number of iterations. Has a strong ability to redundant interference.
Keywords: Hadoop platform; big data; mining; recommendation algorithm; distributed storage.
International Journal of Internet Protocol Technology, 2019 Vol.12 No.4, pp.213 - 220
Received: 12 Sep 2018
Accepted: 29 Oct 2018
Published online: 21 Nov 2019 *