Title: The evolution model for big data storage structure of online learning behaviour based on parallel algorithm

Authors: Ying Zhou

Addresses: School of Information and Technology, Nantong Normal College, Nantong 226001, China

Abstract: In order to overcome the heavy task of big data storage structure evolution computation, this paper proposes a parallel algorithm based network learning behaviour big data storage structure evolution model. This method introduces parallel algorithm, divides the whole dataset into several non overlapping data subsets randomly, mines the local frequent itemsets in the network learning behaviour big data in parallel and hierarchically, and connects the local frequent itemsets. Frequent itemsets can get all candidate sets. The actual support degree of different candidate sets is calculated by scanning datasets, and the evolution model of big data storage structure of network learning behaviour is established. The experimental results show that the operation efficiency of the proposed evolutionary model is as high as 99%, the cost is significantly lower than the other three evolutionary models, and the storage space consumption is the lowest.

Keywords: online learning behaviour; big data storage structure; evolution model; parallel algorithm; candidate sets; data subsets randomly.

DOI: 10.1504/IJAACS.2020.112602

International Journal of Autonomous and Adaptive Communications Systems, 2020 Vol.13 No.4, pp.431 - 447

Received: 22 Nov 2019
Accepted: 06 May 2020

Published online: 22 Jan 2021 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article