Title: Prediction algorithm based on web mining for multimedia objects in next-generation Digital Earth

Authors: Li Zhu; Shengyong Xu

Addresses: College of Engineering, Huazhong Agricultural University, 430070 Wuhan, Hubei, China ' College of Engineering, Huazhong Agricultural University, 430070 Wuhan, Hubei, China

Abstract: New research questions related to big data have brought Digital Earth into a new data-intensive era. In this paper, we present a prediction algorithm based on data mining especially for multimedia objects in next-generation Digital Earth. We collect the useful data and preprocess them in client logs, mine the sequential pattern between the spatial objects and the multimedia objects, and obtain the association rules. The derived association rules can be used for prefetching some candidates in advance. The diverse experiment results show that our prefetching strategy based on mining web logs achieves higher efficiency than the other general or no prefetching ones.

Keywords: prediction algorithms; multimedia objects; web mining; next-generation Digital Earth; association rules; sequential patterns; embedded systems; big data; prefetching strategy; web logs.

DOI: 10.1504/IJES.2015.066145

International Journal of Embedded Systems, 2015 Vol.7 No.1, pp.79 - 87

Received: 14 Apr 2014
Accepted: 07 May 2014
Published online: 03 Dec 2014 *

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