A Survey for Mobility Big Data Analytics for Geolocation Prediction

Guangxia Xu, Shiyi Gao, Mahmoud Daneshmand, Chonggang Wang, Yanbing Liu

Research output: Contribution to journalArticlepeer-review

50 Scopus citations

Abstract

Geolocation prediction (GP) can be applied to geolocation-based services (GBS), which could provide future services for application users and expand its field of application. Typical geolocation prediction schemes include Markov- based and Bayesian network-based methods. Emerging mobility big data (MBD) poses new challenges and opportunities for geolocation prediction. Because of the diversity of geolocation data, geolocation prediction can be divided into two primary parts: the mining popular geolocation region (MPGR), which is the first step in preprocessing geolocation data when building a geolocation prediction model (GPM); and mining personal trajectory (MPT), which is the second step in building a geolocation prediction model. This article aims to survey existing solutions for geolocation prediction in the era of mobility big data. It first introduces the concepts, classifications, and characteristics of geolocation prediction. Then it describes the basic principles and characteristics of mining popular geolocation regions and mining personal trajectory. This article also discusses challenges, opportunities, and future directions of mobility data analytics for geolocation prediction.

Original languageEnglish
Article number7731599
Pages (from-to)111-119
Number of pages9
JournalIEEE Wireless Communications
Volume24
Issue number1
DOIs
StatePublished - Feb 2017

Fingerprint

Dive into the research topics of 'A Survey for Mobility Big Data Analytics for Geolocation Prediction'. Together they form a unique fingerprint.

Cite this