Pith. sign in

REVIEW

When Machine Learning Meets Big Data: A Wireless Communication Perspective

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1901.08329 v2 pith:OWNPB67U submitted 2019-01-24 cs.NI cs.CY

When Machine Learning Meets Big Data: A Wireless Communication Perspective

classification cs.NI cs.CY
keywords datalearningmachinenetworksrequirementswirelessframeworkpredict
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X LinkedIn Reddit HN
read the original abstract

We have witnessed an exponential growth in commercial data services, which has lead to the 'big data era'. Machine learning, as one of the most promising artificial intelligence tools of analyzing the deluge of data, has been invoked in many research areas both in academia and industry. The aim of this article is twin-fold. Firstly, we briefly review big data analysis and machine learning, along with their potential applications in next-generation wireless networks. The second goal is to invoke big data analysis to predict the requirements of mobile users and to exploit it for improving the performance of "social network-aware wireless". More particularly, a unified big data aided machine learning framework is proposed, which consists of feature extraction, data modeling and prediction/online refinement. The main benefits of the proposed framework are that by relying on big data which reflects both the spectral and other challenging requirements of the users, we can refine the motivation, problem formulations and methodology of powerful machine learning algorithms in the context of wireless networks. In order to characterize the efficiency of the proposed framework, a pair of intelligent practical applications are provided as case studies: 1) To predict the positioning of drone-mounted areal base stations (BSs) according to the specific tele-traffic requirements by gleaning valuable data from social networks. 2) To predict the content caching requirements of BSs according to the users' preferences by mining data from social networks. Finally, open research opportunities are identified for motivating future investigations.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.