UM  > Faculty of Science and Technology
Status已發表Published
Affiliated with RCfalse
Improvised Methods for Tackling Big Data Stream Mining Challenges: Case Study of Human Activity Recognition
Simon Fong1; Kexing Liu1; Kyungeun Cho2; Raymond Wong3; Sabah Mohammed4; Jinan Fiaidhi4
2016-02-16
Source PublicationJOURNAL OF SUPERCOMPUTING
ISSN0920-8542
Volume72Issue:10Pages:3927-3959
Abstract

Big data stream is a new hype but a practical computational challenge founded on data streams that are prevalent in applications nowadays. It is quite well known that data streams that are originated and collected from monitoring sensors accumulate continuously to a very huge amount making traditional batch-based model induction algorithms infeasible for real-time data mining or just-in-time data analytics. In this position paper, following a new datastream mining methodology, namely stream-based holistic analytics and reasoning in parallel (SHARP), a list of data analytic challenges as well as improvised methods are looked into. In particular, two types of decision tree algorithms, batch-mode and incremental-mode, are put under test at sensor data that represents a typical big data stream. We investigate whether and to what extent of two improvised methods-outlier removal and balancing imbalanced class distributions-affect the prediction performance in big data stream mining. SHARP is founded on incremental learning which does not require all the training to be loaded into the memory. This important fundamental concept needs to be supported not only by the decision tree algorithms, but by the other improvised methods usually at the preprocessing stage as well. This paper sheds some light into this area which is often overlooked by dataanalysts when it comes to big data stream mining.

KeywordData Stream Mining Big Data Very Fast Decision Tree Resampling Sensor Data
DOI10.1007/s11227-016-1639-5
Indexed BySCIE
Language英語English
WOS Research AreaComputer Science ; Engineering
WOS SubjectComputer Science, Hardware & Architecture ; Computer Science, Theory & Methods ; Engineering, Electrical & Electronic
WOS IDWOS:000385417400014
PublisherSPRINGER, VAN GODEWIJCKSTRAAT 30, 3311 GZ DORDRECHT, NETHERLANDS
Fulltext Access
Citation statistics
Cited Times [WOS]:4   [WOS Record]     [Related Records in WOS]
Document TypeJournal article
CollectionFaculty of Science and Technology
DEPARTMENT OF COMPUTER AND INFORMATION SCIENCE
Corresponding AuthorSimon Fong
Affiliation1.Department of Computer and Information Science, University of Macau, Macau, SAR, China
2.Department of Multimedia Engineering, Dongguk University, Seoul, Korea
3.School of Computer Science and Engineering, University of New South Wales, Sydney, Australia
4.Department of Computer Science, Lakehead University, Thunder Bay, Canada
First Author AffilicationUniversity of Macau
Corresponding Author AffilicationUniversity of Macau
Recommended Citation
GB/T 7714
Simon Fong,Kexing Liu,Kyungeun Cho,et al. Improvised Methods for Tackling Big Data Stream Mining Challenges: Case Study of Human Activity Recognition[J]. JOURNAL OF SUPERCOMPUTING,2016,72(10):3927-3959.
APA Simon Fong,Kexing Liu,Kyungeun Cho,Raymond Wong,Sabah Mohammed,&Jinan Fiaidhi.(2016).Improvised Methods for Tackling Big Data Stream Mining Challenges: Case Study of Human Activity Recognition.JOURNAL OF SUPERCOMPUTING,72(10),3927-3959.
MLA Simon Fong,et al."Improvised Methods for Tackling Big Data Stream Mining Challenges: Case Study of Human Activity Recognition".JOURNAL OF SUPERCOMPUTING 72.10(2016):3927-3959.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Simon Fong]'s Articles
[Kexing Liu]'s Articles
[Kyungeun Cho]'s Articles
Baidu academic
Similar articles in Baidu academic
[Simon Fong]'s Articles
[Kexing Liu]'s Articles
[Kyungeun Cho]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Simon Fong]'s Articles
[Kexing Liu]'s Articles
[Kyungeun Cho]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.