UM  > 科技學院
Improvised Methods for Tackling Big Data Stream Mining Challenges: Case Study of Human Activity Recognition
Fong, Simon1; Liu, Kexing1; Cho, Kyungeun2; Wong, Raymond3; Mohammed, Sabah4; Fiaidhi, Jinan4
2016-10
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
DOIhttp://doi.org/10.1007/s11227-016-1639-5
Indexed BySCI
Language英语
WOS Research AreaComputer Science ; Engineering
WOS SubjectComputer Science, Hardware & Architecture ; Computer Science, Theory & Methods ; Engineering, Electrical & Electronic
WOS IDWOS:000385417400014
Fulltext Access
Citation statistics
Document TypeJournal article
CollectionFaculty of Science and Technology
DEPARTMENT OF COMPUTER AND INFORMATION SCIENCE
Affiliation1.Univ Macau, Dept Comp & Informat Sci, Zhuhai, Macau, Peoples R China
2.Dongguk Univ, Dept Multimedia Engn, Seoul, South Korea
3.Univ New South Wales, Sch Comp Sci & Engn, Sydney, NSW, Australia
4.Lakehead Univ, Dept Comp Sci, Thunder Bay, ON, Canada
First Author AffilicationUniversity of Macau
Recommended Citation
GB/T 7714
Fong, Simon,Liu, Kexing,Cho, Kyungeun,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 Fong, Simon,Liu, Kexing,Cho, Kyungeun,Wong, Raymond,Mohammed, Sabah,&Fiaidhi, Jinan.(2016).Improvised Methods for Tackling Big Data Stream Mining Challenges: Case Study of Human Activity Recognition.JOURNAL OF SUPERCOMPUTING,72(10),3927-3959.
MLA Fong, Simon,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.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Fong, Simon]'s Articles
[Liu, Kexing]'s Articles
[Cho, Kyungeun]'s Articles
Baidu academic
Similar articles in Baidu academic
[Fong, Simon]'s Articles
[Liu, Kexing]'s Articles
[Cho, Kyungeun]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Fong, Simon]'s Articles
[Liu, Kexing]'s Articles
[Cho, Kyungeun]'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.