Hybrid classification system design using a decision learning approach and three layered structure

A Meta learning paradigm in Data Mining

Lamogha Ighoroje, Joan Lu, Qiang Xu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A data classification system is designed consisting of three layers. The second layer is the main focus of this research paper. It describes a meta-learning (learning to learn) concept that uses certain characteristics of the dataset as well as some more general knowledge about supervised and unsupervised machine learning algorithms (e.g. supervised learners tend to perform very well in the presence of a large pre-labelled training sets, etc.) to create some hypothesis. The main aim of this research is to harness general knowledge about a dataset and different machine learning methods to develop a set of meta-rules that when implemented will help to automate and speed up big data classification processes in data mining. An experiment is conducted to verify the hypotheses made using supervised and unsupervised knowledge flows in weka with some datasets taken from weka and UCI machine learning repositories. The performance result of the experiments is used to design a meta-learning algorithm in form of rules. The results from the experiments confirmed that general knowledge known about supervised and unsupervised learning is then harnessed successfully for making learning decisions.

Original languageEnglish
Title of host publication25th International Conference on Information Systems Development, ISD 2016
PublisherAssociation for Information Systems
Pages402-412
Number of pages11
ISBN (Electronic)9788378753070
Publication statusPublished - 2016
Event25th International Conference on Information Systems Development: Complexity in Information Systems Development - University of Economics , Katowice, Poland
Duration: 24 Aug 201626 Aug 2016
Conference number: 25
http://aisel.aisnet.org/isd2014/proceedings2016/ (Link to Conference Details)

Conference

Conference25th International Conference on Information Systems Development
Abbreviated titleISD 2016
CountryPoland
CityKatowice
Period24/08/1626/08/16
Internet address

Fingerprint

Data mining
Learning systems
Systems analysis
Learning algorithms
Unsupervised learning
Experiments
Supervised learning
Big data

Cite this

Ighoroje, L., Lu, J., & Xu, Q. (2016). Hybrid classification system design using a decision learning approach and three layered structure: A Meta learning paradigm in Data Mining. In 25th International Conference on Information Systems Development, ISD 2016 (pp. 402-412). Association for Information Systems.
Ighoroje, Lamogha ; Lu, Joan ; Xu, Qiang. / Hybrid classification system design using a decision learning approach and three layered structure : A Meta learning paradigm in Data Mining. 25th International Conference on Information Systems Development, ISD 2016. Association for Information Systems, 2016. pp. 402-412
@inproceedings{56fde265766c4f1e8816c091778b7bd0,
title = "Hybrid classification system design using a decision learning approach and three layered structure: A Meta learning paradigm in Data Mining",
abstract = "A data classification system is designed consisting of three layers. The second layer is the main focus of this research paper. It describes a meta-learning (learning to learn) concept that uses certain characteristics of the dataset as well as some more general knowledge about supervised and unsupervised machine learning algorithms (e.g. supervised learners tend to perform very well in the presence of a large pre-labelled training sets, etc.) to create some hypothesis. The main aim of this research is to harness general knowledge about a dataset and different machine learning methods to develop a set of meta-rules that when implemented will help to automate and speed up big data classification processes in data mining. An experiment is conducted to verify the hypotheses made using supervised and unsupervised knowledge flows in weka with some datasets taken from weka and UCI machine learning repositories. The performance result of the experiments is used to design a meta-learning algorithm in form of rules. The results from the experiments confirmed that general knowledge known about supervised and unsupervised learning is then harnessed successfully for making learning decisions.",
keywords = "Big data classification, Classification, Clustering, Data mining, Learning to learn, Machine learning, Meta learning, Meta-rules, Security classification, Supervised learning, Unsupervised learning",
author = "Lamogha Ighoroje and Joan Lu and Qiang Xu",
year = "2016",
language = "English",
pages = "402--412",
booktitle = "25th International Conference on Information Systems Development, ISD 2016",
publisher = "Association for Information Systems",
address = "United States",

}

Ighoroje, L, Lu, J & Xu, Q 2016, Hybrid classification system design using a decision learning approach and three layered structure: A Meta learning paradigm in Data Mining. in 25th International Conference on Information Systems Development, ISD 2016. Association for Information Systems, pp. 402-412, 25th International Conference on Information Systems Development, Katowice, Poland, 24/08/16.

Hybrid classification system design using a decision learning approach and three layered structure : A Meta learning paradigm in Data Mining. / Ighoroje, Lamogha; Lu, Joan; Xu, Qiang.

25th International Conference on Information Systems Development, ISD 2016. Association for Information Systems, 2016. p. 402-412.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Hybrid classification system design using a decision learning approach and three layered structure

T2 - A Meta learning paradigm in Data Mining

AU - Ighoroje, Lamogha

AU - Lu, Joan

AU - Xu, Qiang

PY - 2016

Y1 - 2016

N2 - A data classification system is designed consisting of three layers. The second layer is the main focus of this research paper. It describes a meta-learning (learning to learn) concept that uses certain characteristics of the dataset as well as some more general knowledge about supervised and unsupervised machine learning algorithms (e.g. supervised learners tend to perform very well in the presence of a large pre-labelled training sets, etc.) to create some hypothesis. The main aim of this research is to harness general knowledge about a dataset and different machine learning methods to develop a set of meta-rules that when implemented will help to automate and speed up big data classification processes in data mining. An experiment is conducted to verify the hypotheses made using supervised and unsupervised knowledge flows in weka with some datasets taken from weka and UCI machine learning repositories. The performance result of the experiments is used to design a meta-learning algorithm in form of rules. The results from the experiments confirmed that general knowledge known about supervised and unsupervised learning is then harnessed successfully for making learning decisions.

AB - A data classification system is designed consisting of three layers. The second layer is the main focus of this research paper. It describes a meta-learning (learning to learn) concept that uses certain characteristics of the dataset as well as some more general knowledge about supervised and unsupervised machine learning algorithms (e.g. supervised learners tend to perform very well in the presence of a large pre-labelled training sets, etc.) to create some hypothesis. The main aim of this research is to harness general knowledge about a dataset and different machine learning methods to develop a set of meta-rules that when implemented will help to automate and speed up big data classification processes in data mining. An experiment is conducted to verify the hypotheses made using supervised and unsupervised knowledge flows in weka with some datasets taken from weka and UCI machine learning repositories. The performance result of the experiments is used to design a meta-learning algorithm in form of rules. The results from the experiments confirmed that general knowledge known about supervised and unsupervised learning is then harnessed successfully for making learning decisions.

KW - Big data classification

KW - Classification

KW - Clustering

KW - Data mining

KW - Learning to learn

KW - Machine learning

KW - Meta learning

KW - Meta-rules

KW - Security classification

KW - Supervised learning

KW - Unsupervised learning

UR - http://www.scopus.com/inward/record.url?scp=84995902577&partnerID=8YFLogxK

M3 - Conference contribution

SP - 402

EP - 412

BT - 25th International Conference on Information Systems Development, ISD 2016

PB - Association for Information Systems

ER -

Ighoroje L, Lu J, Xu Q. Hybrid classification system design using a decision learning approach and three layered structure: A Meta learning paradigm in Data Mining. In 25th International Conference on Information Systems Development, ISD 2016. Association for Information Systems. 2016. p. 402-412