A Cumulative Timeline Feature Selection Approach for Data Classification

Research output: Book chapter/Published conference paperConference paperpeer-review

Abstract

This paper presents a novel feature vector generation technique of malware data which retains high classification accuracy over an extended time period. The proposed approach is to combine the features and accumulating these features with intervals over time. Experimental results show that the proposed method maintains constant classification accuracy and with a standard deviation of 0.92 over the extended time period. These results strongly support the hypothesis that it is possible to develop any classification strategy that will work well into the future.
Original languageEnglish
Title of host publicationProceedings of the 1st International Conference on Machine Learning and Data Engineering
Pages2-6
Number of pages5
ISBN (Electronic)9780648014737
Publication statusPublished - 2017
Event1st International Conference on Machine Learning and Data Engineering: iCMLDE 2017 - Western Sydney University, Parramatta, Sydney, Australia
Duration: 20 Nov 201722 Nov 2017
http://2017.icmlde.net.au/Home.aspx (Conference website)
http://2017.icmlde.net.au/Proceedings.aspx (Conference proceedings)

Conference

Conference1st International Conference on Machine Learning and Data Engineering
CountryAustralia
CityParramatta, Sydney
Period20/11/1722/11/17
Other1st International Conference on Machine Learning and Data Engineering, Sydney, 2017 (iCMLDE 2017) will bring international and national experts in Artificial Intelligence, Computer Vision, Machine Learning and Data Engineering researcher to Sydney to present and share their latest research and practice. iCMLDE 2017 will be held in Sydney, which is one of the largest cities in Australia. The conference is supported by Global Circle for Scientific, Technological and Management Research (GCSTMR) and Science, Technology and Management Crest, Australia (STAMCA). GCSTMR endeavors to create a platform for young researchers to exchange ideas and share information with experts in their respective fields. GCSTMR encourages and facilitates trans-national collaborations among its members and associates in research, training and professional activities. Multidisciplinary international conference series run by GCSTMR open vast opportunities for researchers to interact on the global platform and take advantage of knowledge sharing in the digital arena.

iCMLDE 2017 will be held at Western Sydney University’s Parramatta campus, which is situated 20km from the Sydney CBD and is well connected by rail, road and ferry networks. The traditional landmarks of Sydney including Sydney Harbour, Sydney Opera House and Harbour Bridge are within easy reach of the conference venue as is the Blue Mountains World Heritage National Park.
Internet address

Fingerprint Dive into the research topics of 'A Cumulative Timeline Feature Selection Approach for Data Classification'. Together they form a unique fingerprint.

Cite this