M. N. Marsono
Faculty of Electrical Engineering, Universiti Teknologi Malaysia

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Proceeding of the Electrical Engineering Computer Science and Informatics

Online Data Stream Learning and Classification with Limited Labels Loo Hui Ru; Trias Andromeda; M. N. Marsono
Proceeding of the Electrical Engineering Computer Science and Informatics Vol 1: EECSI 2014
Publisher : IAES Indonesia Section

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (680.855 KB) | DOI: 10.11591/eecsi.v1.366

Abstract

Mining data streams such as Internet traffic andnetwork security is complex. Due to the difficulty of storage, datastreams analytics need to be done in one scan. This limits thetime to observe stream feature and hence, further complicatesthe data mining processes. Traditional supervised data miningwith batch training natural is not suitable to mine data streams.This paper proposes an algorithm for online data streamclassification and learning with limited labels using selective selftrainingsemi-supervised classification. The experimental resultsshow it is able to achieve up to 99.6% average accuracy for 10%labeled data and 98.6% average accuracy for 1% labeled data. Itcan classify up to 34K instances per second.
Configurable Version Management Hardware Transactional Memory for Multi-processor Platform Jeevan Sirkunan; Chia Yee Ooi; N. Shaikh Husin; Yuan Wen Hau; Trias Andromeda; M. N. Marsono
Proceeding of the Electrical Engineering Computer Science and Informatics Vol 1: EECSI 2014
Publisher : IAES Indonesia Section

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (755.231 KB) | DOI: 10.11591/eecsi.v1.409

Abstract

Programming on a shared memory multi-processor platforms in an efficient way is difficult as locked based synchronization limits the efficiency. Transactional memory (TM) is a promising approach in creating an abstraction layer for multi-threaded programming. However, the performance of TM is application-specific. In general, the configuration of a TM is divided into version management and conflict management. Each scheme has its strengths and weaknesses depending on executing application. Previous TM implementations for embedded system were built on fixed version management configuration which results in significant performance loss when transaction behaviour changes. In this paper, we propose a hardware transactional memory (HTM) with interchangeable version management. Random requests at different contention levels are used to verify the performance of the proposed TM. The proposed architecture is targeted for embedded applications and is area-efficient compared to current implementations that apply cache coherence protocols.