UPM Institutional Repository

Comparison of hidden Markov Model and Naïve Bayes algorithms among events in smart home environment


Babakura, Abba and Sulaiman, Md Nasir and Mustapha, Norwati and Kasmiran, Khairul A. (2014) Comparison of hidden Markov Model and Naïve Bayes algorithms among events in smart home environment. In: International Conference of Recent Trends in Information and Communication Technologies 2014, 12-14 Sep. 2014, Universiti Teknologi Malaysia, Johor, Malaysia. (pp. 1-11).


The smart home environment consists of numerous subsystems which are heterogeneous in nature. Smart home environment are configured in such a way that it comfort driven as well as achieving optimized security and task-oriented without human intervention inside the home. The subsystems, due to their diversified nature develop difficulties as the events communicate making the smart home uncomfortable. The complexity of decision making in handling events stands at the bottleneck in ensuring various tasks executed jointly among diversified systems in smart home environment. In this paper, we propose Hidden Markov Model (HMM) and Naïve Bayes (NB) to test the accuracy and response time of the home data and to compare between the two algorithms. The result experimented shows that the HMM algorithm stands at higher accuracy and better response time than the NB. The implementation has been carried out in such a way that quality information is acquired among the systems to demonstrate the effectiveness of decision making among events in the smart home environment.

Download File

[img] PDF
Restricted to Repository staff only

Download (1MB)

Additional Metadata

Item Type: Conference or Workshop Item (Paper)
Divisions: Faculty of Computer Science and Information Technology
Publisher: Univerisiti Teknologi Malaysia
Keywords: Smart home; HMM; Naïve Bayes; Decision making; Feature selection
Depositing User: Nursyafinaz Mohd Noh
Date Deposited: 19 Aug 2015 08:24
Last Modified: 19 Aug 2015 08:24
URI: http://psasir.upm.edu.my/id/eprint/39909
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item