Senior Lecturer and researcher for more than 10 years in the field of ICT for Rural Development (ICT4RD). Formerly was an IT project manager handling the projects at ISITI, UNIMAS, which include the multi-award winning eBario project, eLamai, eBakelalan, eBuayan,eLarapan and the Telecentre Program for the Orang Asli (TPOA) at Pahang and Kelantan. Besides looking after the sites, he also one of the pioneer in providing training to the PI1M managers and assistant managers for Maxis. He is a Human Resources Development Fund (HRDF) Malaysia certified Train the Trainers (certificate no: TTT/13861) and also a certified CyberSec First Responderâ„¢ (CFR) (certificate no: 2017001338). He had successfully organised and participated many Makerfest/Hackathon event in his university and other universities, which involved the school children from the lower secondary schools to university undergraduates. Currently, he is actively involved in research and consultation on project that involved ICT4RD, renewable energy, service learning and Broadband over Power Line (BPL) technology in the central region of Sarawak. He is also involved heavily in lecturing at University College of Technology Sarawak (UCTS) located at Sibu from network computing to cyber security courses. Besides commitment to the education, research and consultation, he is also the caretaker for the Smart Solar project which is a joint venture with a Korean company to support Sarawak state government in rural electrification project. He also involve with the SHARE project in collaboration with TTC (Japan) and UNIMAS to explore the oneM2M Internet of Things (IoT) Azure cloud based platform for research on possible existing IoT systems integration.
IEEE International Conference on Smart Sensors and Application (ICSSA), 2015
In this paper, a Neural Network Deployment (NND) algorithm is presented to realize and synthesize... more In this paper, a Neural Network Deployment (NND) algorithm is presented to realize and synthesize Multi-Valued Logic (MVL) functions. The algorithm is combined with back-propagation learning capability and MVL operators. The operators are used to synthesize the functions. Consequently the synthesized expressions are applied by the MVL neural operators. The advantages of NND-MVL algorithm are demonstrated by accuracy measurement of MVL neural operator realization. Furthermore, evaluation of NND-MVL algorithm is analyzed by its application, propagation delay and accuracy achieved for training with 4 hidden neurons. In a brief, an effort of training MVL neural operators and utilizing them for logic synthesis is observed.
As described by Oviatt et al.(1997), when two or more modalities work together, the
integration t... more As described by Oviatt et al.(1997), when two or more modalities work together, the integration techniques used for combining different modalities into a whole system is very important. The integration techniques are the main determinants in guiding the design of the multimodal system. To resolve a map-based multimodal system, we propose to use natural language processing (NLP) to identify and describe the syntax and the semantics of prepositions within the spoken speech, in particular, when it relates to maps or directions.
The purpose of this research is to improve the multimodal integration technique, of multiple spee... more The purpose of this research is to improve the multimodal integration technique, of multiple speech and sketch inputs in human computer interaction. In our technique, we integrated the sketch and speech representation model, taking into consideration the normalised relative distances between objects in the sketch, spatial preposition and the deictic words from the speech. We compared the proposed integration technique with the integration technique using only spatial information. The results showed an increase in the matching of correct pairs of spoken and sketched inputs, by 11%, that is, from 52% to 63%. The improved integration technique has shown that the probability of matching the wrong pair of spoken and sketched inputs can be reduced.
IEEE International Conference on Smart Sensors and Application (ICSSA), 2015
In this paper, a Neural Network Deployment (NND) algorithm is presented to realize and synthesize... more In this paper, a Neural Network Deployment (NND) algorithm is presented to realize and synthesize Multi-Valued Logic (MVL) functions. The algorithm is combined with back-propagation learning capability and MVL operators. The operators are used to synthesize the functions. Consequently the synthesized expressions are applied by the MVL neural operators. The advantages of NND-MVL algorithm are demonstrated by accuracy measurement of MVL neural operator realization. Furthermore, evaluation of NND-MVL algorithm is analyzed by its application, propagation delay and accuracy achieved for training with 4 hidden neurons. In a brief, an effort of training MVL neural operators and utilizing them for logic synthesis is observed.
As described by Oviatt et al.(1997), when two or more modalities work together, the
integration t... more As described by Oviatt et al.(1997), when two or more modalities work together, the integration techniques used for combining different modalities into a whole system is very important. The integration techniques are the main determinants in guiding the design of the multimodal system. To resolve a map-based multimodal system, we propose to use natural language processing (NLP) to identify and describe the syntax and the semantics of prepositions within the spoken speech, in particular, when it relates to maps or directions.
The purpose of this research is to improve the multimodal integration technique, of multiple spee... more The purpose of this research is to improve the multimodal integration technique, of multiple speech and sketch inputs in human computer interaction. In our technique, we integrated the sketch and speech representation model, taking into consideration the normalised relative distances between objects in the sketch, spatial preposition and the deictic words from the speech. We compared the proposed integration technique with the integration technique using only spatial information. The results showed an increase in the matching of correct pairs of spoken and sketched inputs, by 11%, that is, from 52% to 63%. The improved integration technique has shown that the probability of matching the wrong pair of spoken and sketched inputs can be reduced.
Uploads
integration techniques used for combining different modalities into a whole system is very
important. The integration techniques are the main determinants in guiding the design of the
multimodal system. To resolve a map-based multimodal system, we propose to use natural
language processing (NLP) to identify and describe the syntax and the semantics of
prepositions within the spoken speech, in particular, when it relates to maps or directions.
integration techniques used for combining different modalities into a whole system is very
important. The integration techniques are the main determinants in guiding the design of the
multimodal system. To resolve a map-based multimodal system, we propose to use natural
language processing (NLP) to identify and describe the syntax and the semantics of
prepositions within the spoken speech, in particular, when it relates to maps or directions.