Ontology based recommendation system for research communities

• Main Author: Abdul Razak, Siti Hajar
• Format: Thesis
• Language: English
• Published: 2015
• Online Access: http://psasir.upm.edu.my/id/eprint/57117/1/FSKTM%202015%2011RR.pdf; http://psasir.upm.edu.my/id/eprint/57117/

The enhancement of Virtual Community of Practice (VCoP) among research universities (RU) in Malaysia has led to a proper system model for new researchers to ensure that they are joining the right research communities effectively. It is important for researchers to capture and express their expertise in a form that can be easily accessed and used by others. Besides, it has led to devise a proper system model to group researchers in a community with mutual research interest since there is a limitation of system model in finding the right people to work together. The formation of virtual research communities in RU is known as VCoP where it provides a flexible way for researchers to interact, reuse, and share knowledge virtually. In this context,researchers are lecturers, as well as postdoctoral and postgraduate students. Thus, this research had adopted the Quick Ontology Mapping (QOM) as an ontology mapping technique since this technique had been proven to be effective and the applications of ontology are to classify, as well as to model the VCoP. In addition, the aim of this research was to propose an ontology-based model in recommending research groups to new researchers, besides proving the effectiveness of the system model that implemented the ontology mapping technique. The proposed system modelwas constructed based on literature review. Then, a case study was conducted upon the research community at the Faculty of Computer Science and Information Technology (FCSIT) in Universiti Putra Malaysia (UPM). The aim of this case study was to validate the elements of the proposed system model. As a result, the research findings showed that: (i) most of the respondents agreed that there was a need to group the researchers in the same field, and (ii) most of the respondents agreed that the proposed system modelcould recommend suitable research groups based on their interests. Apart from that, in order to prove the effectiveness of the proposed system model, this study compared the ontology mapping with simple matching technique. The purpose of this comparison had been to differentiate the mapping technique with and without ontology for it to function in a more relevant manner based on user’s perspective. With that, a system prototype was developed to transform the proposed system modelinto a more workable state. Then, a post survey was conducted to prove the effectiveness of the system prototype. In addition, the results were analyzed by using Rasch Model since this tool has been proven good in analyzing small sample size. The findings indicated that: 58% of respondents completely agreed (Strongly Agreed and Agreed) with the implementation of ontology mapping technique, and (ii) most of the respondents like the way ontology mapping technique recommended the research group. Hence, this research showed that this system model could be used as a guide to recommend research groups to new researchers. Finally, this proposed system modelcould be used by other communities to group their members with similar knowledge. Indirectly,knowledge can be shared and reused effortlessly.