AGENT MODEL DEVELOPMENT BY USING ONTOLOGY IN UNIVERSITY'S ARCHIVING

• Published in: Journal of Theoretical and Applied Information Technology Vol. 87; no. 3; p. 501
• Main Authors: Hasbi, Muhammad, Wardoyo, Retantyo, Istiyanto, Jazi Eko, Mustofa, Khabib
• Format: Journal Article
• Language: English
• Published: Islamabad Journal of Theoretical and Applied Information 01.05.2016
• Subjects: Archiving; Classification; Electronic mail; Email; Information technology; Knowledge representation; Mathematical models; Query processing
• ISSN: 1817-3195

Email classification, based on its subject targetted to the chief. The email should be targetted to the other unit appropriate with its email content. This research is limited with the email domain at university and aimed at making agent model using ontology classify emails based on their subject and content in the university's archiving process. The use of agent to classify emails is done in order the process run autonomously. However, using agent with ontology to classify emails based on their subject and content is still becoming something interesting to study since there is no such research done before. Hence, this research is proposed to develop models to classify emails based on content by using agent with ontology to overcome the problem of archiving. The agent used is the multi-agent which is reactive and proactive that can work autonomously. These agents classify emails and dispose them by ontology to certain unit in the university. The email classification method used is the Naive Bayesian method. The classification is based on the email's content to determine the email's topic which is the keyword of the unit's job description in the university. The keyword then is used to enquery it to the ontology data. The query is aimed at determining the email's target unit in the university. The ontology is the representative of the knowledge of the Unit's job description in the organization structure of a university. The use of ontology can result in the disposition target distributedly in the university's units appropriate with its email subject and content. This model can be run precisely in determining the email's disposition target to a certain unit in the university.