Advances in Computer Science and Information Technology (ACSIT) Print ISSN: 2393-9907; Online ISSN: 2393-9915; Volume 1, Number 1; October, 2014 pp. 11-16 © Krishi Sanskriti Publications http://www.krishisanskriti.org/acsit.html Multiagent System for Admissions in Academic Institution Puneet Modgil 1 , M. Syamala Devi 2 1 Research Scholar, Department of Computer Science and Applications, Panjab University, Chandigarh, India punmod@rediffmail.com 2 Professor, Department of Computer Science and Applications, Panjab University, Chandigarh, India syamala@pu.ac.in Abstract: Admission of students is an important activity of any academic institution along with many other activities like conducting examinations, result preparation and course scheduling. All these activities are time consuming, distributed and involve number of knowledge intensive sub-activities. If these knowledge intensive sub-activities are automated using agents, the entire work can be efficiently performed without much human intervention. An Agent is a software entity that can automatically perceive environment and can behave reactively as well as proactively. A Multiagent System is distributive artificial intelligence system in which a number of agents interact with each other to solve problems cooperatively to achieve the desired objectives. In this paper a multiagent system for admission activity called Multiagent System for Admissions (MASA) is designed and developed. In MASA, the knowledge intensive sub-activities of admission such as filling the admission form of admission seeker with very little information given by seeker whereas retrieval of rest of information from relevant sources; and passing of information about admitted students to the external academic system are automated by means of agents. MASA is analyzed and designed using agenttool3 which is supported by Organization based Multiagent System Engineering (O-MaSE) framework. MASA is implemented using Java Agent Development Environment (JADE). The interface of entire system to end users is provided using Java Enterprise Edition 7 technologies. It is concluded that Multiagent Systems can automate number of activities which otherwise are performed by humans only, and improve system performance with respect to time and accuracy. 1. INTRODUCTION An academic institution consists of many activities which are distributed and knowledge intensive in nature like admission of students, course scheduling, exam scheduling and result preparation. Admission of students is one of the indispensable activities of any academic institution especially in courses where admission is done on merit basis, a knowledge intensive and distributed activity. The first sub-activity in admission activity is filling of forms by admission seekers. Secondly, all the forms are scrutinized for checking the details filled in by seekers. Thirdly, additional weightage is assigned to forms by scrutinizer based on the course desired and previous record of the seeker. Then scrutinizer creates the merit list with respect to percentage of seeker and weightage assigned. This merit list is displayed to admission seekers and given to admission coordinators. After that on a pre-fixed day, interview of admission seekers is taken by admission coordinator according to merit list and seats are allocated. Lastly, information of all the admitted student is sent to concerned authorities like university to which institution is affiliated. 1.1 Multiagent system An Agent is a software entity that can automatically perceive environment and can behave reactively as well as proactively. The system where a number of agents interact with each other to achieve some common goal is called Multiagent System (MAS). More specifically Multiagent System is distributive artificial intelligence system in which a number of agents interact with each other to solve problems cooperatively to achieve the desired objectives. Agents use knowledge based reasoning approaches for solving problems assigned to them. Autonomy, reactivity, pro-activeness and social ability are important agent properties. The emphasis of work in Multiagent System is on behavior co-ordination amongst agents. A global solution arises from the co-operation between agents. Thus agents are well-suited for solving the problems which are knowledge intensive and distributed in nature. This paper defines how various sub-activities of admission activity that are knowledge intensive, are automated with help of agents, thus producing a Multiagent System. In section 2, related research is discussed. Section 3 explains created Multiagent System for Admissions (MASA) in context of its functionality, process used to create it along with its analysis and design. Implementation is detailed in section 4. Section 5 provides conclusion and scope of future work. 2. RELATED WORK Many systems and their activities are automated with the help of Agent-based Multiagent systems like I-MINDS: A Multiagent System for Intelligent Computer- Supported