Research Summary

During my academic tenure at Stevens Institute of Technology I worked with Dr. Jeffery Nickerson , Dr. Michael zur Muehlen. The research focus of the Agents Research Center at Stevens-Tech was on various Agent Technologies, Human-Agent interaction, and their applications to the business application. I started working with this team during the Fall of 2002 as a Research Assistant. My responsibilities included: Assisting my supervisors in their routine research, work on special research projects, etc.As a contribution to the research efforts of the Team during Fall 2002, I had synthesized the research history in Intelligent Agents area, in the form of a Time Line (See below for a link to the timeline).

For "Agent Research History Time Line" in Graphical format : Click Here

In the Fall 2003, however, I transfered to University of Maryland at Baltimore County as a doctoral student. I am involved in Intelligent Systems including Agents related research under the supervision of Dr.Victoria Yoon at the Information Systems Department of UMBC. Our research focus is to study the application of Intelligent Systems to various business scenarios/problems.

Based on this motivation and experience I have been able to expand the array of my research interests into following and also have been successful in turning my efforts in successful publications.

Research Interests

• Application of Intelligent Agent Technology to Various Business Problems (eBusiness, CRM, etc.)
• Application of Intelligent Agent Technology to Knowledge Management
• Context-Aware Computing with Special focus on Application of Agent Technology
• Deception Detection with Special focus on Agent Societies
• Agent-User Interactions from information processing standpoint
• CRM Systems, Architectures and Applications

Publications/Projects

FALL 2003

On Formal Role of Agents in eBusiness Environment. IS 760 Research Paper. (for a copy Contact)

Abstract: This paper is a literature review paper on application of Intelligent Agents in eBusiness Environments with a special focus on Customer Relationship Management Systems. The paper reveiws characteristics, benefits and challenges related to Intelligent Agents in business environment. It also reviews the literature that covers the communication amongst these agents as well as their communication with humans. It is an effort to develop a comprehensive understanding on the subject matter at hand. Although, I have tried to develop a comprehensive perspective by no means this paper is conclusively comprehensive overview of all the literature available in this domain.

Adaptive Interfaces and Intelligent Agents. IS 760 Seminar. (for a copy Contact)

Abstract: This presentation is a brief overview of major aspects related with Intelligent Agents and Adaptive Interfaces. I have briefly discussed characteristics, benefits, roles, nomanclature, challenges, communication and such other aspects related to Agent Technologies.

 

SPRING 2004

Vyas, A. & Ozok, A. (2005). Usability of Agent Technology in m-Commerce. Proceedings of HCI International Conferences, Las Vegas, USA, Aug. 21-27. (for a copy Contact)

Abstract: The prospects of m-Commerce have been touted since Internet’s inception in early-1990s but most of them have met with a comprehensive disappointment statement by users and other critics, considering several limitations of mobile devices, users, and m-Commerce paradigm as a whole. Many researchers have proposed a software-agent based model for enhancing users’ performance using mobile devices, performing commercial activities with existing spectrum of challenges prevalent in m-Commerce domain. Although, these efforts are not exempt from further challenges. One of the major challenges faced by these researchers has been to verify Agents’ usability in m-Commerce. In this paper, we propose an agent framework based on a few frameworks proposed earlier and then identify various usability aspects related to application of agent technology to m-Commerce domain. We believe that addressing various usability aspects will not only make the Agent Technology Systems more promising for m-Commerce adoption but will also help minimize the user disappointments and consequent apathy towards m-Commerce.

Gender Differences in Accessing Internet from Home & Offices in USA. IS 804 Research Paper. (for a copy Contact)

Abstract: Ever since its conception Internet has been the source for various interesting studies; including studies on Internet Usage and several factors (demographic) impacting the usage. As the Internet gets popular the vision of such studies get more and more clear. This study reviews demographic differences in Internet Usage at workplace or at home as a function of Age, Race and Sex. The data used by this study was gathered by Pew in December 2002. The study performed here tests the overall premise that Internet Usage at work or at home will be different in terms of Age, Race and Sex of the internet users. Most of these hypotheses are rejected with noted difference using Ordinal Logistics Regression Analysis as a procedure in SAS statistical tool. Based on the limited scope of the study we then recommend to extend the study in a specific direction to generate future research insight regarding Internet Usage in U.S. society.

 

FALL 2004

Vyas, A. & Yoon, V. (2004). Location-Aware Intelligent Agent Syste (LIA) for Publish/Subscribe Middleware in Mobile Environments. Proceedings of the 3rd Workshop on e-Business (Web '04), Washington DC, pp. 75-81. (for a copy Contact)

Abstract: The number of information consumers on the move is increasing rapidly following the popularity of mobile devices and subsequent technological advancements. Proposing a robust system for dynamic content dissemination to these mobile devices transpires to be of prime research interest. Making such dissemination location-aware as well as intelligent is even more challenging yet rewarding. Our work focuses on creating a distributed infrastructure suitable to support such scalable content dissemination. We offer a formal framework, Location-Aware Intelligent Agent System (LIA) in integration with Publish/Subscribe middleware to comprehensively address dynamic content dissemination and related issues. We discuss the operational form of our framework in terms of PUSH and PULL strategies. There are two main features of LIA: First, the framework is scalable and adaptable without compromising information security or network organization. Second, it maintains participant anonymity with effective information channeling.

Vyas, A. & Ozok, A. (2005). Intelligent Human-Agent Interaction: Attributes That Need To Be Honed. Proceedings of HCI International Conferences, Las Vegas, USA, Aug. 21-27. (for a copy Contact)

Abstract: Use of intelligent and interactive systems in daily human life has been a favorite theme in science fiction. With advent of intelligent agent systems along with advances in computing and communication technologies, those fictions are beginning to appear almost feasible. Additionally, developing systems that provide comprehensive automation as opposed to developing new metaphors and tools that enhance users’ abilities to directly manipulate objects of a system seems more plausible when system complexities and information overload are on a steep rise. Intelligent agent systems are claimed to provide such comprehensive automation. However, for these agent systems to gain ground and prove to be of real use to the users, specific attention needs to be paid to the user-agent interaction related issues. Theoretically, Agent Systems often make assumptions about their environments, specifically, in terms of predictability of action effects, user motivations, static behavior of the environment and so forth. There is a striking correspondence between these assumptions and the properties of their communication interfaces. In this paper, we highlight some important issue/challenges pertaining to such interaction between users and intelligent agent systems.

 

SPRING 2005

Vyas, A. & Zhou, L. (2005). On Detecting Deception in Agent Societies. Proceedings of the 2005 IEEE/WIC/ACM International Conference on Intelligent Agent Technology (IAT '05), Compeigne, France, Sept. 19-22, pp. 491-494. (for a copy Contact)

Abstract: Driven by the advances in intelligent agent technologies, agent societies have taken on prominent roles in various sciences and applications. Research on risks posed to such societies such as deception by malicious agents has recently begun gaining ground. Drawing on the deception literature from social sciences and artificial intelligence, we propose a deception model which extends and integrates the past research on deception from multiple disciplines. In order to facilitate deception detection, we also develop a procedural model for deception detection. Application of both models to a real-world case scenario demonstrates that both models can be used to instill a better understanding of deception and to detect and remedy deception effectively.

 

OTHER PUBLICATIONS

Vyas, A. & Yoon, V. (2005). Information Management in Mobile Environments Using Location-Aware Intelligent Agent System (LIA). In Handbook of Research in Mobile Business: Technical, Methodological and Social Perspectives. Ed. Bhuvan Unhelkar, Idea Group Publications, In print.

Abstract: Recent rise in the level of comfort and demand to access various types of information using mobile devices can be attributed to the advancements in Wireless as well as Internet technologies, leading us to the new era of Mobile Computing. Location-Based Services (LBS) are engendering new passion in mobile services utilizing users’ location information. Such spatio-temporal information processing entails the need for a dynamic middleware that accurately identifies changing user location and attaches dependent content in real-time without putting extra burden on users. Our work focuses on creating a distributed infrastructure suitable to support such scalable content dissemination. We offer a conceptual framework, Location-Aware Intelligent Agent System (LIA) in integration with Publish/Subscribe middleware to comprehensively address dynamic content dissemination and related issues. We discuss the operational form of our framework in terms of PUSH and PULL strategies.