There are many opportunities for students to perform research with the faculty. Graduate students perform either one- or two-semester research under the supervision of a faculty mentor. The results of the research are often published in regional and national conferences such the yearly Sigma Xi conference that is held at Saint Joseph's university every April.

Jing Zhou, Saint Joseph's University, Mentor: Dr. Babak Forouraghi

Abstract: The project implements a dynamic supply chain operation for the online shopping business by using web services. An extension model of web services, migrating web service, is used to ensure the capability of the client/server connection. The migrating web service provides more flexibility to client-server communication and improves the web service reliability.

Baladitya Voleti, Saint Joseph's University, Mentor: Dr. Stephen Cooper

Abstract: In this work, we examine design patterns as they apply to real (academic) software engineering projects. We attempt to answer the question whether design patterns are useful in the software engineering process. If they are, in which process models do they seem most effective and where specifically consideration of design patterns is best used. We look at the decorator, singleton, adapter and the proxy design patterns and how they could have been used in two software engineering (course) projects: a web crawler and a project involving graphical display of information.

Preethi Ramaswamy, Saint Joseph's University, Mentor: Dr. Babak Forouraghi

Abstract: Web services are simple, self contained applications which perform functions, from simple requests to complicated business processes. The Web Services model of computation uses WSDL, UDDI (Universal Description & Discovery Integration) and SOAP/XML. The UDDI registry is like a phone directory to look up a web service. The WSDL description retrieved from the UDDI directory is used to access the web service. WSDL (Web Services Description Language) allows the software systems of one business to extend to use those of the other directly. The services are invoked over the World Wide Web using the SOAP (Simple Object Access Protocol)/XML protocol. Each of the components is XML based.

Jing Zhou, Saint Joseph's University, Mentor: Dr. Babak Forouraghi

Abstract: Today, many businesses are realizing the value of delivering business intelligence via the Internet. Business intelligence is a broad category of applications and technologies to make better business decisions. Web service is the future of business intelligence for global enterprise analysis and inter-company collaboration. The enterprises use web services to make connections between businesses partners, suppliers and customers.

This project shows how to create web services with different languages from different systems for business use, and how web services transform information sharing between business applications. The project also includes an e-commercial application taking advantage of web services facilities to do the businesses selling on the web. This project proves web services' interoperability.

Ye Guo, Saint Joseph's University, Mentor: Dr. Jonathan Hodgson

Abstract: The aim of this project is to implement a dynamic clustering system that reorganizes froogle.s search results. Froogle is a price match system provided by google. Our clustering system retrieves the price information from froogle with respect to the keywords given by users. We then use Weka tool kit to dynamically cluster the search results, making the output more accessible to users.

Jimil Patel, Saint Joseph's University, Mentor: Dr. Babak Forouraghi

Abstract: Clustering is partitioning of a data set into subsets (clusters), so that the data in each subset share some common trait. We proposed an approach to identify structurally similar XML documents and cluster them. Clustering XML documents has applications in filtering results of web searches, identifying hidden patterns, data mining and bioinformatics. XML documents are tree like structure so clustering XML documents by structure is closely related to Tree Clustering Problem. We first reduce the dimensionality of the XML documents by Principal Component Analysis. Then we use Tree-Edit distance algorithm that estimates similarity in terms of hierarchical relationships of their nodes and apply K-Means clustering algorithm to cluster actual XML documents.

MAbhineet Mittal, Saint Joseph's University, Mentor: Dr. Babak Forouraghi

Abstract: The world of Internet has been experiencing an exponential growth of discussion boards lately, which involve discussion threads ranging from design of cars and engines to analytical chemistry, particle physics, music and so on. It is often the case that users of these discussion boards easily become lost amid such multitudes of information. The purpose of this research project was to incorporate data mining into operation of a web-based discussion board in order to help users better focus their search in their areas of interest.

Mohit Mathur, Saint Joseph's University, Mentor: Dr. Babak Forouraghi

Abstract: Key-Exchange between different clients active on a network is one of the most important features which if implemented incorrectly can lead to serious security loop-holes. In this talk, I'll discuss the mechanism of key exchange using a public key encryption mechanism. Rather than just sending the encryption keys on the network, they are first encrypted using an efficient algorithm and then transferred over to the network clients. I'll discuss the transfer of files/messages between clients on a wireless network using AES encryption, and the use of RSA for its efficient key-exchange.

Sravanthi Matta, Saint Joseph's University, Mentor: Dr. Babak Forouraghi

Abstract: In this talk we introduce a distributed algorithm, called the Ant Colony System (ACS) which is applied to the Traveling Salesperson Problem (TSP). The ACS is a heuristic procedure inspired by the workings of real ants which cooperate with each other using indirect means of pheromone trails built on the edges connecting cities to find good solutions to TSP.

Patricia Hasson, Saint Joseph's University, Mentor: Dr. Stephen Cooper

Abstract: Students often have a great deal of difficulty specifying their requirements as part of their team project in the software engineering class. Omissions, errors, and vagueness all lead to students having a much harder time in design, code, and test when these problems surface. We wanted to know whether using Z, to help teams further refine their specifications, would produce better specifications. We performed a case study of three different types of student projects, to help determine the advantages/disadvantages of using Z for the requirements specification. We found certain decided advantages for some types of projects, and some significant drawbacks for others. This work has been submitted to the 17th Conference on Software Engineering Education and Training (CSEE&T 2004).