Learning Abstract Data Types and Inheritance

with Interactive Multimedia

Glenn D. Blank, William M. Pottenger,

Shreeram A. Sahasrabudhe and Shenzhi Li

glenn.blank@lehigh.edu, pottenger@cse.lehigh.edu

Computer Science and Engineering

Lehigh University, Bethlehem, PA 18015

Abstract

We are developing new interactive multimedia for computer science courses, including

introductory CS0/CS1, upper level undergraduate software engineering, and graduate level

object-oriented software engineering (OOSE). For the software engineering courses, we have

developed and evaluated units on abstract data types (ADTs) and inheritance. ADTs formalize

the meaning of classes in connection with object-oriented design by contract (Meyer, 1988). We

have found that students have difficulty mastering this material based on lecture and assignments

alone. Results from evaluation indicate that students learn more from multimedia than lecture.

Another direction for interactive multimedia is inquiry-based learning, where students discover

emerging trends in software engineering.

Introduction

CIMEL is a multimedia framework for Constructive, collaborative Inquiry-based E-

Learning. CIMEL provides a framework for multimedia-enhanced education throughout the

computer science curriculum. Our experimental results confirm that interactive multimedia

courseware can significantly improve learning, both in terms of objective knowledge and task-

oriented skills.

In object-oriented software engineering, abstract data types (ADTs) can bridge a gap

between analysis and implementation of classes. As Meyer (1988), Thomas and Weedon (1995)

and others have shown, ADTs provide a way to describe the semantics of classes identified

during analysis in explicit, language-independent notation. The pre-conditions and post-

conditions of all functions describe a contract between the supplier and clients of each ADT:

clients are responsible for pre-conditions; suppliers are responsible for post-conditions. That way

the programmers supplying an ADT know clearly what they must implement and programmers

using an ADT know clearly how to use it. We think ADTs and design by contract are an

excellent idea, but judging from textbooks, it hasn’t caught on as well as UML for analysis or C+

+ or Java for implementation. Our experience has been that many students, even in the graduate

level course, have difficulty learning how to write ADTs, based on lecture and textbook readings

alone. Our hypothesis was that interactive multimedia, by presenting concepts and facilitating

learning by doing in constructive exercises, could help students understand and use such

material.

Software engineering is a rapidly changing area. To keep up with developments, students

need to develop lifelong learning skills. Another goal of the CIMEL project is to facilitate

inquiry-based learning, where students learn by doing research (and also how to do research).

This aspect of our multimedia framework involves a methodology and tools for discovering

emerging trends in scientific literature. We developed a multimedia tutorial in which students

learn how to discover emerging trends pertaining to inheritance in object-oriented programming.

The User Interface

Figure 1 illustrates several features of the CIMEL user interface:

A TRACK LIST at left displays the content of a lesson as a sequence of screens. The menu

uses check marks to show progress and highlights the current screen in red. An external

XML file maintains a menu of screen titles, facilitating maintenance of course content.

Figure 1: Screen Capture from CIMEL multimedia

Multimedia personae model a diverse community of teachers and learners. The personae

currently include two professors (one shown here), a teaching assistant, a reference

librarian, and three students. In addition to graphical images, they speak in audio and/or

text boxes. Personae model students and instructors studying material together, working

through interactive exercises, and suggest exploratory research on relevant topics using

online information.

The icons at the bottom give learners access to various tools, including the COLLABORATE

and EXPLORE (emerging trends) tools under development for this project, as well as the

BlueJ and JavaEdit programming environments developed elsewhere.

The PREFERENCES icon presents a panel of options letting the user adapt the

environment according to his or her personal learning style, including turning text boxes

or audio on/off, toggling auto-advance or wait for next page, setting the timing rate where

there is no audio narration, volume control etc. A user may change these settings at any

time during these sessions and they will be recorded locally and on a network drive for

the next session.

A JUST THE FACTS mode lets users switch to viewing non-interactive content (text and

graphics) presented in HTML pages. From there, one can switch back to rich media mode

via hyperlinks anchored to the corresponding Flash page. There are also links to

interactive screens, which remain in Flash. (Using Flash, CIMEL can deploy pages via

the web or CD-ROM. The vector-based graphics of Flash makes it easy to display the

user interface in full screen, whatever the particular screen resolution may be.)

The user interface design has been driven by a process of user-driven evaluation, including focus

groups and usability surveys. This process has helped us to design an interface that accommodate

the diverse preferences of students: some enjoying the rich media and others wanting a more

straightforward “just the facts” presentation, many liking both text and audio while others found

either sound or text annoying.

Interactivity and constructive exercises

While our approach is to present enough didactic material in the multimedia that it can be

a standalone learning experience, interactivity is frequent and rich in CIMEL content. Personae

provide feedback to all responses, which are in turn logged to a web-based tracking system for

evaluation purposes. Experimental evidence, focus groups, surveys and student interviews make

it clear student learning is roughly proportional to the richness of interactivity. A design goal of

CIMEL is to facilitate diverse student learning styles to keep the multimedia effectiveness for a

broader audience (Blank, Sahasrabudhe & Heigl, 2002).

Interactive quizzes and constructive exercises help students learn by doing. Personae

provide feedback guiding a student through each exercise. In Figure 2, the TA persona changes

expression and provides feedback hinting at what is wrong with a user choice in a multiple

choice question. Because the multimedia supplies feedback to all user interactions, right or

wrong, students can learn whether or not they know the answers.

Figure 2: TA persona responds to a wrong choice in a multiple-choice question

Constructive exercises are much more complex, challenging a learner to build solutions

to problems by dragging and dropping pieces of structures into place, incrementally. Figure 3 is

a snapshot from a series of screens, in which a professor and learner together gradually build an

abstract data type. In this exercise, the learner goes through a series of screens, incrementally

building up an ADT specification for apples. The learner has already constructed the signatures

section of an Apple ADT by first choosing a member function, then in figure 3 selecting that

function’s arguments. Later on in the exercise, the learner builds the preconditions and

postconditions for the Apple ADT. Finally, the learner runs simulations running the functions,

preconditions and postconditions, testing them for completeness. At each step, feedback helps

the learner learn from mistakes as well as correct actions.

Figure 3: Excerpt from a constructive exercise

Another type of interactive exercise facilitates inquiry-based learning. For example, after

studying ADT for collections, a professor persona asks the learner to investigate the design of

similar ADTs in the Java Development Kit (JDK). The following screen reinforces this activity

by asking a follow-up question, designed to get the learner to reflect on what he or she has

learned from the inquiry-based exercise.

More complex inquiry-based exercises provide students with opportunities to explore the

research literature related to a course, using an emerging trends detection tool. An emerging

trend is a topic area for which one can trace the growth of interest and utility over time. We have

developed a methodology for Emerging Trend Detection based on textual data mining (Roy,

Gevry & Pottenger 2002) (Kontostathis, Galitsky, Pottenger, Roy & Phelps, 2003). A

multimedia tutorial teaches students a methodology for trend detection in the main topic area of

inheritance in object-oriented software. Following the tutorial, an inquiry-based exercise asks

students to search for emerging trends inheritance using the methodology.

Figure 4: Interface for query input

First, students are asked to search recent conferences and workshop proceedings to

identify candidate emerging trends (links are provided in the multimedia). This step requires

domain knowledge that the students have gained previously in the class. Next, the students

employ the tool shown in Figure 4 to search for additional trends or to glean further evidence for

previously discovered candidate trends. This tool enables students to query a popular search

engine and prune and analyze the results returned. As portrayed in Figure 4, a student enters a

candidate trend and/or a main topic area (such as Inheritance), combined with a list of ‘helper

terms’. ‘Helper terms’ are words or phrases that are commonly found in association with

emerging trends. The student can also constrain the range of years that will be represented in the

result set returned by the search engine. Similarly, the student can specify the minimum

occurrence frequency of both helper terms and the candidate trend.

Suppose the student enters the candidate emerging trend “aspect-oriented programming”

in the interface shown in Figure 4. Figure 5 depicts the results returned by the inquiry-based

Figure 5: Results returned after filtering

learning tool. The output interface provides support for searching for additional candidate

trends, or for identifying confirming evidence for a given trend, through the two buttons labeled

Noun-Phrases and Advanced. Using these features, students can identify terms using

combinations of words and parts of speech assigned to words.

Rich inquiry-based learning exercises of this nature give students a better appreciation for

the fundamentals presented in the courses, as well as a perspective of where technology is

headed. Learning how to do research is itself a valuable skill for lifelong learning.

Evaluation of ADT multimedia

Formal evaluation is crucial to determine whether multimedia works in improving

learning both in terms of objective knowledge and performance on task. In the fall of 2002, we

conducted a study on Abstract Data Types (ADTs) in a graduate level course on Object-Oriented

Software Engineering (OOSE). We designed an experiment to test the multimedia effectiveness

in student learning, both in terms of objective knowledge and in terms of students’ ability to

perform a task designing ADTs for a sample problem involving several classes and inheritance.

Mean scores on objective tests improved significantly, suggesting that the multimedia does

indeed contribute to objective learning of this content (Blank et al. 2002). However, the results

for the results for task learning were less clear than the results for objective knowledge. These

mixed results may have been due in part to the design of the experiment and in part to the design

of the multimedia. Learning from our experience, we have improved both.

With respect to multimedia design, focus groups and usability surveys led us to improve

both the user interface and content in many ways. For example, the JUST THE FACTS mode now

provides an alternative to the Flash media-rich version of the content. Students in both first year

and graduate level focus groups were roughly split about whether they liked media-rich

techniques such as animated, graphical personae and audio narration in the alpha version. Those

who liked the rich media included most of the women and non-majors, while those who found

these media a bit annoying and wanted to get right to the nuts and bolts tended to be male and

more experienced in computer science. Even students who prefer the media-rich presentation

would like a way to review the material in a less media-rich form. This result suggested that it

would be useful to provide HTML pages with the text and key graphics from Flash content

pages, dropping the use of personae and audio. Interactive exercises remain as is. Finally, a

summary with a checklist now appears at the end of the module, summarizing what one should

consider in designing ADTs. Each item in the checklist includes a link back to the relevant

screen, for review. Students in the OOSE focus group strongly agreed that a checklist would

give them a better idea how ADTs were designed as well as a good review.

In the spring of 2002, we conducted a study in an upper level undergraduate software

engineering course. We switched to a 2x2 design, dividing a class of about 72 upper level

undergraduates randomly into four groups: one fourth got neither the multimedia nor lecture, a

second got just the multimedia, a third got just a lecture, and the fourth got both the multimedia

and lecture. All four groups got the same homework problem description (see

www.cse.lehigh.edu/~cimel/eval/beta/ADTassignment.htm). After completing the assignment,

students took an online post-test of twenty multiple-choice questions presented in random order.

Our hypotheses were that the multimedia group would perform better than those getting

nothing and that the multimedia groups (with or without lecture) would also out-perform the

lecture-only group, on both posttest and task grades.

The results confirmed our hypotheses. The students getting multimedia scored higher,

on both the assignment, F(1,57)=12.25, p<.01, and the post-test, F(1,58)=15.39, p<.001, than

those without the multimedia (regardless of whether they saw the lecture). The lecture did not

have a significant effect on either assignment grades, F(1,57)=1.11 or the post-test, F(1,58)=0.55,

and there was no interaction between groups, for either the assignment, F(1,57)=0.01, or the

post-test, F(1,58)=0.43. Thus the new multimedia has a significant effect on learning, both in

terms of the objective knowledge (the post-test) and task knowledge (the assignment).

Good as these results are, closer examination of the task results indicated that further

improvement of the multimedia was possible. With or without the multimedia, most students

missed an important component of the assignment—reusing a given ADT (List) to construct the

semantics of a new ADT (Bowl). This observation led to a redesign of the multimedia to make

this point clearer: rearranging materials on ADT for collections as a separate chapter and adding

another exercise that explicitly gets learners to solve problems from existing ADTs. Further

improvements have also been made in response to feedback to student comments on a usability

survey. In the fall of 2002, we repeated the experiment of the year before, in the graduate level

OOSE course. The results this time showed improvement on both post-test and task, and most of

the students included constructive semantics in their ADT design. Thus, formative evaluation is

helping to drive our development process.

These experiments highlight that an increase level of interaction in multimedia increases

its effectiveness in teaching. This is evident since the student learning has improved over the past

experiments (Blank, Sahasrabudhe & Heigl, 2002), as a result of incorporating more interactive

multimedia. Multimedia design and development for computer science education is a continuous

process. We must also candidly admit that it is also a time-consuming process—though we hope

there will be a payoff in terms of modules that can be widely used and also in terms of lessons

learned for future development.

Evaluation of inquiry-based emerging trend detection exercises

We conducted another series of experiment test our methodology for detecting emerging

trends in inquiry-based exercises. In the fall of 2001, twenty-one students in a graduate level

OOSE class participated in this experimental evaluation to discover emerging trends. The class

was divided into two groups, groups A and B, each group having an approximately equal number

of students. Students from both groups A and B were expected to have attended the lectures of

the class. They were also expected to have introductory knowledge in the main topic area before

participating in this experiment. This was necessary as at different steps of the experiment they

needed to apply their domain knowledge to justify their choices of emerging versus non-

emerging trends. Also, all the students had access to their textbooks, reference books and

handouts given in the class. Both groups A and B attempted an exercise that involved

identification of three emerging trends in OOSE. In addition, group B was provided with the

methodology. Group B was also provided with a practical case study that demonstrated the

process of detecting emerging trends as outlined in the methodology. After completing the task,

students in group A were given the methodology and case study and required to resubmit their

solutions using the methodology.

Using a lower tail t-test we found, with a confidence level of 99%, that the mean

precision of sample 2 (followed the methodology including resubmissions from Group A) was

significantly greater than the mean precision of sample 1 (without the methodology). Also, with

a confidence level of 95%, mean precision of sample 2 (followed the methodology excluding

resubmissions from Group A) was significantly greater than sample 1 (without the

methodology). We thus concluded that students who used the methodology performed

significantly better on the task of Emerging Trend Detection (Roy, Gevry & Pottenger 2002).

In the spring of 2002, a similar experiment was conducted using the multimedia tutorial

in an undergraduate programming languages class. As with the first experiment, students in the

experimental group achieved significantly greater precision in detecting emerging trends when

using the methodology implemented in the multimedia tutorial (Roy 2002).

Related Work

While there is a great deal multimedia educational material, relatively little focuses on

computer science education, and even less is as deeply interactive as CIMEL. We compare

CIMEL to a brief selection of comparable systems oriented toward computer science education.

ProgramLive (Gries & Gries, 2002) is a rich multimedia tutorial of the Java programming

language. ProgramLive’s interface represents a notebook, within a browser. There are tabs to

the side of the notebook display that can be used for the navigation of the material, as well as

pop-up explanation of key terms. The CIMEL user interface also plays through a browser, but

avoids mixing interface metaphors by eliminating all of the usual buttons of a browser. The

interface thus immerses the learner in an environment uniquely associated with the material.

Another difference between ProgramLive and CIMEL is in their approach to feedback. CIMEL

provides an explanation of each of the wrong answers as the student makes these mistakes rather

than just providing the correct answer. This helps the student to gain a better understanding of

the thought process involved in solving problems.

An ongoing project at Massey University (Jesshope, Heinrich & Kinshuk, 2000) is

developing and evaluating an integrated system for web-based education. This system uses web-

based delivery of course material including interactive multimedia presentations, problem

solving and simulation environments in which students learn by doing. Like CIMEL, TILE

(Technology Integrated Learning Environment) provides students with an interactive multimedia

environment, and developers with a framework for managing, authoring, monitoring and

evaluating multimedia. The most salient differences are that the CIMEL framework lets students

go beyond the lessons through collaboration with experts (e.g., instructors, TAs, research

librarians and other student) as well as through tools that allow the student to explore current

research trends in course-related literature.

The Interactive Learning Modules (ILM) presents web-based multimedia tutorials,

created with the Director authoring environment (Millard, 1999). ILM provides a mechanism for

the creation of supplementary material for lectures, and collaborative problem solving

environments. The system is highly modular to encourage the usage of parts of the lesson

material in different courses. Similarly, the CIMEL multimedia framework is being developed

with modularity in mind, where each screen is a separate Flash movie, and screens are organized

hierarchically into sections and chapters. (We chose Flash instead of Director or Authorware

because its vector-based graphics lets us easily use the whole screen, regardless of resolution.)

The CIMEL dynamic tracks interface will let instructors and students create and traverse their

own learning track, corresponding to their unique requirements.

Conclusions and future work

Experimental results have confirmed the effectiveness of CIMEL multimedia. Along

with surveys and focus groups, they have also guided improvements to the user interface and

content. In addition to the units on ADTs and inheritance and emerging trends, we have

completed a survey of topics in software engineering and a unit on user interface design, which

we believe are suitable for use in various computer science courses. Interested parties are invited

to review sample content by logging on at http://www.cse.lehigh.edu/~cimel/prototype.html. As

this web page explains, rich multimedia works best with high-speed connections (with 56K

connections, sound often gets out of sync with the presentation). More documentation about the

CIMEL project is available at http://www.cse.lehigh.edu/~cimel.

Our current and future efforts are directed to developing material for a course for either

CS0, covering the breadth of computer science, or CS1, adding coverage of Java. This material

will parallel the content of a new textbook, The Universal Computer: Introducing Computer

Science with Multimedia (Blank, Barnes & Kay, 2002). The next textbook revises an earlier one

(Blank & Barnes, 1998) and the new multimedia recasts all of the multimedia associated with the

earlier title in the new user interface. Contact the lead author of this paper for more information

about the book or new multimedia.

For the CS1 course, we introduce Java using an objects-first approach and BlueJ

(Kölling, 2001). Earlier, we had developed material introducing Java with a traditional, syntax-

first approach, starting with a “Hello, World” program, then presenting classes in a second

chapter on Java. After the first author presented material from the second chapter at a SIGCSE

workshop on BlueJ (March 2002), other participants convinced him that a syntax-first approach

is problematic for Java, because it puts too much emphasis on distracting details, such as the

signature of a main method. With the BlueJ environment, it is possible to get students started

immediately with classes and objects. A pedagogical heuristic: students learn best what they

learn first (and repeatedly). A corollary for Java: classes and objects are what students need to

learn best. We therefore redesigned the chapter on classes and objects so that it would parallel

the first chapter of a new textbook introducing BlueJ (Barnes & Kölling, 2002). In the fall of

2002, we conducted another experiment, comparing how well students in a CS0/CS1 class

learned from the first chapter of the textbook versus our corresponding multimedia. Our results

show that students made significant gains from pre-test to post-test, with either the textbook or

the multimedia, but they learned more from the multimedia (t = -3.337, df = 76, p<.001), and the

multimedia adds to what they learn from the textbook (t = -6.527, df = 34, p<.001) . At

publication time, we are still analyzing grades on the Java homework assignment. Our tentative

conclusions are that first year students can learn Java “objects first” using BlueJ, and that

interactive multimedia helps.

As the COLLABORATE suggests, CIMEL will also comes with collaborative tools (Wang,

Kessler, Blank & Pottenger 2002). These tools will support instant messaging and chat among

two or more participants, and remote SHOW ME demonstration of applications on a computer

desktop. The SHOW ME tool. provides, on one user’s computer display, a view of the graphical

windows of an application (or entire screen, if specified) on another person’s display. Mouse

pointers are shown on both displays and outline, filled area, arrow, and text annotations can be

made directly on the application’s interface (as shown in figure 6). Communication sessions that

occur through these tools can be recorded as a video clip and added to a searchable multimedia

frequently asked questions (mFAQ) database. We plan to begin evaluation of these tools soon.

Figure 6: Annotating a programming environment using the SHOW ME collaborative tool

A broad goal of our project is to expand access to computer science education, attracting

more novices, especially women and minorities, and students with diverse learning styles, to

study and appreciate the field as a broad intellectual discipline. To this end, we will create

interactive multimedia and associated tools, both to accompany a new textbook for first year

students (and possibly high school students), as well as for uses independent of the book in other

CS courses, such as CS1 with Java and software engineering at the undergraduate and graduate

level. We also plan to extend the tracking capability so that instructors and other researchers can

visualize and evaluate student interaction with the multimedia, generic tutorials and automated

tools to help students discover emerging trends in computer science, and collaborative tools to

help students when they are “stuck” in the learning process.

Acknowledgements

This project was funded in part by National Science Foundation Grant Number EIA-0087977.

We thank our program managers Anita LaSalle for her support and encouragement. Thanks to

the many students who worked on the development of the multimedia and associated software,

notably David Gevry, David Goldfeder, Jeffrey Heigl, Martin Herr, Harriet Jaffe, Sumit Jain,

Chris Janneck, Aaron Sherrick, Soma Roy, David Servas, Qiang Wang and Fang Wei. Thanks to

Mark Shulman, Althea Smith and Pamela Wikkerink-Blank for portraying the faces of the

professor and TA personae. Thanks to Edwin J. Kay and M. Jean Russo for their help with the

design and analysis of our experimental data. Above all, co-authors Blank and Pottenger give

thanks to Yeshua the Messiah, for giving us the inspiration for this project, and for using it to

deepen our friendship and an ever-widening community.

References

Barnes, D. & Kölling M. Objects First With Java: A Practical Introduction Using BlueJ,

Englewood Cliffs, NJ: Prentice Hall, 2002. BlueJ software and documentation are available at

http://www.bluej.org.

Blank, G. D., Barnes, R. F. & Kay, E. J. The Universal Computer: Introducing Computer

Science with Multimedia. Manuscript is available from first author. The preface is also available

at http://www.cse.lehigh.edu/~glennb/um/0preface.pdf and the first chapter is available at

http://www.cse.lehigh.edu/~glennb/um/1intro.pdf.

Blank, G. D. & Barnes, R. F.. The Universal Machine: A Multimedia Introduction to

Computing. Burr Ridge, IL: McGraw-Hill, 1998.

Blank, G. D., Pottenger W. M., Kessler G. D., Roy S., Gevry D. R., Heigl J. J., Sahasrabudhe S.,

& Wang Q. (2002). Design and Evaluation of Multimedia to Teach Java and Object Oriented

Software Engineering. Proceedings of American Society for Engineering Education. Montreal,

Canada. Available at http://www.cse.lehigh.edu/~cimel/papers/ASEE2002.pdf.

Blank G. D., Sahasrabudhe S. A. & Heigl, J. J. (2002). Adapting Multimedia for Diverse Student

Learning Styles. Eastern Conference of the Consortium for Computing in Small Colleges.

Bloomsburg University, PA. To appear in Journal of Computing in Small Colleges. Available at

http://www.cse.lehigh.edu/~cimel/papers/LearningStyles.pdf.

Gries D. & Gries P. (2002). ProgramLive - Master Java Programming in a Self-paced Learning

Environment. New York, John Wiley and Sons, Inc.

Jesshope C., Heinrich E., & Kinshuk (2000). On-line Education using Technology Integrated

Learning Environments. Massey University, New Zealand. Retrieved October 14, 2002 from

www-tile.massey.ac.nz/publicns.html.

Kontostathis A., Galitsky L., Pottenger W. M., Roy S., & Phelps D. J. (2003) A Survey of

Emerging Trend Detection in Textual Data Mining. In A Comprehensive Survey of Text Mining,

Michael Berry, Ed., Springer-Verlag, (to appear).

Meyer B. (1988, 1998), Object-Oriented Software Construction. Englewood Cliffs, NJ: Prentice

Hall.

Millard D. M. (1999). Interactive Learning Modules for Electrical Engineering Education and

Training. Proceedings of American Society for Engineering Education.

Roy S. (2002). A Multimedia Interface for Emerging Trend Detection in Inquiry-based Learning.

M. S. Thesis, Computer Science and Engineering Department, Lehigh University.

Roy S., Gevry D., & Pottenger W. M. (2002). Methodologies for Trend Detection in Textual

Data Mining. In the Proceedings of the Textmine '02 Workshop, Second SIAM International

Conference on Data Mining, Washington, DC.

Thomas P. & Weedon R. (1995, 1997). Object-Oriented Programming in Eiffel. Addison

Wesley.

Wang Q., Kessler G. D., Blank G. D., & Pottenger W. M. (2002). Demonstration Tools for

Collaborative E-Learning. Poster at the ACM Conference on Computer Supported Collaborative

Work (CSCW 2002), New Orleans, LA.

Author Biographical Sketches

GLENN D. BLANK

Dr. Glenn D. Blank (glenn.blank@lehigh.edu) is an Associate Professor of Computer Science

Lehigh University. He is the principal investigator and lead multimedia designer for the CIMEL

project and The Universal Computer.

WILLIAM M. POTTENGER

Dr. William M. Pottenger (billp@cse.lehigh.edu) is P.C. Rossin Assistant Professor of Computer

Science and Engineering at Lehigh University. His research in Emerging Trend Detection as a

co-principal investigator for the CIMEL project builds on his research program in textual data

mining.

SHREERAM A. SAHASRABUDHE

Shreeram A. Sahasrabudhe (sas4@lehigh.edu) received his B.E. in Computer Technology from

Nagpur University, Nagpur, India in 2001. He is a currently pursuing an M.S. degree in

Computer Engineering at Lehigh University. He is a Research Assistant working under Professor

Glenn D. Blank on the CIMEL Multimedia Team.

SHENZHI LI

Shenzhi Li (shl3@lehigh.edu) received her M.S. in Computer Science and Technology from

Southeast University, Nanjing, China in 2001. She is a currently pursuing a Ph.D. degree in

Computer Science at Lehigh University. She is a Research Assistant working under Professor

William M. Pottenger on the CIMEL Multimedia Team.