Suggestions for Final Year Projects for the academic year 2010-2011

The following are suggestions for final year projects that I am willing to supervise for the year 2010/2011. I am allowed to supervise only three projects. If you are interested please pass by my office to discuss the projects as well as your interests.

Software synthesis for parallel platforms

New computers exhibit more and more cores and processors. We expect that an application would run faster when we move from a platform with one processor to a platform with more processors. However, this does not happen. The reason is that legacy software is written to work in sequential mode. If we were to benefit from the parallelism, we will need to rewrite the software to use the parallel power. The purpose of this project is to propose a new methodology to write software so that it is portable to new platforms and still can use their parallel power. This will build on previous work from last year's FYP on a similar project.

Disciplines: Software, hardware computer architecture and digital systems, intelligent systems Advisors: Fadi Zaraket team: (Nourreddine et al.)

Engineering and math speech engine

This project aims at taking the speech of an engineer and to produce a mathematical typeset equation. The work involves developing methods and tools to capture and analyze signals. It also involves building rules to transform signals into mathematical symbols, then grammars to form equations of these symbols and signals. The last step is a correction mechanism to correct the equation if needed.

Disciplines: Software, signal processing, intelligent systems Advisors: Fadi Zaraket and Fadi Karame team: (Batlouni et al.)

Content-based controller for Arabic network traffic

Parental controls help parents allow their kids to benefit from the internet without fearing the danger of violent and obscene material. These controls partially rely on processing the content to detect and rate the target material. This project will provide a solution for Arabic traffic as currently these solutions are mostly manual. In this project we will use an Arabic morphological analyzer developed here at AUB to detect the bad content. The project will also involve networking and intelligence work to set up control lists and correction mechanisms.

Disciplines: Software, Networking, intelligent systems Advisors: Fadi Zaraket team: Open

Media search

Do you own a digital camera? How many times did you look for half an hour at your own collection of digital photos and videos to find one photo or few seconds of video. This project aims at automating the search for personal images on your local hard disk. Google, Microsoft among other companies are working on solving this problem at large for an infinity of images on the web. We will solve a smaller problem and we will define it in an easier way to provide a better answer. At the end of this project, we will pick a cute name for our software and sell the product to one of the big guys. The team needs to have good programming skills and a passion on building software. This will build on previous work from last year's FYP on a similar project.

Disciplines: Software, signal processing, intelligent systems Advisors: Fadi Zaraket and Fadi Karame team: Open

Automatic testing and verification for software (GUI, privacy, and security case studies)

Testing GUI is a nightmare. Most current techniques use record and play to record a sequence of GUI events, save golden output at selected breakpoints, and then play the recorded event and compare the results. These techniques are very sensitive to any change to code, functionality, and even hardware such as resolution and color settings since the saved golden output highly depends on these factors. In this project, we will build a GUI automation tool that will allow specifying properties of GUI formally so that we can automatically check whether the properties hold for the GUI with modern techniques such as shape analysis techniques.

Disciplines: Software, graphics and visualization, intelligent systems Advisors: Fadi Zaraket and Wassim Masri team: Open

Strategy diversification with trading robots

Trading robots try to estimate risk and gain margins in moving markets. They also suggest an asset diversification plan that suits the profile of a specific users. More advanced trading robots analyze news reports to detect patterns and relations to market quotes. They all use preset trading strategies suitable for the profile choices. In this project we will use existing trade simulators to investigate and deploy a new methodology in trading robots. The idea is to diversify trading strategies instead of diversifying assets.

Disciplines: Software, intelligent systems, financial engineering Advisors: Fadi Zaraket team: Open

fyp1011.txt · Last modified: 2014/08/13 09:34 (external edit)
Except where otherwise noted, content on this wiki is licensed under the following license:CC Attribution-Noncommercial-Share Alike 3.0 Unported
Recent changes RSS feed Donate Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki