I've listed myself at Technorati. Technorati Profile

The senior-most year at our college, IOE (http://www.ioe.edu.np) has a course that requires everyone to complete a Project. We call it Major Project, reason being the junior year has similar but small scale project thats called Minor Project. A lot of students I've seen are really confused as to what topic to choose to do their projects in.

Now that I've completed my undergraduate and am working as a full time employee, let me share my experiences.

Before I decided my project, there were a lot of options. I've seen earlier projects related to Networking, Database Management Systems, MIS, AI related, Image processing, etc etc. I had to narrow down my choices. MIS and DBMS always seemed the easier options. Those were out of my options early on. Projects related to Computer Networks sounded interesting. So did projects related to Artificial Intelligence. Eventually I decided to take work on AI and Image processing. The reason being, my elective was Image Processing and it was really fun, and Abhishek had really good vision at that time about what we wanted to achieve .

Our project was Real Time Face Tracking and Recognition (http://rtftr.sourceforge.net). Our target was to make the system modular so that we could plug-in face detection and recognition algorithms and compare their result. We divided our team into two groups, the Face Recognition group and Face Detection group. Anjan and myself worked on the Face Detection group. I studied Neural Network based Face recognition and Anjan researched into the Cascaded Adaboost Algorithm. Likewise Abhishek dived into the world of Gabor wavelets and Lakesh was more involved into PCA and subspace LDA for Face Recognition.

Before we divided our respective topics, we spent a lot of time reading papers on current state of the face detection and recognition technology. There were comparative papers and literary surveys on the Internet. We held meetings at regular intervals to discuss on where and when to begin, set deadlines and share our gained knowledge.

We received a great response from our supervisors and our examiners on the final presentation. And huge huge credit goes once again to Abhishek since his coordination skills and analytical skills are really exceptional.

My point is, Senior years project should be more research based and less about implementations. It matters less, how polished your project comes out in the end. Rather it matters how much effort you put into gain knowledge on your field. Students might be quite tempted to copy an existing system and implement with their own code. There were projects titled XYZ Management Information System. You will have more than plenty of time in your professional life to build those systems. Instead I recommend students to work on new ideas and implement algorithms to see how they fit in to solve real world problems. We did that during the period of our project.

There are some wonderful tips at http://www.ehow.com/how_13147_choose-topic-senior.html about how to choose a thesis topic. Of many points mentioned there, I really think two of them.
1. Focus on where your interest lies.
2. Have constant communication with your teachers and professors.

Besides, another important thing to keep in mind is, it is not always necessary to come up with your own brilliant idea. A lot of work are already done by your seniors. Countless man-hours are devoted every year for new ideas. Go through all the titles and see if you like any of them. Then start building your project on top of it. The benefit for you is, you will have a solid researched topic for your own project, and you can start doing the interiors on already constructed framework.

I hope my views are of help to the confused minds out there.