This page shows this year's ideas for projects for third year undergraduates and MSc students. Each project comes with an indication of its difficulty and its suitability for different types/grades of student. Do not be put off if you think you'd really really like to do one of these projects but feel inappropriately skilled.

Intelligent keyboard

This is a taxing project for undergraduate students - the work is not that difficult (but it's not that easy either) - rather, there is lots of it!  It would be possible for two students to work together on this as the project naturally has components which have to be intelligently integrated.

When people making 'typing errors' they can do so for several reasons.  The fingers can slip. Finger troubles can be habitual/personal, or fatigue induced, or just 'chance'. The typist may not know how to spell. The typist may be guessing at the spelling based on the sound. More than one error may be present (finger-slippage whilst typing an incorrectly learned spelling). Errors may be based on 'sound / character slippage' rather than guess-work or learned error - e.g. there for their; to for two, etc.  - and thus involve some syntax if they are to be seen as errors.  Is a wrongly placed, or omitted, " ' " a spelling error?

Id like a keybaord to bee intellagent and sort out all thses posssibilities, perhaps on an indivudalised bases. :-)
 

Reference parser for auto input to Endnote.

Suitable for exceptional software engineer.

Endnote is a software package which runs with Word. It has one or more libraries of reference details (a database, really) entries in which can be selected and formatted according to a chosen style, and then 'dropped in' to a Word document. The problem is getting the reference details into the libraries from existing documents. The project is simple: devise software which will read reference details, parse them and export them in a format suitable for importing to Endnote.

Blog skin evaluator.

Suitable for exceptional software engineer.

Blog skins come in a variety of styles, formats, etc..... but in essence they are just chunks of code. The project is relatively simple: write a parser which takes blog skins, parses them, and evaluates them according to criteria (e.g. suitability for those with visual impairment, funky quotient, whatever).

Pantome: 3D data structure unification and display.

Suitable for exceptional software engineer with Cognitive Science / AI interests.

The project requires a student to re-engineer a project from some years back with improved graphics and speed. The work is relevant to those interested in speech and language processing.

PROJECT AREAS

In general I am prepared to supervise projects in HCI, Speech/Natural Language, Cognitive Science.  If you have your own project ideas in these areas then email me in the first instance - before coming along to see me.  Do this also if you want to pick one of the above projects. This will serve to establish precedence if I get swamped and have to start turning students away.

Information updated on 27th April 2009
This page maintained by W.H.Edmondson
The University of Birmingham
William's Home Page