G52GRP 2010/11 Projects

Click here for an alphabetical list of students and group allocations.

Project Descriptions and Groups 2010/11

Group ID: gp10-jaa

Supervisor: Dr. Jason Atkin (jaa)

Group mailing list: gp10-jaa@cs.nott.ac.uk

Group members:

Description:

Vehicle routing problems are common for delivery and logistics companies. They involve picking up and delivering goods using a set of vehicles. The aim is to allocate the vehicles to clients so that some objective (such as minimum travel time) is minimised. The goal of this project is to be able to represent these problems in a generic way and to visualise solutions. I envisage this having at least three components: a database to store problems and solutions to the problems, a visualisation engine to display the results to the user and a simple solver to find simple solutions to the problems. The database, thus, effectively forms the interface between the solver and visualisation, potentially allowing others to replace either of these components later. This presents the group with a number of options, emphasizing one or more of the database manipulation, graphical user interface or algorithmic problem solving elements according to their interests and abilities.

Group ID: gp10-jqb

Supervisor: Dr. Jaume Bacardit (jqb)

Group mailing list: gp10-jqb@cs.nott.ac.uk

Group members:

Description:

Bioinformatics is a very fascinating research area where many disciplines of science such as mathematics, computer science, engineering, etc. are put together to solve biological problems and bring new insight into our understanding of how life works. Within the bioinformatics context one of the most relevant topics of research is proteomics, the study of the role and structure of proteins and, in particular, the prediction of the structure of proteins (PSP). From a compute science point of view, PSP is a problem than involves complex data management. There ara repositories of protein data scattered around the world; moreover, protein data has to be heavily preprocessed before it can be feeded into data mining techniques, etc.

The goal of this project is to design and develop a web-based protein data management system, with the following characteristics:

• Aggregation of protein data from multiple sources
• Efficient structured storage of protein information
• Interface for the design and generation of protein-based data mining datasets
• 3D visualisation of protein's structural aspects, both in the interface as well as interfacing with existing protein visualisation suites

Group ID: gp10-rcb

Supervisor: Prof. Roland Backhouse (rcb)

Group mailing list: gp10-rcb@cs.nott.ac.uk

Group members:

Description:

The aim of this project is to develop a cycle route planner particularly oriented to the needs of students of the University of Nottingham. There are several cycle route planners already available for use. The first task will be to assess these against the needs of Nottingham students. Also, it will be necessary to determine how they gather and process their information. The next step will be to implement and test an improved system for use within the university.

Group ID: gp10-srb

Supervisor: Dr. Steven Bagley (srb)

Group mailing list: gp10-srb@cs.nott.ac.uk

Group members:

Description:

This group project will develop a software package that can create a personalized newspaper using the stories presented on a typical news website, such as the BBC News. The final output of this group project will hopefully be software that produces a typical broadsheet newspaper-sized layout that can be printed on the large-format printer.

Some of the possible tasks that your software will need to do are:

• Scrape the website to extract any new articles and content available.
• Provide the user with the ability to customize the newspaper to their own interests (for example, selecting which sections to include -- e.g. to include the Science and Technology section and drop the Sports section).
• Sort the content into the various levels of story provided in a newspaper.
• Layout the extracted and selected articles to form a newspaper.

Group ID: gp10-pxb

Supervisor: Dr. Peter Blanchfield (pxb)

Group mailing list: gp10-pxb@cs.nott.ac.uk

Group members:

Description:

A significant number of individuals are living in relative isolation. For instance, this is a common predicament for the elderly and for the disabled. Sometimes they will have concerned friends or relatives. But such contacts may not live nearby and, besides, being “visited” is not the same as “going out together”. Moreover, being able to “go out” provides individuals and their companions with a richer basis for their joint conversation. Of course, going out is also the way in which individuals feel more part of their community. Can technology help give isolated individuals some greater sense of this “getting out”, a vehicle for richer conversation and greater social and cultural engagement?

The design proposed for this project involves configuring mobile phone and wearable cameras to support a “virtual walking” conversation through IP telephony: a mobile walker linked to a screen-based partner. These walks might be orchestrated among friends and relatives, but a web-based scheduler is also sought whereby willing “walkers” can identify their availability and isolated individuals can book a period of such shared contact.

The full project description is available here

Group ID: gp10-vxc

Supervisor: Dr. Venanzio Capretta (vxc)

Group mailing list: gp10-vxc@cs.nott.ac.uk

Group members:

Description:

Scrabble is a word game in which two to four players score points by forming words from individual lettered tiles on a game board marked with a 15-by-15 grid. The words are formed across and down in crossword fashion and must appear in a standard dictionary.

The project consists in producing an electronic version of it with a graphical playing board and letter tiles. Important elements in programming it will be: randomly giving tiles to the players, keeping tracks of those that have already being used; scoring of points; correct implementation of the rules about multiple scoring and so on.

Further elements that could be developed are: the possibility to play on-line; checking the admissibility of words using an electronic dictionary; the possibility to play online over a network.

Group ID: gp10-axc

Supervisor: Dr. Andrew Crabtree (axc)

Group mailing list: gp10-axc@cs.nott.ac.uk

Group members:

Description:

The home is increasingly the focus of technology development following the widespread uptake of broadband and home network solutions. In 2009, over 70% of UK homes had Internet access, with 63% of all UK households connecting via broadband, and 45% doing so via wireless networks. The trend is increasing year on year and is accompanied by a shift in service access, media usage, and the proliferation of computational devices in the home. It results in new patterns of interaction and communication in and between homes and household members. While much of the emphasis in IT research has focused on the wholesale development of the smart home, the bulk of the market actually consists of homes which already exist where technology adoption is done in a piecemeal way according to user need.

The goal of this project is to develop applications and services for the “networked home” — i.e., for homes that have a broadband connection and wireless access. Your project can be on any topic that is relevant to domestic life: media usage, communication and collaboration, mobility, energy consumption, etc. The only requirement is that your software works in a broadband environment and supports existing activities in the home.

For further information, see this page.

Group ID: gp10-jqg

Supervisor: Dr. Julie Greensmith (jqg)

Group mailing list: gp10-jqg@cs.nott.ac.uk

Group members:

Description:

“You arrive home after a long day at work, open the fridge to find...a single egg, one lonely onion and some almost mouldy cheese......”. The aim of this project is to develop a software kitchen assistant tool. This tool must be able to provide recipes which match a supplied list of available ingredients. This is similar to the recipe search at bbc.co.uk/food, but is not constrained to just three search items. The software should provide a number of matching recipes and rank the suggestions according to how well they match. The recipe database can be online and community maintained and be able to store recipes uploaded by users, and provide recipe ratings. In addition, shopping lists of ingredients should be exported to an external smart phone application. As an extension a mobile version of the site should be created so users can create and save their shopping lists on the go. This is the ultimate kitchen management tool, useful for aspiring Heston Blumenthals everywhere, anywhere.

Group ID: gp10-cah

Supervisor: Dr. Colin Higgins (cah)

Group mailing list: gp10-cah@cs.nott.ac.uk

Group members:

Description:

This project involves using historical data on the outcomes of football (soccer) matches to predict future results. Large amounts of such data are now available on the web. The prediction can be via any method suitable, most likely an AI method such as genetic algorithms or neural networks. When the system is up-and-running several days predications can be compared to both odds available on the outcomes and the final results of the matches, to see whether the systems predicts better than the human population (which influences the odds). It will be beneficial to chose something with only two outcomes such as over/under a total number of goals or “asian handicaps”.

Technically this project involves obtaining and parsing the html to obtain past results, analysing these to see which items are the best predictors, building an AI based predication system and comparing this to odds available and actual results.

Group ID: gp10-gtr

Supervisor: Dr. Gail Hopkins (gtr)

Group mailing list: gp10-gtr@cs.nott.ac.uk

Group members:

Description:

This project will involve developing an application which uses tangible devices, or phidgets, to present information about remote locations to users. These phidgets are electronic devices which can be plugged into a computer's USB port to offer extra functionality. Examples include motors and sensors.

The group will be required to decide on how they want to use the phidgets and to design, implement and evaluate an application for the purpose remote activity monitoring.

Group ID: gp10-dsk

Supervisor: Dr. David Kirk (dsk)

Group mailing list: gp10-dsk@cs.nott.ac.uk

Group members:

Description:

In this project the software team will develop a collaborative media sharing application to be run on mobile devices. The application will be designed to support the ad hoc sharing of media (this could for example include photos and/or videos) within social settings.

The idea is to prototype and develop a mobile application, the specific requirements of this are to be determined but that will allow users to define ad hoc groups of people who they wish to share media with during an event and then have access to that media at a later date in a shared repository somewhere.

It is assumed that the definition of 'groups' for sharing will be based on some notion of proximity sensing using blue-tooth (or similar) connections on the mobile devices.

Part of the project could potentially include the editing and manipulation of the media within the application, thus going beyond simple sharing.

Solutions should be based on observations of real world activities and involve potential users in their development through formative evaluation. Possible technical solutions will require database, interface, networking and mobile application development.

Group ID: gp10-bnk

Supervisor: Dr. Boriana Koleva (bnk)

Group mailing list: gp10-bnk@cs.nott.ac.uk

Group members:

Description:

A growing area of interest is concerned with how interactive systems can be used to better inform the choices people make in their everyday lives. A subclass of such systems, that is attracting a lot of recent research attention, is focusing on facilitating more environmentally sustainable behaviours. This project will develop a number of different interfaces that make inhabitants aware of their energy consumption to encourage them to reduce their usage. The project will explore both different approaches for motivating energy saving behaviour (e.g. providing general awareness of usage, displaying quantities and concrete consequences, making social comparisons and running competitive games ) and different interfaces to access the data (e.g. web interface, mobile phone, ambient display). In this project you must engage in a creative process of design, proposing design concepts and iteratively developing prototypes based on formative user evaluations.

Group ID: gp10-jds

Supervisor: Dr. Dario Landa-Silva (jds)

Group mailing list: gp10-jds@cs.nott.ac.uk

Group members:

Description:

The purpose of this project is to develop a computer game that simulates the delivery of relief supplies to people in need as part of a disaster management plan. The computer program should display a map with the locations of the groups of people that need the supplies. The only way to deliver the supplies is by air drop and therefore, the supplies are contained in crates attached to a number of parachutes which can be dropped in any area of the map. Once the crates are dropped, the people has to walk to the nearest crate in order to collect the relief supplies. The walking distance between two areas in the map with no obstacles in the way is calculated using a straight line. Determining the best locations to drop the crates in order to minimise the total distance walked by all the people is a logistics optimisation problem that can be solved with operations research techniques. The game should present a relief mission scenario and give the user a limited amount of time to determine the best drop locations. Then, the solution given by the user should be compared to the optimal solution which will be computed by the program using an appropriate algorithm. There is scope to expand the game for example by incorporating: different difficulty levels, obstacles in the map, limited capacity of the crates, etc.

Group ID: gp10-bai

Supervisor: Dr. Bai Li (bai)

Group mailing list: gp10-bai@cs.nott.ac.uk

Group members:

Description:

This project focuses on the creation of a user interface for viewing and analysis of Magnetic Resonance (MR) images. The interface should allow the display of 3D MR images, as well as 3 side views of a 3D image. The user interface should also be easy to use. Publicly available 3D visualisation software libraries maybe used. Specific image analysis applications will be lesion detection from MR images of Multiple Sclerosis (MS) patients with visible lesions. The user interface should provide functions for loading MS images, detecting lesions, etc.

Group ID: gp10-bsl

Supervisor: Dr. Brian Logan (bsl)

Group mailing list: gp10-bsl@cs.nott.ac.uk

Group members:

Description:

In his book, “Vehicles: Experiments in Synthetic Psychology”, Valentino Braitenberg explores a series of thought experiments designed to show how complex behaviour can result from very simple reactive agent architectures. Braitenberg's vehicles use direct, excitatory and inhibitory couplings of sensors to motors: sensors respond to features in the environment, e.g., heat, light, obstacles etc.; motors move the vehicle in response to signals from the sensors; connections carry signals from the sensors to the motors and either cause them to turn or inhibit them from turning. Although they are very simple, such vehicles can give rise to complex emergent behaviour through interaction with other vehicles and their environment. In his book, Braitenberg describes a range of vehicles, including those (he) imagined to exhibit “cowardice”, “aggression”, even “love”.

The aim of this project is to implement a Braitenberg Vehicles simulator, which could be used to explore new kinds of vehicle, and for teaching the principles of reactive agent architectures. Ideally, this would be a graphical applet that allows a user to configure and execute a number of vehicles of different sorts to see how they interact. See the Braitenberg Vehicles simulator from Kovan University for an example. The user should be able to save vehicle designs, and to to record experiments for future playback. Critically, the vehicles should be 'aware' of each other and interact with objects in their environment, such as obstacles (which the Kovan applet doesn't do). The project could be extended in a number of different ways, e.g., to allow the user to steer one of the vehicles, or by adding realistic physics to the vehicles' interactions.

Group ID: gp10-nhn

Supervisor: Dr. Henrik Nilsson (nhn)

Group mailing list: gp10-nhn@cs.nott.ac.uk

Group members:

Description:

Guitar Hero is a combined musical and computer game that has become very popular recently. This project is inspired by that game, but aimed at keyboard players instead, and with a more educational scope. It is envisioned that the game will be played by attaching a MIDI keyboard to the computer.

One can imagine many kinds of games. The simplest is perhaps showing the score for a musical phrase to the player, picked by the computer at random from a library (or maybe even randomly generated), and asking him or her to play it as accurately as possible guided by a metronome click. Points are scored depending on how accurately the phrase is played, and the difficulty of the phrase (e.g. tempo, how involved it is, key, ...). One can imagine variations with one note at a time or chords, and one or two hands simultaneously. Other variations include asking the player to play with a certain rhythmic feel, like swing, as opposed to as accurately as possible.

To make the game more fun, and more like Guitar Hero, the metronome click could be replaced by a backing track (either MIDI or audio) for well-known tunes.

For ultimate fun, consider on-line play. It's likely not feasible to broadcast the performance of a player in real-time unless they are on the same local net, but one could easily capture a performance as a MIDI sequence, and then sending that to the other players so they can hear how well or badly their opponents played a particular phrase. One could consider allowing the players to set each other musical challenges by picking phrases from well-known songs as an alternative to the computer picking phrases at random.

In short, plenty of scope to develop this project in different directions! As a base-line, the project should implement at least a simple variation of the game.

Group ID: gp10-exo

Supervisor: Dr. Ender Ozcan (exo)

Group mailing list: gp10-exo@cs.nott.ac.uk

Group members:

Description:

Description: This project involves in creating a natural and easy to use intelligent user interface by combining techniques from computer vision, machine learning, computer graphics and human-computer interfaces. Interactive multimedia such as computer simulations and animations received increased attention over the years as supplementary teaching tools and have now become integral components of most engineering and science curriculums. One way to boost the utility of such simulations and animations is to make them easier to use. In this project, a pen-based interface will be implemented and combined with a speech recognition system (Windows SAPI). The goal is to recognize spoken commands and interpret free-hand drawings for constructing weighted and unweighted graphs which functions as a front-end to a shortest path and minimum spanning tree (MST) algorithm simulators. The usability of the different types of interfaces (such as, a WIMP based interface and hybrids) will be evaluated.

Initial Reading:

• H. Dibeklioglu, T. M. Sezgin, E. Ozcan. A Recognizer for Free-Hand Graph Drawings. International Workshop on Pen-Based Learning Technologies, Catania, Italy, May 24-25 (2007).
• A. Blessing, T. M. Sezgin, R. Arandjelovic, P. Robinson, A multimodal interface for road design. Workshop on Sketch Recognition, International Conference on Intelligent User Interfaces, Sanibel, FL, February (2009).

Group ID: gp10-ajp

Supervisor: Dr. Andrew Parkes (ajp)

Group mailing list: gp10-ajp@cs.nott.ac.uk

Group members:

Description:

Automated planning of routes for car journeys is a common service on the web and within SatNavs. Although very useful, the routes provided can often be unsatisfactory. A common complaint is that they send users down very minor roads. If the only information available is the speed limit on a road then it might give an over-optimistic estimate of (safely) achievable transit times. The aim of this project is to develop a system that will extract other information about the proposed routes and use it to provide a better estimate of their relative suitability. Sources of information can be publically-available map data. Methods to measure suitability might well include some form of simulation of a driver along the routes, under estimated traffic conditions. The project should also account for different kinds of drivers and vehicles and for different driver preferences.

Group ID: gp10-qiu

Supervisor: Dr. Guoping Qiu (qiu)

Group mailing list: gp10-qiu@cs.nott.ac.uk

Group members:

Description:

An objective non-referenced image and video quality metric is fundamental to many image processing tasks including compression, restoration and high dynamic range compression (tone mapping). For example, in archive video restoration or post production, some segments of the video may be of perfectly good quality and there is no need to do anything, some segments may contain defects that can be easily restored automatically and yet some may contain serious defects that will need manual intervention. With a good quality metric, we can automatically identify various video segments and apply the appropriate processing solutions.

One of the biggest obstacles in building an image/video quality metric is the lack of data. Many existing image metrics are based on limited number of images evaluated by a small number of subjects under laboratory environments [1]. Collecting good quality large datasets is an essential first step in building a high quality image/video quality metric. We have been experimenting using Web2.0 technology to harness the power of millions of Internet users for collecting subjective image quality assessment data (see http://www.hdri.cs.nott.ac.uk/ and http://www.hdri.cs.nott.ac.uk/siq/).

In this project, we will develop sophisticated games with a purpose (GWAP) [2] to collect subjective quality assessment data.

The game has to be designed and implemented in such a way that it is not only engaging and fun to play, but very importantly can collect high quality useful data for image/video quality assessment.

References:

• 1. Z. Wang, A. C. Bovik, H. R. Sheikh and E. P. Simoncelli. Image quality assessment: From error visibility to structural similarity. In IEEE Transactions on Image Processing, vol. 13, no. 4, pp. 600-612, Apr. 2004
• 2. L Von Ahn, L Dabbish. Designing games with a purpose. In Communications of the ACM, vol. 51, no. 8, pp. 58 – 67, August 2008

Group ID: gp10-rxq

Supervisor: Dr. Rong Qu (rxq)

Group mailing list: gp10-rxq@cs.nott.ac.uk

Group members:

Description:

Definition of Portfolio Management at Wikipedia:

In finance, a portfolio is an appropriate mix or collection of investments held by an institution or an individual. Holding a portfolio is a part of an investment and risk-limiting strategy called diversification. By owning several assets, certain types of risk (in particular specific risk) can be reduced.

This group project aims to develop a financial decision support system for bank managers or individuals to analyse and choose a set of assets of investments which returns the maximum expected profits while considering its associated risk. You will need to develop a system which can:

• fetch the latest financial data from online resources, e.g. Yahoo! Finance
• support analysis on the fetched data
• select the optimal set of assets, under certain constraints from the user, which returns the maximum profits under the risk limit
• provide user account management functions, support record keeping, invoicing or client reporting, etc

In doing this, a review on existing portfolio management tools should be carried out. A computational method or an algorithm (either an exhaustive calculation or a simple search algorithm) should be developed to intelligently choose the optimal or the best possible set of assets within given time.

Group ID: gp10-mvr

Supervisor: Dr. Milena Radenkovic (mvr)

Group mailing list: gp10-mvr@cs.nott.ac.uk

Group members:

Description:

The aim of this project is to develop a web-based system for booking tickets for air travel. Given date, origin, destination, and number of passengers, along with possible constraints or preferences, such as desired times of departure and arrival, travel with a specific company or airline network, fewest hops, cheapest tickets, etc., the system should suggest possible routes and allow the customer to select and book one of them. To do this, the system should interact with a database of flight information detailing flights, prices (possibly different prices depending on circumstances), constraints, current availability, etc.

Compared to a real such system, a lot of simplifying assumptions will have to be made. Nevertheless, it is expected that some background research on how this kind of systems work, what information is available in real flight databases, and how it is used is carried out. Then, based on this, a central aspect is to come up with good, general, and principled algorithms for suggesting and optimising routes.

The booking server should support multiple simultaneous clients, so mechanisms to avoid e.g. double bookings or discovering that some legs of a multi-leg trip are no longer available need to be implemented (without permanently “losing” seats that in the end are not booked).

The project should also support a (mock) payment system, so that it can charge the passengers after the passengers have input the information about the trip.

Finally, the system needs to be able to send notifications (in form of e-mails and smss) to the customers when there are unexpected changes to the scheduled flights.