Details of previous projects I have supervised.
Below is a list of the projects available for students in 1998-9.
Mathematics includes notations which are ambiguous, except that mathematicians can easily disambiguate them given their context. For instance, in a calculus textbook, no one would think that
dy/dx
referred to a fraction, and start trying to cancel the d's. The question arises whether a computer program can perform similar interpretations. In Latex (Lamport, 1988), for instance, the above would be written as
\frac{dy dx}
suggesting that it is indeed a fraction. This can be a problem when rendering mathematics in non-visual forms, which is one of the major objectives of the Maths Project.
The objective of this project would be to see to what extent a mathematical text (in Latex) can be analysed and modes or styles of presentation (e.g. calculus, trigonometry, etc) identified. One approach to this might me to use tools such as Lex and Yacc.
The student will be expected to produce a Web page describing the project.
People who have memory problems (possibly due to being elderly or being an academic) often rely on lists. For instance before going on holiday, they will make out a check list of the items they need to take. Of course one of the problems is remembering what to put in the list. The idea behind this project is to build an intelligent assistant for compiling such lists.
There would be two approaches combined. One is for the system to apply pre-stored knowledge about the types of items needed for a particular type of journey and the other would be to learn about the individual's patterns of requirements.
Hence there will be one database of items tagged with information about the type of holiday on which they are required (skis may be required for a winter alpine holiday, a passport will be required for an overseas trip and so on). Having given some information about the holiday, the system could thus prompt for items that might be required. The user can choose whether a suggested item will be added to the list or not. The system would learn from this. Next time it was asked to generate a list for a similar holiday it would be biased towards suggesting the same items as were included last time (e.g. if skis were not taken on the last alpine holiday, the user is probably a walker and not a skier).
The project would involve designing suitable representations of the data, such that it embodies the information required and the similarity links between items. It would also have to develop a user model. For this it might use techniques similar to those developed by Spooner (1996). Wilson, Evans et al.(1997) and Aldrich (1997) describe a rather different use of information technology to assist people with severe memory impairment.
The student will be expected to create a web page describing the project.
The objective of this project would be to build a kit from which simulations of roller coaster rides could be built. The user of the kit (probably a physics student) would specify the dimensions of the components (height of the drops, radius of curves etc) using the equations of motion. Once constructed, the user could set a car off rolling on it and then find out if it works - whether the car gets safely to the end of the ride.
The project - and the resulting software - might be developed to increasing levels of complexity. For instance, the initial version might build only two-dimensional rides and not take account of friction. Such complications might be added later.
The project would require good programming skills - to build a graphical kit that is easy to use. A knowledge of mechanics would be necessary. (Books such as Maxwell, 1952; Huang, 1967) should help). Some thought will be required regarding the pedagogical aspects; how much work will the computer do for the student, for instance? (O'Shea and Self, 1983, will help with some of the questions regarding the design of computer-aided learning software).
The student will be expected to create a web page describing the project.
Stop Press
The Mechanics in Action Project may be very relevant to this project.
The simplest thing to do with your email is to read it, but often that is not appropriate. For instance, when people are going away they often set up programs (such as gone.away(1)) which will automatically reply to their email with an appropriate message.
The idea of this project will be to build a set of tools with which the user can build up quite complex programs that will respond to email in different ways.
The tools will be based on Unix pipes. For instance, suppose there are tools
Given such tools, the action of gone.away could be achieved by a shell program such as:
|
from='sender' |
Extract sender's address |
|
if !inlist $from gone.addr |
Has this sender already received a reply? |
|
then reply $from gone.msg |
If not - send it |
|
echo $from >> gone.addr |
Record who has mailed |
|
fi |
|
|
cat >> gone.mbox |
Save the message received |
Pipe to /usr/alistair/gone
then the script would operate on incoming mail.
Other tools that might be provided include:
The student undertaking this project might devise other useful tools - and implement them.
The obvious implementation language would be Perl (Schwartz, 1993) and, given its facilities, the implementation would not be difficult. There would therefore be scope to extend the project much further. A particularly valuable extension would be tools to automatically deal with spams.
The documents in /sys/doc/rfc/rfc821.txt.gz and /sys/doc/rfc/rfc822.txt.gz may be useful.
The student will be expected to create a web page describing the project.
One of the apparent attractions of graphical user interfaces (GUIs) is the fact that users can easily learn the way that the components operate because they can infer the meanings of their symbols. For instance, the Macintosh Wastebasket icon is recognizable as a picture of a real-world item and its use as a means of disposal can easily be deduced - and remembered. However, the growing use of icons in GUIs has the effect that their meanings become increasingly arbitrary. (See the toolbars, ribbons and palettes of any piece of Microsoft software).
According to Macken, Perry et al., 1993 symbols such as computer icons can have properties including readily inferable meaning (RIM) and easily rememberable meaning (ERM). Furthermore they suggest that symbols may have arbitrary conventional meaning (ACS) which are `more difficult to infer and remember' (op. cit.) However, they present no evidence for this assertion and it does seem to contradict the findings of Grudin (1989) who found in the context of keyboard driven interfaces that, once learned, arbitrary key mappings are readily remembered.
The purpose of this project would be to find out where the truth lies. A human factors experiment would be carried out whereby subjects would be presented with icons (probably taken from the software supplied with Zetie, 1995). They would first be asked to suggest what meanings they would infer for each of the icons within a particular context. this would measure the degree of RIM. For the icons which they get `wrong' they would be given the correct meaning and some time later re-tested to see whether the given meanings are remembered.
The degree of arbitrariness could be tested by attaching more or less meaningful descriptions to the icons. In other words, if it is possible to give the user some kind of `story' to explain the mapping of name to symbol then the meaning is not entirely arbitrary, whereas in other cases one might impose a meaning which has absolutely no apparent mapping to the symbol (rather as Grudin does with his key mappings).
A further extension to the study would involve regular users of graphical software. Do they know the meanings of all the icons available? If not, can they learn and remember them? The study by Mayes, Draper et al. (1988) would suggest that users do not remember the exact contents of interface components (in this case icons), but that it would be sufficient for the icon to act as a reminder of its function.
The student will be expected to create a web page describing the project.
One of the powerful aspects of vision is the ability to view more than one thing at a time, really the facility to switch quickly from one to another. That is one of the reasons we have computer monitors with large screens and window-based interfaces; it is valuable to be able to compare the contents of different windows. The non-visual senses are not as good at such pseudo-parallel processing. This is one of the major problems in adapting computer interfaces for use by blind people (Edwards, 1988; Mynatt, 1994; Mynatt, 1997).
The objective of this project would be to look at the feasibility of implementing parallel inspection of auditory data in an application in which it is particularly important. The application is biological, that of comparing DNA sequences. The degree of homology between two sequences (of amino acids or nucleic acids) is a strong clue to the function of that sequence. Pairwise or multiple alignments are made to exhibit these homologies: Identical or functionally similar residues are juxtaposed, with gaps inserted to optimise the alignment. A biologist looking at such an alignment can spot important regions of the sequence.
An obvious approach would be to use two streams of synthetic speech (Edwards, 1991), but there are a number of questions to be tackled as to how best to do this. For instance, should headphones or stereo speakers be used - with one voice per channel? Should different voices be used, if so, which are best? There may also be a role for use of non-speech sounds (Hereford and Winn, 1994; Brewster, 1994).
The project would involve software implementation and user testing.
The student will be expected to create a web page describing the project.
The Department has acquired a prototype joystick with tactile feedback. A pantograph-style device is attached to the wrist, such that the lever rests on the back of the hand. The user moves the lever with the other hand, receiving feedback about the position of the end of the lever and hence the position of the cursor on the screen.
The device has been designed as an alternative to the mouse for blind computer users (Edwards, 1987; Edwards, 1989), to be used in conjunction with other non-visual feedback. It needs to be evaluated. Does it work, does the extra channel of information enhance the interaction, making it quicker or less error-prone? What applications is it useful for?
The project would involve designing and carrying out some form of controlled test. Robson (1994) describes the kinds of techniques required, while Stevens and Edwards (1996) explains some of the difficulties in doing this kind of study. This project might be carried out in conjunction with ADNE7.
The student will be expected to create a web page describing the project.
The ability to spatialize sounds in three-dimensional space offers new possibilities for human-computer multimedia interaction (e.g. Crispien and Felbaum, 1995). A variety of techniques for processing sounds in this way (Begault, 1994; Keating, 1996). However, they are not always as useful as might first seem (Crispien, op. cit.) for a variety of reasons. A simple, low-technology alternative is to use a set of speakers and to simply vary the balance between them (Crispien and Petrie, 1993). The Department has acquired a circuit board which will perform this simple processing and the objective of this project would be to evaluate it. The first phase of the project would be to take some measurements as to how accurately users' perceptions are. For instance, how closely can they locate the sound source, how well do they perceive the shapes traced by a moving source. The second phase of the evaluation would attempt to find out how useful such a system might be in particular applications, with particular reference to the potential use by blind people.
The student will be expected to create a web page describing the project.
The Maths Workstation has a complex, multimodal interface. Most manipulation is based on the use of a keyboard, but the system was designed such that it is possible to use speech input for control inputs. Evaluation of the keyboard-based interaction (Stevens, Edwards et al., 1996) did not reveal any significant usability problems, but nevertheless there is the suggestion that a speech-based dialogue with the system (i.e. speech input evoking speech output) might be a particularly natural form of interaction. The objective of this project would be to investigate that.
The first problem is to decide what to measure. As suggested above, errors in the keyboard-based system were few so that error rates will not be a good thing to count. Instead, the suggestion is that the speech input will enhance the users' sense of direct engagement (Hutchins, Hollan et al., 1986). So, the first objective of the project will be to decide exactly what direct engagement is - and how to measure it. Then test will be designed and carried out to test the hypothesis.
The student will be expected to create a web page describing the project.
Aldrich, F. (1997). A case study of Neuropage: A reminder system for memory-disabled people. in Computers in the Service of Mankind: Helping the Disabled, (London), IEE. Digest number 97/117 pp. 10/1-10/3. (I have a copy)
Begault, D. R. (1994). 3-D Sound for Virtual Reality and Multimedia. Boston: Associated Press.
Brewster, S. A. (1994). Providing a structured method for integrating non-speech audio into human-computer interfaces. DPhil Thesis, University of York.
Crispien, K. and Felbaum, K. (1995). Use of acoustic information in screen reader programs for blind computer users: Results from the Tide project GUIB. in I. P. Porrero and R. P. de la Bellacasa (ed.), The Eurpoean Context for Assistive Technology: Proceedings of the Second Tide Congress, (Paris), IOS Press. pp. 306-311.
Crispien, K. and Petrie, H. (1993). Providing access to GUIs for blind people. in Proceedings of the 19th Convention of the Audio Engineering Society.
Edwards, A. D. N. (1987). Adapting user interfaces for visually disabled users. Unpublished PhD Thesis, Open University.
Edwards, A. D. N. (1988). The design of auditory interfaces for visually disabled users. in E. Soloway, D. Frye and S. B. Sheppard (ed.), Human Factors in Computing Systems: Proceedings of Chi '88, (Washington), pp. 83-88.
Edwards, A. D. N. (1989). Soundtrack: An auditory interface for blind users. Human Computer Interaction 4(1): pp. 45-66.
Edwards, A. D. N. (1991). Speech Synthesis: Technology for disabled people. London: Paul Chapman.
Grudin, J. (1989). The case against user interface consistency. CACM 32(10): pp. 1164-1173.
Hereford, J. and Winn, W. (1994). Non-speech sound in human-computer interaction: A review and design guidelines. Journal of Educational Computing Research 11(3): pp. 211-233.
Huang, T. C. (1967). Engineering Mechanics. Reading, Massachusetts: Addison-Wesley.
Hutchins, E. L., Hollan, J. D. and Norman, D. A. (1986). Direct manipulation interfaces. in D. A. Norman and S. W. Draper (ed.) User Centred System Design. Hillsdale, New Jersey: Lawrence Erlbaum Associates. pp. 87-124.
Keating, D. A. (1996). The generation of virtual acoustic environments for blind people. in P. M. Sharkey (ed.), Proceedings of the First European Conference on Disability, Virtual Reality and Associated Technologies, (Maidenhead), University of Reading. pp. 201-208.
Lamport, L. (1988). Latex: Users guide and reference manual. New York: Addison Wesley.
Macken, E., Perry, J. and Haas, C. (1993). Richly grounding symbols in ASL. Sign Language Studies (December):
Maxwell, J. C. (1952). Matter and Motion. New York: Dover.
Mayes, J. T., Draper, S. W., McGregor, A. M. and Oatley, K. (1988). Information flow in a user interface: The effect of experience and context on the recall of MacWrite screens. in D. M. Jones and R. Winder (ed.), People and Computers IV: Proceedings of HCI '88, (Manchester), Cambridge University Press. pp. 275-289.
Mynatt, E. (1994). Auditory presentation of graphical user interfaces. in G. Kramer (ed.), Auditory Display: Sonification, Audification and Auditory Interfaces (Proceedings of ICAD '94), (Santa Fe), Addison Wesley. pp. 533-556.
Mynatt, E. D. (1997). Transforming graphical interfaces into auditory interfaces for blind users. Human-Computer Interaction 12(1 & 2): pp. 7-46.
O'Shea, T. and Self, J. (1983). Learning and Teaching with Computers. Brighton: Harvester Press.
Robson, C. (1994). Experiment, design and statistics in psychology. 3rd Edition, London: Penguin Books Ltd.
Schwartz, R. L. (1993). Learning Perl. Sebastopol, California: O'Reilly.
Spooner, R. I. W. (1996). A computerised writing aid for dyslexic people, University of York, Department of Computer Science, Research Student Thesis Proposal
Stevens, R., Edwards, A., Linehan, C., McCarthy, J., Weber, G., Bauwens, B., Elsom-Cook, M., Marichal, G. and Deschepper, D. (1996). Handling of mathematics, Tide Maths Project, Deliverable No. D6.
Stevens, R. D. and Edwards, A. D. N. (1996). An approach to the evaluation of assistive technology. in Proceedings of Assets '96, (Vancouver), ACM. pp. 64-71.
Wilson, B. A., Evans, J. J., Emslie, H. and Malinek, V. (1997). Evaluation of NeuroPage: A new memory aid. Journal of Neurology, Neurosurgery and Psychiatry 63: pp. 113-115.
Zetie, C. (1995). Practical User Interface Design. London: McGraw-Hill.
This page maintained by Alistair Edwards alistair@cs.york.ac.uk
http://www.cs.york.ac.uk/~alistair/projects/projects.html