1010101010101010101001100101010
1010101010101010101010101010101010101001100101010100 1010101010101010101010101010101010101001
10101010101010101010101010101 1010101010101010 1010101010101010101010101010101010101 101010101010101010101010 1010101010101010101010101010101010 1010101010101010101010101010101010101001100101010100110
101010101010101010101010
1010101 1010101010101010101010101010101010
Research at the Department of Information Technology
Do y ou learn mor
e fr om r
eading a book than fr
om surfing on the Internet?
Producing steel is an ar
t Softwar
e de velopment f
or and with users Release the po
wer of everyday objects To er r is human… Can computers fix our mistak
Smar t to get computers to w
ork together
Thr ough mediators,
mor e users can access data fr
om man
Wher e in the genome ar
e ther e genes that cooperate?
Technological de
velopment in the ser
vice of humanity
Telemedicine pr
ovides access to medical exper
ts r egar Gains or losses on the stock mark
et?
Computer pr ograms an aid in making complex bids
Production: Ord&Form AB, Uppsala
Photo: Martin Ceije, Roland Grönroos (p. 16) and EyeQnet (p. 25) Copy: Helena Kämpfe Fredén and Ord&Form AB, Uppsala Print: Sandvikens Tryckeri AB, Sandviken
18
4 6 8 10 12 14 20 22 24 26 28
The potential for what can be done in information technology is both fascinating and thought-provoking. In this brochure we present some of the exciting research projects that are under way at the Depart- ment of Information Technology. In the following pages, you will read about how your car will soon be talking to your cell phone, about programs constructed to detect errors or about computers soon being able to figure out what genes cause cancer. And this is just a tiny por- tion, a sampler of the extensive research going on at our department.
The Department of Information Technology is unique in Sweden.
Our undergraduate programs are unusually broad in scope, and in many cases research done here ranks among the best in the world.
The department is a strong and committed collaborative partner in basic research with other institutions, in international projects and not least in cooperation with industry, business, and the public sector.
Work at the department is imbued with the joy of discovery, com- bined with in-depth scientific expertise. It is our hope that some of that feeling will come across in the articles we are publishing here. We also hope that the articles will provide you with a first glimpse of what information technology is all about, and that this will sow the seeds of thought about what our common future will be like. A future
Foreword
Page 3 FOREWORD Page 4 USABLE SYSTEMS Page 6 LEARNING
Page 8 SELF-CONFIGURING NETWORKS Page 10 AUTOMATIC PROGRAMVERIFICATION Page 12 COMPUTER CLUSTERS
Page 14 DATABASE TECHNOLOGY
Page 16 DEPARTMENT FACTS
Page 18 COMPUTING SCIENCE & GENETICS Page 20 KINETIC ANALYSIS
Page 22 TELEMEDICINE Page 24 OPTIONS
Page 26 ALGORITHM DESIGN
Page 28 AUTOMATIC STEEL PRODUCTION Page 30 RESEARCH OVERVIEW
Contents
4 USABLE SYSTEMS
Software development for and with users
User-centered system design is about understanding how people use technology
We have all had trouble using VHS players and cell phones, not to mention maddening Internet services. The only way to improve these is to work alongside users to understand how they use technology. This is what user-centered system design is all about.
Cooperation between system developers and users is not easy. Deve- lopers often feel that users don’t understand what is in their own best interest, that they can’t make up their minds, and that they are often anti-technological to start with. For their part, users think that soft- ware developers are more interested in playing with new technologies than in producing usable systems for them.
User-centered system design is a process that focuses on making systems more usable based on close cooperation between developer and user throughout the entire development cycle. The advantage is not merely attaining higher quality in the products constructed, but
also reducing occupational health problems that may arise owing to defi cient technologies.
To be able to develop user-centered methods and techniques that function in practice, it is crucial for research to be pursued in collabo- ration with industry and other organizations. Researchers at the de- partment work together with IT consultants, authorities, trade unions, and industry on their own everyday problems, helping to incorporate usability efforts in their development. The most recent project invol- ves cooperating with Third World countries that wish to develop user- centered methods. If India, for example, is to develop programs that can be used in the Western world, developers there will have to coo- perate with users who are thousands of miles away, with another cul- ture, other time zones, and other languages.
Jan Gulliksen
“User-centered system design is a process that focuses on making systems more usable based on close cooperation between developers and users.”
6 LEARNING
Do you learn more from reading a book than from surfi ng on the Internet?
The art of learning computer science
How do students learn university-level computer science? This is being studied by a didactics team at the Department of
Information Technology in an ongoing research project. The goal is to understand how learning takes place, in order to provide teachers with ideas for how computer courses should be set up to produce good study outcomes.
The project targets what types of aids students use for learning. Not only that; the researchers are probing how students apprehend the subject matter. Is there any correlation between your choice of study aid, how you use the aid, and what you learn? Study aids in this case can include textbooks, lectures, working with other students, www documents, various forms of computer aids, practical exercises, and lab projects. Different students use different combinations of aids. On top of this, the aids are used in different ways. Can different categories of students be discerned based on what aids they choose and how they use them? If so, is there any correlation between these different cate-
gories and how well they succeed as learners?
The project focuses on introductory courses in object-oriented programming and numerical analysis. The researchers are also looking at more advanced courses where students are learning about compu- ter communication through international student collaboration in practical projects. Conclusions will be primarily based on interviews with students.
The results can lead to recommendations for teachers and others who arrange courses regarding how courses in the subject of compu- ter science should be set up to produce good results. What aids should be provided, how should subject matter be presented? In this way the project can help improve learning outcomes in various types of com- puter science education, not only at the university level, but also in high schools and in further education.
The type of knowledge the project generates is expected to also be useful as a basis for the training of future computer teachers.
Michael Thuné
“Is there any correlation between choice of study aids, how you use study aids, and what you learn? The goal of the project is to understand how learning takes place.”
8 SELF-CONFIGURING NETWORKS
Release the power of everyday objects
Research helps technology communicate freely
Cell phones, digital cameras, and cars are examples of everyday products that have built-in computers. These gadgets are often
‘unaware’ of each other, and this limits the service they can offer us. Now researchers are working on helping different gadgets communicate with each other.
When we move around, these gadgets and their built-in computers are going to be close to other gadgets. In theory they should be able to sense each other and create a network. The best-known example of this linking of computers is the Internet. The Internet is designed to tie together computers that look pretty much the same and can do roughly the same things. But with the number of mobile eve- ryday gadgets growing explosively, the question arises what these things would be able to do if they could talk to each other.
This is the question researchers in computer communication are tackling. They are studying the conditions for linking together compu- ters to various nets of collaborating units with one goal in mind: to facilitate communication.
In principle it is possible to compare the computers that surround us today with people in an organization. Everyone has a special task that, when carried out correctly, adds to the common result. But if we
also share our experiences, knowledge, and resources with one another, we can truly benefi t from working in the same organization and toward the same goals.
If we could get our everyday gadgets with built-in computers to exchange information about capacity and functionality, then they would be able to utilize each other’s services to provide the user with an entirely different kind of service compared to today’s. The technical term for this is ‘self-confi guring networks,’ meaning that the computers themselves manage the exchange of information needed for them to be able to cooperate.
On the way to an environment where technical aids cooperate, de- pending on how close they are to each other, there are a number of challenges for researchers. One project aims to get computers that are to be linked together to dare to trust one another. Imagine taking pic- tures with your digital camera to be sent via your cell phone to a friend. Your cell phone then needs to understand that it should allow the camera to use it to send the pictures. The challenge lies in creating the fi rst sense of trust between the machines, a trust that can then be further strengthened.
Mattias Wiggberg
“One project aims to get computers that are to be linked together to dare to trust one another.”
10 AUTOMATIC PROGRAM VERIFICATION
To err is human… Can computers fi x our mistakes?
Developing software that automatically detects errors
On June 4, 1996, the Ariane 5 launcher exploded less than a minute after take-off. The accident was not the result of a mechanical fault but rather due to an error in the design of the guidance software. At the Department of Information Technology a research team is developing software to automatically detect and circumvent mistakes of this type.
Today it is possible to completely design and verify complex compu- ter chips before the fi rst prototype has been built. First engineers de- sign the chip using specialised software, then a computer can simulate this chip and automatically fi nd weak points with the help of mathe- matical methods.
These developments have been rapid. The fi rst Pentium processors made mistakes when they divided one number by another. Today most makers of computer chips use software to discover and correct
design glitches before the product goes to market. Since chips are getting more and more complex, researchers are forced to steadily improve their methods for making ever faster verifi cation programs. A further area where automatic verifi cation is useful is communications protocols, such as those used in mobile telephones to make it possible for people to communicate with each other. The fi rst generation of mobile phones was limited to voice transfer, but modern equipment can transfer images and video fi lms. New protocols are needed. Every protocol has to be able to guarantee that data is received by the pro- per destination within a reasonable period of time.
There are many other applications as well. The fact that computers are making their way into more and more systems means that the fi eld of new uses is constantly expanding. The need to develop new algorithms is growing apace.
Johann Deneux and Pritha Mahata
“A computer can simulate a chip and automatically fi nd weak points with the help of mathematical methods.”
12 COMPUTER CLUSTERS
Smart to get computers to work together
Collaborative project DSZOOM developing effi cient cluster techniques
Anyone who has ever used a search engine on the Internet has very probably made use of the services of a computer cluster.
The primary advantage of such clusters is that they can solve problems faster while being cheap to construct. Certain types of programs cannot be run effi ciently using today’s clusters, however, new research is bringing online clusters that can meet the challenge.
A computer cluster is two or more computers connected in a network that utilizes cluster software to allow them to work together as a unit.
A good example of a computer cluster is the world’s largest search engine, Google, which consists of more than 10,000 PCs. Computer clusters are not a new concept; they have been in use since the 1970s, when they were primarily used to improve the reliability of computer systems.
Today’s largest servers have around 100 processors in a cabinet and can cost millions of dollars. There are several physical constraints on building bigger servers, including space and generation of heat.
Certain types of programs have nonetheless required large and ex- pensive servers, that is, it has not been possible to run them on cheap clusters. A team of researchers at the Department of Information Tech- nology is now at work creating new cluster software. This work is being done in a project called DSZOOM, which is a collaborative effort in- volving Uppsala University, Sun Microsystems, and an institute that supports cooperation between the academic world and industry, called PSCI (Parallel Scientifi c Computing Institute). The aim of the research is to enable the same program to be run on a DSZOOM cluster as on a larger and much more expensive server. This will make it possible to purchase several smaller computer systems and connect them in a cluster, thus bringing down the total cost of the system considerably.
Zoran Radovic and Håkan Zeffer
“Certain types of programs can’t be run effi ciently using today’s clusters. The aim of the research is to enable the same program to be run on a DSZOOM cluster as on a
14 DATABASE TECHNOLOGY
Through mediators, more users can
access data from many different databases
Mediators make it easier and faster to seek complex information
Using databases available on the Internet you can fi nd answers to questions ranging from medicine to space technology.
Sometimes you want to search for and combine data from several different databases. To make it easier and faster to develop such searches across different types of data, researchers at the Uppsala DataBase Laboratory have developed a new technique, so-called mediators, that mediate information from several different data sources and combine them to achieve a ready answer.
Mediators are a kind of software for retrieving information from dif- ferent external data sources. Through the mediator the user sees the data in a special display. This can be likened to a virtual database whose structure and content has been culled and converted from the different data sources spread out across the Internet. The user can easily ask questions within the fi eld he or she is working in, such as astronomy or medicine. In this way the user and the database can communicate in the same ‘language.’
Since data sources differ, a mediator system has to know how dif- ferent displays can be calculated from the contents of the sources
and how to deal with similarities and differences between similar data in various sources. What’s more, mediators must take into consi- deration the fact that the databases can be distributed across the en- tire Internet. Thanks to special technology, several mediators across the Internet can collaborate and utilize each other’s knowledge. This makes it feasible to construct effi cient software systems for integra- ting large and complex distributed amounts of data.
Researchers are developing mediator technology for space physics, medicine, engineering science, and education. For space applications, for example, a database manager is being developed for questions about data produced by a distributed digital radio-telescope being developed in Europe. In medicine they are dealing with database questions addressing parts of brain images.
Modern product design requires access to and cooperation bet- ween many different kinds of distributed data. For educational pur- poses it is necessary to access and combine many divergent sorts of data described using an Internet meta-data standard called the ‘se- mantic web.’
Tore Risch
“Thanks to special technology, several mediators across the Internet can collaborate and utilize each other’s
knowledge.”
Department of Information Technology at Uppsala University
Education and research of top international quality
tion at the department. The department is building on activities in IT that have been carried on at Uppsala University since the mid 1960s.
Research
In many cases work pursued here is on the cutting edge internatio- nally, in both pure and applied research. The department is active in a large number of collaborative projects across various fields of business and the public sector, as well as the academic world. In all, some 100 Ph. D. students and 25 full professors are engaged in different research projects. Many of the professors come from industry with continuing ties. External funding covers 70% of the entire research budget.
A number of research centers are associated with the department:
ASTEC (Advanced Software Technology), ARTES (A network for The Department of Information Technology is unique in it´s
breadth. Our activities span from the gathering and management of data via signal processing, computational and control
engineering to communication of results with the aid of database management and human-computer interaction. A common ground is provided by research in theoretical computing science, real-time systems, and computer architecture.
World class
Education and research at the department are of the highest interna- tional quality. Roughly 4,000 students study here each year, and about 30 research teams are based here. The strong focus on research im- pacts and provides an excellent foundation for undergraduate educa-
Real-Time Research and Education in Sweden), and PSCI (Parallel and Scientifi c Computing Institute), as are centers of excellence: the national graduate school Mathematics & Computing (FMB) and UPP- MAX (Uppsala Multidisciplinary Center for Advanced Computatio- nal Science). The Center for Image Analysis is run in collaboration with the Swedish University of Agricultural Sciences conducting re- search in computerized image analysis.
Facts in brief
• 4,000 students annually
• 240 employees, including PhD students
• 80 teachers, including 25 full professors, who divide their time between research and teaching
• over 100 PhD students
• 14 programs of study, a total of 150 course offerings per academic year
• 43 PhDs and 53 licentiate degrees between 1999 and 2003
• 260 computer work stations for students
• 3 research centers and 2 centers of excellence associated with the department
• 1 member of the Royal Academy of Science
• 3 members of the Royal Academy of Engineering Science
the development of new systems, technologies, and applications in IT. Computer Science, a four-year program, offers a high degree of choice of specialization and courses and leads to a master of science in computer science. Systems in Technology and Society is a master’s in engineering program focusing on how technology functions in so- ciety and contains in-depth study in the arts and social sciences. Other programs with various degrees of choice and profi ling are the Natural
18 COMPUTING SCIENCE & GENETICS
Where in the genome are there genes that cooperate?
Interdisciplinary project analyzing genetic data
Cancer is one example of a partially hereditary disease that is associated with several different genes. The risk of contracting cancer is a so-called polygenic trait. At the Department of Information Technology a team of researchers is developing computational methods for determining where in the genome there are genes that infl uence polygenic traits in animals.
What we look like, how quickly we grow, and how susceptible we are to disease depends largely on our genes. Traits that are infl uenced by several genes are called polygenic traits. An individual´s risk of developing cancer is one example of such a trait, meat quality in pigs is another. The project is a collaboration between researchers in gene- tics, statistics, and scientifi c computing. Genetic data from a group of animals is used to formulate a statistical problem. The computational
task is to calculate the solution to this problem, i.e. to fi nd out where in the genome the interesting genes are located.
The study of both meat quality and cancer is complicated by the fact that the underlying genes can work together as a team. This me- ans that we must look for several genes simultaneously, which entails gigantic multidimensional optimization problems. One can compare this to selecting players for a soccer team. If there are 25 players to choose from, there are more than 4 million possible ways to pick a team of 11 players. It is impossible to test all teams, but on the other hand it is not enough to study how single players handle the ball on an empty fi eld, the most important thing for the success of the team is how the players perform together. The genetic analysis is further complicated by environmental infl uences on the traits, and by the fact that we do not know how many genes are involved in each case.
Kajsa Ljungberg
“What we look like, how quickly we grow, and how
20 ANALYSIS OF HUMAN MOTION
Technological development in the service of humanity
Systems engineers help patients with neurological injuries
Humans can instantly recognize another human being by that person’s gait. Describing the movements of that person
scientifi cally is a gigantic problem. Scientists at the Department of Information Technology are busy analyzing patterns of
movement with the objective of helping people with neurological injuries or joint prostheses.
Research related to human motion has been pursued for more than 100 years, and the basic problem remains the same: how to describe the mechanics of the human being’s complicated locomotor system.
To formulate the mechanical equations is a major problem in itself. To go on to solve those equations requires advanced methods from nu- merical analysis and fast computers to do the calculations. Research in the fi eld can be said to be technological development in the service of
humanity. It’s about analyzing patterns of movement in children with neurological injuries and people who have had their patterns of move- ment altered by stroke, Parkinson’s disease, hip or knee prostheses, or amputation.
Most treatments in orthopedics and physical therapy are intended to enhance the patient’s function, and this makes it crucial for gait analysis to determine the patient’s mobility before, during, and after a treatment. Researchers at the department collaborate with the motor laboratory at Uppsala University Hospital (Akademiska sjukhuset) and with orthopedic specialists on how best to amputate to ensure the best possible rehabilitation. CP in children is the application that has made the greatest progress, and in the US it is now accepted clini- cal practice to perform gait analyses ahead of operations.
Håkan Lanshammar
“It’s about analyzing patterns of movement in people who have had their locomotor system altered by stroke, Parkinson’s disease, hip or knee prostheses, or amputation.”
22 TELEMEDICINE
Telemedicine provides access to medical experts regardless of where you live
The bottom line is how usable the technology can be made
Medical images are used in telemedicine for remote consultation.
Diagnosing patients on the basis of images is not easy. This is why it is important to design the technology so that it does not place any barriers between humans and computers.
Today many hospitals produce digital medical images. One example is computer tomography, which makes it possible to create pictures that show the structure and cavities of the brain. Another is magnetic reso- nance, which can yield extremely sharp and detailed images of organs and the skeleton. These pictures contain information that is essential for correct diagnosis, but they are not easy to interpret.
The images are often sent to an experienced specialist for consulta- tion. It is hoped that telemedicine, sending pictures via the Internet, for example, will make these activities more effi cient. Another impor- tant aspect is greater equality, in that access to expert knowledge is made universal, regardless of where you live or happen to be.
The drawback has been the diffi culty associated with providing usable computer support in health care. The systems available have proven to be hard to use or simply have not suited the complex envi- ronment that care providers work in.
This is why the design of the computer systems is crucial to getting results. With knowledge about design and human beings, specialists in human-computer interaction can devise and construct systems that work better in health care.
CHILI is one example of how human-computer interaction has helped construct a system that is simple to use. The system was desig- ned with usability and personnel effi ciency in mind. It turned the most common needs into simple moves—sending images for consulta- tion requires only three clicks. More than 7 million pictures have been sent back and forth using the system.
Erik Borälv
“With knowledge about design and human beings, specialists in human-computer interaction can devise and construct systems that work better in health care.”
24 OPTIONS
Gains or losses on the stock market?
Equations solved on the computer can fi nd the right price of an option
Anyone trading in options is most often speculating in the ups and downs of the stock market. With the help of computers, advanced equations can calculate the right price of an option.
A so-called European call option gives the owner the right to buy a certain stock at a specifi c price at a pre-determined time in the future.
For anyone trading in options, it is extremely important to be able to calculate quickly and readily what the value of an option should be.
For the simplest options there are formulas that solve this problem.
But for more complicated options there are no ready-made formulas.
In these cases, it is necessary to turn to the computer for help.
One way to use a computer to help price a European call option is to solve the partial differential equation (PDE) that was presented in 1976 by the American researchers Black, Scholes, and Merton. By fi nding an approximate solution to this PDE, you get a reasonable price for this option. Scholes and Merton were awarded the Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel in 1997, also known as the Nobel Prize in Economics, for their pione- ering work in the fi eld.
The method can be generalized so that it is possible to fi nd a PDE
that describes the value of a so-called portfolio option, which gives the owner the right to buy several different stocks. This equation is multidimensional, since it is based on all of the relevant stocks. Re- searchers at the Department of Information Technology solve these PDEs based on several stocks simultaneously using so-called fi nite difference methods. In many cases this requires the solution of several very large systems of equations with several million unknown variab- les. The size of the equation systems grows dramatically with the number of stocks involved. The solution can take a long time to calcu- late, even for today´s fastest computers, if it is even possible to fi t the system of equations in the computer memory. To reduce the size of the systems of equations, and thereby cut the time needed to solve the problem, researchers at Uppsala have developed adaptive met- hods. These methods keep the computational errors made under con- trol. It is thereby possible to calculate more carefully in segments that require extra caution and less carefully in other segments. This makes it possible to arrive at a solution to the PDE much more rapidly, and problems that were previously impossible to solve, owing to the com- puter memory constraints, can now be solved.
Jonas Persson
“For anyone trading in options, it is extremely important to be able to calculate quickly and easily what the value of an option should be.”
26 ALGORITHMDESIGN
Computer programs an aid in making complex bids
Algorithm design creates new negotiating techniques
In electronic commerce it is important to deal with bids like: “If I am awarded contracts 2 and 9 I will give a 5 percent discount.”
These so-called combinatorial bids lead to optimization problems that are diffi cult to solve. But, with the aid of theoretical
algorithmic research and knowledge of real-life negotiation situations, it is possible to develop mechanisms that allow a controlled type of combinatorial bidding.
Algorithm design is about constructing computer programs solve pro- blems and carry out tasks rapidly and effi ciently. In many cases huge amounts of data and complex issues are involved, where the human brain wouldn’t stand the slightest chance of fi guring out the solution.
A clear example can be found in advanced electronic commerce. In a negotiation where a few suppliers are competing for a number of cont- racts, there are often great advantages in allowing combinatorial bids like: “I want to compete for all ten contracts, but I can only shoulder four of them at most.” If we have a hundred contracts and ten suppliers, the number of possible combinations of suppliers and contracts
amounts to 1020.
Combinatorial bidding belongs to a class of problems in which it is extremely diffi cult to guarantee that you will fi nd a solution in a short
time. Generally speaking you cannot even be certain that you will fi nd a good approximate solution. Most researchers believe that sometime in the future it will be possible to prove that no general and rapid algo- rithm can exist.
Nevertheless, to be able to deal with situations involving these com- plicated calculation problems, a number of successful approaches can be brought to bear:
• By combining algorithmic methods with knowledge of real-life negotiation situations, it is possible to develop methods that can deal with many types of negotiations in practice.
• Based on research in a number of disciplines, like microeconomics, game theory, and algorithm design, it is possible to develop speci- ally adapted market mechanisms that allow a controlled type of combinatorial bidding. At the same time necessary optimization can be effi ciently achieved.
The research being pursued at the Department of Information Tech- nology is not only intellectually stimulating but moreover results in industrial spin-off activities, where methods and techniques have be- come advanced IT products.
Arne Andersson
“I want to compete for all ten contracts, but I can only shoulder four of them at most.” If we have a
28 AUTOMATIC STEEL PRODUCTION
Producing steel is an art
Doing it automatically is information technology
Today steel is produced mostly by the converter process, an expensive and energyintensive method. Now the next generation of steering systems is being brought forth in a research project with metallurgists and control engineers working together. This is how tomorrow’s steel production will be less energy consuming, more environmentally friendly, and safer.
The converter process has been in use for more than 50 years and today it is the overwhelmingly dominant method for producing steel.
The purpose of the process is to reduce the amount of carbon and silicon and other impurities in the raw iron coming from the blast furnace. This is done by blowing pure oxygen against the molten metal’s surface and oxidizing the impurities. The cost of running the converter process is high, primarily because it is slow and uses huge quantities of pure oxygen.
The goal of converter process control is to achive the proper che- mical composition of the steel and a given temperature for the molten metal. Since neither the temperature nor the content of carbon or silicon can be measured in real time, the process is steered manually.
A skilled and experienced blower knows how to make use of the mea- surable and perceptible quantities to estimate what the consequences of steering measures would be.
Blowing oxygen at supersonic speed on molten steel at 1,700°C is no joke, and many things can go wrong. Therefore, an automatic sys- tem for steering the converter process is high on the wish list of the steel industry.
At the Department of Information Technology control engineers are collaborating with metallurgists from the Royal Institute of Tech- nology in Stockholm in a research project that aims to produce a new reliable and effi cient system. To develop such a system, fi rst all chemi- cal and physical phenomena that occur in converter blowing need to be described by equations. Mathematical modeling not only provides insights into how the process works but also into how the process can be improved and where the limits of the technology lie. This is where close cooperation is needed between metallurgists, control engineers, physicists, and mathematicians.
Alexander Medvedev
“Blowing oxygen at supersonic speed on molten steel at 1,700°C is no joke, and many things can go wrong. Therefore, an automatic system for steering the converter process is high on the wish list of the steel industry.”
Algorithms and Data Structures
Algorithms and Software for High-performance Computers Automatic Control
Bioinformatics Biomedical Engineering Complexity Theory
Computational Aspects of Financial Mathematics Computational Electromagnetics
Computer Science Education Computer Architecture Computer Networks Computer Tomography Compiler Technology
Twelve articles don’t tell the whole story
We haven’t mentioned many other fields of research
The purpose of this brochure has been to give a brief glimpse of the scope of ongoing research at the Department of Information Technology. For reasons of space we can’t tell the whole story of the various projects, nor can we talk about all the research being pursued. There are actually a large number of other fields covered by research at the department. Most of these are listed below, although we still haven’t listed them all. If you want to know more or stay abreast of developments in our work, you are welcome to visit our home page at www.it.uu.se/research.
Data Bases Distributed Systems
Design and Evaluation of Usability E-commerce
Estimation Fault Detection Fluid Dynamics
Formal Methods for Specification and Analysis of Programs Human-Computer Interaction
Image Processing Learning Systems
Mechanical Systems Control Real-time Systems
Process Control with Applications to Environment and Medicine Signal Processing
Spectral Analysis
Radar and Synthetic Aperture RadarResource Alloccation Scientific Computing
Software Technology
Support Systems in Working Life Applications System Identification
Verification
Wireless Communication
ORD&FORM AB, UPPSALA
Department of Information Technology PO Box 337
