Fuzzy Systems and Their Application

to Information Assurance,

Geosciences, and Environmental Sciences:

Summer 2010

**Instructors:**

- Vladik Kreinovich, email
vladik@utep.edu

office COMP 215, office phone (915) 747-6951

office hours: MTWR 9-9:30 am, MTW 10:35-11 am, or by appointment. - Luc Longpre, email longpre@utep.edu

office COMP 227, office phone (915) 747-6804

office hours: TR 2:30-4 pm, or by appointment

** General motivations.** Computers have been actively applied in many
areas of science and engineering. For example, to predict weather,
we simulate how the weather parameters change -- by programming
the corresponding numerical models.

However, it is well known that models and equations are only a part of knowledge. For example, as every person who works in applications knows, often mathematically correct solutions to these equations make no physical sense. We need an expert -- e.g., a professional geologist -- to separate physically meaningful from physically senseless solutions.

A medical doctor uses a large amount of data to make a decision, but in most cases, we cannot replace the doctor with an algorithm: there is an important part of the doctor's expert knowledge that cannot be easily described by an algorithm.

There is no need to invoke experts to explain the need for expert knowledge: most people can drive cars, but it is very difficult to design an algorithm for driving cars.

The difficulty is that we can easily explain our rules for driving -- just like a medical doctor can often explain her reasons for making a good diagnostic decision -- but this explanation will be often formulated by using words from natural language, like "close" (the car in front of me was too close, so ...), or "high" (the patient's fever was high, so ...). So, for a computer program to use this knowledge, we must be able to describe such words in terms that a computer can understand and process - i.e., in terms of numbers.

Fuzzy logic is a general name for different techniques that transform knowledge described by words from natural language into a numerical algorithmic form. Some of these techniques are similar to probabilistic and statistical techniques traditionally used in science and engineering, some use more sophisticated approaches such as multi-valued logic -- with additional truth values like "probably true" or "very possibly false").

These are the techniques that we will study in this course.

**Why applications to geosciences and to environmental sciences.**
Geosciences and environmental science are two examples of
practically important research areas in which, in principle, we
know the equations but the data is sparse and thus, we need to
supplement it with expert knowledge.

- In geosciences, we have a large amount of surface data, but it is much more difficult to get data from the depths.
- In environmental sciences, populated areas of the Earth are mostly well covered by sensors, but areas like Arctic or Antarctica -- not to mention the oceans -- are barely covered.

**Why applications to information assurance.** Information
assurance -- securing computer communications, securing integrity
and confidentiality of the data, etc. -- are important but
extremely difficult areas of research, difficult because, as everyone
understands, good protection algorithms help, but we still need
the expertise of a skilled system administrator.

Information assurance is one of the main research and education directions of our department:

- In the College of Engineering, we have an official Center for Information Assurance of which Luc Longpre is a Director, and
- UTEP has recently received designation as a National Center of Academic Excellence in Information Assurance Education.

**Why study fuzzy techniques now.** On March 18-20, 2011,
one of the major international conferences on fuzzy techniques,
an annual conference organized by the North American Fuzzy
Information Processing Society, will be held in El Paso. This
conference is usually sponsored by IEEE, as a result of which
Proceedings of this conference become part of the widely
used IEEE IExplore publications database. This is an excellent
opportunity to present our results.

We expect that many students taking this class will not only learn the new material, they will also participate in ongoing research projects -- and several of these projects will eventually hopefully lead to co-authorship of papers published in the proceedings of this conference. The plan is that graduate students who are already working in these directions will lead groups of interested students from the class who volunteer to help with their projects.

**Handouts instead of a textbook.** Our emphasis in this course are on
new applications of fuzzy techniques, specially on application to
information assurance, geosciences, and environmental sciences.
While there are good textbooks on fuzzy techniques in general,
the applications in which we are interested
are so new that they have not yet been adequately
reflected in these textbooks. As a result, in this class, instead of
a textbook, we will use research papers and other handouts.

**What we expect from the students taking this class.**
As we have mentioned, the main purpose of fuzzy techniques is to
use numerical methods to process expert information, and that
many ways of generate the corresponding numbers are similar to the
methods used in probability and statistics. In view of this, for
taking this class, it is helpful to know the basics of
numerical methods and the basics of probabilities and statistics.
While having the corresponding class as a pre-requisite is desirable,
it is not required: students who still remember the basics of
probability from their Precalculus and Discrete Math courses are
also welcome.

**Non-CS students.** Due to the inter-disciplinary character of
potential applications, we welcome not only CS students, but
also students from geosciences, environmental sciences,
computational sciences, and from other disciplines
in which fuzzy techniques can be (and are) useful such as
bioinformatics, engineering (especially control), etc. The presence
of such students will help application problems.

We realize that some non-CS students may not be as good in programming different fuzzy-related algorithms as our own CS students. With this in mind, we plan to tailor homeworks and other assignments to such students, so that these tailored assignments are more focused on applications and somewhat less on programming.

**Plan of study:** In the first part of the class:

- we start with describing the basic fuzzy techniques;
- after this, we briefly describe the relevant problems of geosciences and environmental sciences;
- we will also describe what is cyberinfrastructure and how it can be useful in these (and other) applications;
- then, we learn how fuzzy techniques can be used in solving applications problems of science and engineering -- with a special emphasis on geosciences and environmental sciences.

- we will briefly introduce the main techniques and challenges related to information assurance; and
- we analyze how fuzzy techniques can be applied to these problems.

- some students may select to work with graduate students who have already been working on relevant projects -- with the objective of producing high-quality research results;
- some students may want to select a project related to an application of fuzzy techniques to their own area of interest; we will be glad to help these students;
- a more routine type of project will consist of a report on a published technical paper on fuzzy techniques and/or their applications.

**Tests and Grades:** There will be two tests and one final exam.
Each topic means home assignments (mainly on the sheets of paper,
but some on the real computer). Some of them may be graded.
Maximum number of points:

- first test: 10
- second test: 25
- home assignments: 10
- final exam: 35
- project: 20

A good project can help but it cannot completely cover possible deficiencies of knowledge as shown on the test and on the homeworks. In general, up to 80 points come from tests and home assignments. So:

- to get an A, you must gain, on all the tests and home assignments, at least 90% of the possible amount of points (i.e., at least 72), and also at least 90 points overall;
- to get a B, you must gain, on all the tests and home assignments, at least 80% of the possible amount of points (i.e., at least 64), and also at least 80 points overall;
- to get a C, you must gain, on all the tests and home assignments, at least 70% of the possible amount of points (i.e., at least 56), and also at least 70 points overall.

**Online class evaluations:** Close to the end of the class,
a website will be made available to the students for online
class evaluations. These evaluations are an important part of
the class. While remarks that you type in are strictly
anonymous, we the faculty will be informed on who submitted
their evaluations and who did not. To encourage you to submit
your evaluations on time, we will give extra points to those
who submit the class evaluations ahead of the deadline.

**Standards of Conduct:** Students are expected to conduct
themselves in a professional and courteous manner, as
prescribed by the Standards of Conduct. Students may discuss
programming exercises in a general way with other students, but
the solutions must be done independently. Similarly, groups may
discuss project assignments with other groups, but the
solutions must be done by the group itself. Graded work should
be unmistakably your own. You may not transcribe or copy a
solution taken from another person, book, or other source,
e.g., a web page). Professors are required to - and will -
report academic dishonesty and any other violation of the
Standards of Conduct to the Dean of Students.

**Disabilities:**
If you feel you may have a disability that requires
accommodation, contact the Disabled Student Services Office at 747-5148, go
to Room 106 E. Union, or e-mail to
dss@utep.edu.