I
S.89-74
40 ?
SIMON FRASER UNIVERSITY
MEMORANDUM
To: ?
Senate
?
From:
L. Salter
Chair, SOAP
Subject:
Curriculum
School of Computing
revisions
Science -
?
Date:
?
November 16, 1989
Reference: SOAP 89-56
Action undertaken by the Senate Committee on Academic Planning/Senate Graduate
Studies Committee gives rise to the following motion:
Motion:
"That Senate approve and recommend approval to the Board of Governors
as set forth in S.89-74 the curriculum revisions including
Foundations of Programming Languages
Database Systems
Operating Systems
Robot Vision
Query Processing in Database Systems
Automated Learning and Reasoning
Functional Programming
Principles of Database and Knowledge-Base
Systems
Algorithmic Graph Theory
Natural Language Processing
Expert Systems
Advanced Topics in Simulation and Modelling
Biomedical Computing
Computer Mapping
Advanced Seminar II
Advanced Seminar Ill
Directed Reading II
New courses
?
CMPT73O-3
.
?
CMPT740-3
CMPT76O-3
* CMPT821-3
CMPT841 -3
CMPT 826 - 3
CMPT 831-3
CMPT843-3
CMPT814-3
CMPT825-3
CMPT827-3
Deletion of CMPT 840 - 3
CMPT 861-3
CMPT862-3
CMPT892-3
CMPT893-3
CMPT895-5
* Registrar's Note:
This is a modernization of the existing course
CMPT 821 "Pattern Recognition and Image Processing"
SIMON FRASER UNIVERSITY
School of Computing Science
?
MEMORANDUM
Graduate Studies Committee,
?
FROM: Joseph Peters,
Faculty of Applied Sciences
?
Director of Graduate Studies,
School of Computing Science
Graduate Program Reorganization
?
DATE: September 21, 1989
The attached document is a proposal for reorganization of the graduate programs in the School of Computing
Science. Our
M.Sc.
program is now 10 years old and our Ph.D. program is 7 years old. Our enrolment in the 89-3
semester is approximately 47 M.Sc. students and 31 Ph.D. students. To date we have graduated approximately 75
M.Sc. students and 8 Ph.D. students. Our graduate programs have not been changed in any major ways since they
were introduced. With the benefit of our experience, we feel that it is time for some changes. Since the number of
changes is large, the proposed structures of our graduate programs are presented as a new calendar entry followed
by explanatory notes. The notes discuss the purposes of the various components of the programs and the ways in
which they meet the goals below. It is hoped that this form of presentation will focus attention on the coherence of
the proposed organization better than a tedious before and after comparison with the previous organization. The
main goals of this reorganization are the following.
• ?
1. ?
Reduce the average time for M.Sc. and Ph.D. students to complete degree requirements.
2.
Increase the enrollment in our graduate courses.
3.
Ensure that both M.Sc. and Ph.D. students obtain reasonable breadth.
4.
Simultaneously reduce the trauma and increase the accuracy of the Ph.D. breadth requirement (currently
Comprehensive Examinations).
5.
Introduce more flexibility of movement between the M.Sc. and Ph.D. programs so that new students can be
more accurately placed.
6.
Encourage students to start thinking about research topics earlier in their graduate programs.
The major changes are as follows.
I. Six courses have been identified as M.Sc.-level courses stressing broad coverage of a research area. Four of
the courses are existing courses that have been renumbered with 700 numbers to distinguish them from 800-
level research topics courses. The other two 700-level courses are new courses.
2.
M.Sc. students are requited to choose at least 3 of their courses from the 700-level offerings.
3.
Ph.D. candidates can now satisfy some of their breadth requirement by taking 700-level courses.
4.
A course requirement of 9 semester hours has been introduced for Ph.D. candidates.
5.
A Research Topics Seminar series has been introduced to encourage students to think about research prob-
lems earlier in their graduate programs.
.
TO:
RE:
6.
The first two semesters of the M.Sc. and Ph.D. programs have been designed together to permit easy transfer
of exceptional M.Sc. students into the Ph.D. program. Transfers from the Ph.D. program to M.Sc. program
are also possible with little lost time.
7.
Six courses that are no longer offered have been eliminated and 11 new courses (including 3 new 700-level
courses) which match the current research interests of the faculty have been introduced to replace them.
Most of the new courses are currently being offered as Special Topics courses. Five courses have also been
renumbered to accommodate other changes. The titles and/or abstracts of 12 existing courses have been
modified to be more accurate and descriptive.
.
C
I
.
Part I.
?
Calendar Description.
For descriptions of Faculty and Areas of Research, Research Facilities, Laboratory for Computer and Com-
munication Research, and Degrees Offered, see the 1989-1990 calendar entry.
Admission Requirements
To qualify for admission to the M.Sc. program, a student must satisfy the general university regulations and
must have a Bachelor's degree, or the equivalent, in Computing Science or a related field. To qualify for admission
to the Ph.D. program, a student must satisfy the general university regulations and must have a Master's degree, or
the equivalent, in Computing Science or a related field.
To qualify for admission to the Ph.D. program before completion of an M.Sc. program, a student must be
enrolled for at least two semesters in the M.Sc. program, must complete the M.Sc. Course Work requirement, and
must complete the breadth requirements for the Ph.D. program, as described below, within 12 months of first
enrollment in the M.Sc. program. Permission is also required from the Computing Science Graduate Program
Committee and the Senate Graduate Studies Committee.
For further information, refer to Section 1.3 of the
Graduate General Regulations.
•
?
Degree Requirements - M.Sc. Program
1.
Course Work
The minimum course requirement for the M.Sc. degree is 18 semester hours of graduate level course credit.
At least 15 of these 18 semester hours must be in Computing Science and at least 9 of these 18 semester hours must
be 700-level Computing Science courses. The Graduate Program Committee may require additional undergraduate
or graduate courses or a different number of 700-level semester hours in certain circumstances.
2.
Research Topics Seminar Series
The Research Topics Seminar Series is presented in the Fall and Spring semesters to acquaint graduate stu-
dents with the research interests of the faculty. All M.Sc. students are required to attend the Research Topics Sem-
inars presented during their first year of enrollment in the M.Sc. program.
3.
Research
(a)
The student is required to submit a Research Proposal to the School for approval by the student's Supervisory
Committee. The Supervisory Committee consists of the student's Senior Supervisor, at least one other
faculty member from the School of Computing Science, and other committee members as appropriate.
(b)
The student is required to present a seminar based on his/her thesis research.
(c)
The student is required to submit and defend a thesis based on independent work by the student. For regula-
tions governing the composition of M.Sc. Examining Committees and the conduct of M.Sc. Thesis Examina-
tions,
see Sections 1.9 and 1.10 of the
Graduate General Regulations.
3
Degree Requirements - Ph.D. Program
?
.
1. ?
Breadth Requirement
A candidate for the Ph.D. degree must satisfy the Graduate Program Committee that he/she possesses suit-
able breadth of knowledge in Computing Science. Normally, this requirement is satisfied during the first year of
enrolment through a combination of courses and Comprehensive Examinations as described below although excep-
tions are possible at the discretion of the Graduate Program Committee in consultation with the Ph.D. Breadth
Committee. The Ph.D. Breadth Committee is elected annually by and from the faculty of the School of Computing
Science.
A Ph.D. candidate must demonstrate breadth in Theoretical Computing Science and four of the following
five areas: Artificial Intelligence, Programming Languages and Systems, Database Systems, Computer Design and
Organization, Computer Systems. Breadth in an area can be demonstrated by taking the 700-level course in the
area or by writing the Comprehensive Examination in the area. In most cases, 700-level courses taken while in the
M.Sc. program can be used to satisfy the Ph.D. Breadth Requirement. Breadth in at least two areas must be
satisfied by writing Comprehensive Examinations. The Comprehensive Examinations test breadth of knowledge at
the M.Sc. level, and are normally written at the end of April during the first year of enrolment. A candidate should
inform the Ph.D. Breadth Committee, in writing, no later than the end of January of the first year of enrolment,
which examinations he/she intends to write.
The standing of a Ph.D. candidate is reviewed within 12 months of first enrollment by the Ph.D. Breadth
Committee. If the Ph.D. Breadth Committee decides that the candidate has not satisfied the breadth requirement, it
may ask the candidate to
(i)
do additional courses, projects, exams, or other work,
(ii)
transfer to the M.Sc. program, or
(iii)
withdraw from the University.
?
1
The second of these recommendations must also be approved by the Graduate Program Committee and the Senate
Graduate Studies Committee.
2.
Course Work
In most cases, Ph.D. candidates will be required to take 9 semester hours of Computing Science graduate
level course credit during the first year of enrolment in the Ph.D. program. The Graduate Program Committee may
waive part or all of this requirement, or may require additional undergraduate or graduate courses in certain cir-
cumstances.
3. ?
Research Topics Seminar Series
The Research Topics Seminar Series is presented in the Fall and Spring semesters to acquaint graduate stu-
dents with the research interests of the faculty. All Ph.D. students are required to attend the Research Topics Sem-
inars presented during their first year of enrollment in the Ph.D. program. Ph.D. students who have satisfied this
requirement while in the M.Sc. program are not required to satisfy it again.
4.
?
Research
The major portion of the Ph.D. program consists of original research under the direction of a Supervisory
Committee. A Ph.D. Supervisory Committee consists of a Senior Supervisor, at least one other faculty member
3
0 ?
from the School of Computing Science, and other committee members as appropriate.
(a)
The student is required to pass a Research Area Examination in his/her chosen research area. The student
must demonstrate that he/she possesses both the ability, and adequate knowledge of the chosen research area,
to pursue and complete original research at an advanced level. The Research Area Examination is conducted
by the student's Supervisory Committee and should normally take place within six months after the student
has satisfied the Breadth Requirement.
(b)
The student is required to submit a written Research Proposal to the School for approval by the student's
Supervisory Committee. The Supervisory Committee must be satisfied that the proposed research is
appropriate in level and scope for a Ph.D. thesis.
(c)
The student is required to present a seminar based on his/her thesis research. This seminar is normally
presented a few weeks before the candidate's thesis defence.
(d)
The student is required to submit and defend a thesis based on substantial original research. For regulations
governing the composition of a Ph.D. Examining Committee and the conduct of Ph.D. Thesis Examinations,
see Sections 1.9 and 1.10 of the
Graduate General Regulations.
For further information and regulations for both the M.Sc. and Ph.D. degrees, refer to the
Graduate General
Regulations.
Graduate Courses
CMPT 601
. 5 ?
Computing Science Education I
CMPT 602-5 ?
Computing Science Education II
CMPT 710-3 ?
Computational Complexity
(formerly CMPT 810)
This course provides a broad view of Theoretical Computing Science with an emphasis on complexity theory.
Topics will include a review of formal models of computation, language classes, and basic complexity theory;
design and analysis of efficient algorithms; survey of structural complexity including complexity hierarchies, NP-
completeness, and oracles; approximation techniques for discrete problems.
CMPT 720-3 ?
Artificial Intelligence
(formerly CMPT 820)
Artificial Intelligence brings concepts such as computation, process, sub-procedure, data structure, and debugging
to bear upon questions traditionally raised by Psychologists, Linguists, and Philosophers. In this course we will
study a representative sample of work in the field. This will include programs which process written English, "see,
play games, prove theorems, and solve problems.
Exclusion: CMPT 410.
CMPT 730
. 3 ?
Foundations of Programming Languages
This course will cover basic concepts in the area of programming languages. The course will be largely of a
theoretical nature and will concentrate on fundamental concepts of lasting importance, rather than topics of current
interest.
CMPT 740
. 3 ?
Database Systems
I
5-
4
Review of introductory database concepts; query optimization; concurrency control; reliability and crash recovery;
distributed databases; object-oriented databases; knowledge base management systems.
CMPT
750-3 ?
Computer Architecture
(formerly CMPT 850)
Parallel processing: SIMD & MIMD systems, pipelining, data flow architecture; microprogramming; control
memory minimization, optimization and verification of micro-programs.
CMPT 760-3 ?
Operating Systems
This course will discuss design issues relating to the functionality and performance of modem workstation operat-
ing systems, such as methods for sharing memory, file and data objects, and choice of communication protocols.
The special needs of high-performance multiprocessor systems and real-time systems will also be addressed.
CMPT 811-3 ?
Distributed Algorithms
This course is an introduction to computation in distributed systems with an emphasis on the design and analysis of
distributed algorithms. We will study many of the distributed algorithms that have been proposed for such prob-
lems as election, selection, sorting, spanning trees, and routing. Several models of distributed computing will be
discussed.
CMPT 812-3 ?
Parallel Computation
This course is a theoretical treatment of parallel complexity theory concentrating on algorithms and models.
Topics will include models of parallel computation, parallel complexity hierarchies, basic tools and techniques for
the construction of parallel algorithms, and selected advanced topics.
CMPT 813-3 ?
Computational Geometry
This course covers recent developments in discrete, combinatorial, and algorithmic geometry. Emphasis is placed
on both developing general geometric techniques and solving specific problems. Open problems and applications
will be discussed.
CMPT 814-3 ?
Algorithmic Graph Theory
Algorithm design often stresses universal approaches for general problem instances. If the instances possess a spe-
cial structure, more efficient algorithms are possible. This course will examine graphs and networks with special
structure, such as chordal, interval, and permutation graphs, which allows the development of efficient algorithms
for hard computational problems.
CMPT 815-3 ?
Algorithms of Optimization (formerly CMPT 860)
This course will cover a variety of optimization models, that naturally arise in the areas of Management Science
and Operations Research, which can be formulated as Mathematical Programming problems.
CMPT 821-3 ?
Robot Vision
This course discusses issues and research results pertinent to robot vision. Topics include depth recovery, for robot
navigation, three-dimensional object recognition and scene analysis, model-based approach, parallel vision
machines and algorithms, and case study of comtemporary robot vision systems.
CMPT 822-3 ?
Computational Vision
A seminar based on the Artificial Intelligence approach to vision. Computational vision has the goal of discovering
the algorithms and heuristics which allow a two-dimensional array of light intensities to be interpreted as a three-
dimensional scene. By reading and discussing research papers - starting with the original work on the analysis of
.
.
S
line-drawings, and ending with the most recent work in the field - participants begin to develop a general overview
of computational vision, and an understanding of the current research problems.
CMPT 823-3 ?
Formal Topics in Knowledge Representation
This course surveys current research in formal aspects of knowledge representation. Topics covered in the course
will centre on various features and characteristics of encodings of knowledge, including incomplete knowledge,
nonmonotonic reasoning, inexact and imprecise reasoning, meta-reasoning, etc. Suggested preparation: a course in
formal logic and a previous course in artificial intelligence.
CMPT 824-3 ?
Issues in Logic Programming
This course covers the computation model of logic programs, the theory of logic programs, the Prolog language
(both pure Prolog and real-life Prolog, with such features as meta-logical predicates, cuts and negation, extra-
logical predicates, and pragmatic issues); advanced Prolog programming techniques, such as nondeterministic pro-
gramming, incomplete data structures, logic grammars, and meta- interpreters; and applications, such as game-
playing, equation solving, compilers.
CMPT 825-3 ?
Natural Language Processing
In this course, theoretical and applied issues related to the development of natural language processing systems are
examined. Investigations into parsing issues, different computational linguistic formalisms, natural language
semantics, and discourse related phenomena will be considered and an actual natural language processor will be
developed.
CMPT 826-3 ?
Automated Learning and Reasoning
This course covers topics which are shared both by Al and Cognitive Science. Current Al research papers are
examined from the perspective of Cognitive Science, and vice-versa. Topics covered in a given semester will vary,
depending on the instructor, but most of the following topics will be addressed in any given semester: Connection-
ist models of intelligence; "human-like" automated deduction; reasoning by analogy-topics in natural language;
automated concept learning; and computational approaches to semantics.
CMPT 827-3 ?
Expert Systems
This course will analyze the Artificial Intelligence theory and practice underlying Expert Systems and survey a
number of the pioneering Expert System applications. Topics will include reasoning engines, the rule-based
approach, search, model-based representations, constraint propagation, reasoning maintenance, uncertainty,
knowledge acquisition, plus practical issues in Expert System development.
CMPT 830-3 ?
Compiler Theory
Precedence, LL(k), LR(k) grammars; SLR(k), LALR(k), L(m)R(k) and LR(k) parsing techniques, transduction
grammars; general compiler organization, code generation and optimization; memory allocation for object pro-
grams; garbage collection.
CMPT 831-3 ?
Functional Programming
This course will cover functional programming including introduction to a functional programming language, pro-
gram transformation and verification, implementation of functional programming languages, and other selected
•
?
topics which may include parallel evaluation of functional programs, analysis of performance, and advanced appli-
cations.
CMPT 841-3 ?
Query Processing in Database Systems
6
Algorithms for data-intensive operations for disk-based, main-memory-based, loosely distributed and tightly cou-
pled databases; analytical and empirical performance studies of database systems.
CMPT 842-3 ?
Concurrency Control in Database Systems
Transactions, recoverability, serializability theory, schedulers, locking, timestamping, optimistic schedulers, mul-
tiversion database systems; Recovery, commit protocols, termination protocols; Replicated database systems,
quorum-based concurrency control; Distributed snapshot taking, distributed deadlock detection, reliable storage
systems; Concurrency control in object-oriented database systems.
CMPT 843-3 ?
Principles
of Database
and Knowledge-Base Systems
An advanced course on database systems which covers the following topics: semantic data modelling, engineering
databases and spatial databases, object-oriented data models and systems, deductive database systems, semantic
query optimization, learning and induction in database and knowledge-base systems, and architectures of data-
intensive knowledge-base systems.
CMPT 851-3 ?
Fault-Tolerant Computing and
Testing
This course will cover concurrent error detection, self-checking networks, design for testability, and built-in self-
test. Existing fault-tolerant systems will be studied.
CMPT
852-3 ?
VLSI Systems Design
This course links two fields that traditionally have been considered two separate entities: computer architecture
and integrated circuit design. The vehicle used to demonstrate the interaction of layout issues and architectural
concepts is metal oxide semiconductor technology.
CMPT 853-3 Computer-Aided Design
/ Design Automation for Digital Systems (formerly CMPT
863)
Algorithms for logic synthesis and physical CAD/DA. Emphasis on routing, placement, partitioning, and gate-
level logic synthesis.
CMPT 891-3 ?
Advanced Seminar
CMPT 894-3
?
Directed Reading
CMPT 898
?
M.Sc. Thesis
CMPT 899,
?
Ph.D. Thesis
Special Topics Courses
In any semester a limited number of Special Topics courses may be offered subject to student demand and
faculty availability. Details of any Special Topics courses will be posted several months before they are offered.
CMPT 881-3
Special Topics in Theoretical Computing Science
CMPT
882-3
Special Topics in Artificial Intelligence
CMPT
883-3
Special Topics in Programming Languages
CMPT 884-3
Special Topics in Database Systems
CMPT 885-3
Special Topics in Computer Architecture
CMPT
886-3
Special Topics in Operating Systems
CMPT 887-3
Special Topics
in Hardware Design
IN
Part H.
?
Explanatory Notes.
1.
Admission Requirements
The requirements for admission to the Ph.D. program before completion of the M.Sc. program have been
modified to allow students to change programs with little lost time. Also, see the discussion in section 4 below.
2.
Research Topics Seminar Series
In this seminar series, each faculty member will give one or two lectures on topics of his/her choice. The
seminars will be given in the Fall and Spring semesters each year. All graduate students will be required to attend
the seminars presented during the first year of enrolment. The seminar series serves several purposes. Most impor-
tantly, it provides new students with an overview of faculty research interests. It should make it easier for students
to select thesis topics and could shorten the average completion time of M.Sc. students by more than 4 months. It
will also get Ph.D. students thinking about research as early as possible. A second benefit is that it provides a sem-
inar series in the Fall when external speakers are scarce. Attendance at the seminars will be open to all members of
the school. This would give faculty members a good opportunity to learn about the research of their colleagues.
3.
700-Level Graduate Courses
.
?
?
Currently, CMPT 810 is the only graduate course that is explicitly intended to cover a broad area at the M.Sc.
level. Some of the other regularly offered graduate courses and a few special topics courses are also fairly broad
courses accessible to M.Sc. students. The rest of our graduate offerings are advanced research courses. Currently,
we offer close to 20 graduate courses each year. In Part I of this document, 6 courses are identified as stressing a
broad coverage of a research area. They are given 700 numbers to distinguish them from the 800-level research
topics courses.
The general philosophy for these courses is that they should be accessible to the majority of students entering
the M.Sc. program. This means that they should not assume too much specialized previous knowledge. It should
be possible for students from widely differing backgrounds to obtain reasonably broad knowledge of a research
area without being penalized for moderate deficiencies in their backgrounds. The 700-level courses are intended to
fit in with some of the following proposals concerning breadth requirements. Therefore, they will have fairly stand-
ardized curricula and will have a final exam as a major component.
Three of the six proposed 700-level courses, CMPT 730, CMPT 740, and CMPT 760 are "new" courses.
CMPT 810, CMPT 820, and CMPT 850 have been renumbered CMPT 710, CMPT 720, and CMPT 750, respec-
tively.
T
he titles and calendar descriptions of these three courses have been updated, but the content remains the
same as previous offerings. Since these are not new courses, no new course forms have been included. CMPT 760
has already been offered three times as CMPT 886, so it is actually an existing course. However, since we want to
retain the 886 Special Topics number, it is necessary to propose CMPT 760 as a new course and a new course form
is included in the Appendix. Much of the material for CMPT 740 is taken from CMPT 841 which has been offered
• ?
several times previously. Advanced material has been added to CMPT 841 to replace the material that has been
moved to CMPT 740. Although neither course is completely new, new course proposals are included in the Appen-
dix. CMPT 730 is a new course with nothing similar elsewhere in our graduate or undergraduate offerings.
8
4.
Course Work for the M.Sc. Program
?
0
The course requirements for the M.Sc. program remain unchanged at 6 courses, with at least
5
from the
School of Computing Science. The additional requirement that 3 of these courses be 700-level courses ensures
moderate breadth while still allowing considerable room for specialization. Since most students will probably
choose one of their 700-level courses to be in their intended research area, the effect is to require breadth in two
areas and depth in one area (the research area). The requirement of 700-level courses also permits a smooth
transfer of students into the Ph.D. program as discussed below. The current requirement that all M.Sc. students
must take CMPT 810 has been removed (but a similar requirement has been added to the Ph.D. program).
5.
Ph.D. Breadth Requirement
The current breadth requirement for Ph.D. candidates is to pass a set of five Comprehensive Examinations at
the end of the first year of enrolment. There is no course requirement. The examinations are at the senior under-
graduate level and have served well as a filter, but they have also caused some problems. Currently, Ph.D. candi-
dates are spending their entire first year attending undergraduate courses and studying for the exams. This is obvi-
ously undesirable. A second problem is the psychological effects on the students. Even though the pass rate is over
85%, writing five 3-hour exams in a two week period is a very stressful experience.
The proposed Ph.D. breadth requirement is a combination of 700-level courses and Comprehensive Examina-
tions. A Ph.D. candidate is now required to demonstrate breadth at the M.Sc. level in five of six areas. (One of the
previous areas has been split into two.) Theoretical Computing Science is a compulsory area. This is consistent
with the current requirement of CMPT 810 in the M.Sc. program. Breadth can be demonstrated by taking a 700-
level course or by writing a Comprehensive Exam. At least two Exams must be written, but the choice is the
student's. It is expected that most students will choose to write the exams in their areas of strength and take
courses to fill in background in weaker areas.
This approach to ensuring breadth is more constructive than the previous set of Comprehensive Examina-
tions. Students can now correct background deficiencies by taking M.Sc.-level courses instead of undergraduate
courses, and the exams in the 700-level courses test moderate breadth at the M.Sc. level, instead of the undergradu-
ate level. It is also reasonable to set the Comprehensive Exams at the M.Sc. level because students will be writing
the exams in their areas of strength. Giving choice to the students should eliminate much of the fear about
Comprehensive Exams without eliminating the filter that faculty members consider to be necessary.
A further advantage of the proposed system is. that it provides a good mechanism for determining in which
program a student should be. It is often difficult to. determine, on the basis of GRE scores, transcripts, and letters of
reference, whether a student from an unfamiliar school has the background to attempt a Ph.D. More than two-
thirds of our 500 applicants each year are in this category. The proposed system permits a deferral of the decision,
when necessary, until we have enough information to make this decision. Since the structures of the M.Sc. and
Ph.D. programs are similar for the first two semesters, a student can move from one graduate program to the other
with little loss of time or effort.
Another advantage of the course requirement is that Ph.D. students will acquire GPA's. This is important
because most scholarships and fellowships are awarded on the basis of GPA's.
/0
9
6.
Course Work for the Ph.D. Program
A requirement of 3 graduate level Computing Science courses has been added to the Ph.D. program. This
requirement is redundant for most students since most students will be taking three courses to satisfy the Breadth
Requirement. The purpose of the course requirement is to ensure that the current requirement of five Comprehen-
sive Examinations is not weakened by the introduction of 700-level courses. For example, a student could satisfy
his/her M.Sc. course requirement by taking only 700-level courses and these courses could then be used to satisfy
the Ph.D. Breadth Requirement. Without the Ph.D. course requirement, this student would only have to write two
examinations during the first year of enrolment. In contrast, without the Ph.D. course requirement, an M.Sc. stu-
dent who took 700-level courses in his/her areas of interest and then specialized by taking 800-level courses would
be penalized in the Ph.D. program because he/she would be forced to either write Comprehensive Exams in his/her
weaker areas or take two or three more courses during the first year of enrolment.
7.
Completion Time
The effect of the changes above should be a decrease in the average length of both the M.Sc. and Ph.D. pro-
grams by as much as six months. Currently, the average completion time for the M.Sc. program is 7.5 semesters.
A typical M.Sc. student takes courses for 3 semesters and then starts to look for a supervisor and a research topic.
This often takes at least one more semester, so that the student is just starting on thesis research in the fifth semes-
ter. The average M.Sc. student then spends approximately a year finishing a thesis. Under the proposed system, an
. average M.Sc. student will have completed
5
courses and will have a good idea of possible thesis topics after two
semesters. It is then quite reasonable to expect the student to have completed all courses and be doing research by
the end of the third semester. This gives a projected average completion time of about
5.5
semesters.
Ph.D. students are now spending their entire first year taking undergraduate courses in preparation for the
Comprehensive Examinations. It is usually not until part way through the fourth semester that they start to think
about research. Under the proposed system, most students will finish all breadth requirements in two semesters.
8.
Ph.D. Research Area Examination and Research Proposal
The Research Area Examination is the same as what is now called the Oral Candidacy Examination. It is an
oral examination intended to test depth in a student's research area. The Research Proposal is a written document
that is more specific to the student's research topic. It is also a current requirement.
9 ?
Logistics
The calendar descriptions in Part I are set up to give the Graduate Program Committee and the Ph.D. Breadth
Committee enough flexibility to deal with unusual cases. Two possible administrative problems and their solutions
are the following.
(a) Sometimes it may not be appropriate for a student to take some of the 700-level courses because the courses
are similar to courses already taken (either at S.F.U. or elsewhere). An M.Sc. student can simply be asked to
. ?
satisfy his/her course requirement with other 700-level courses; there's nothing wrong with more breadth.
Ph.D. candidates in this situation could be asked to write the final exam in a 700-level course without requir-
ing the term work. They also have the option of writing the Comprehensive Examination in any area.
II
10
(b) Occasionally, a student who has previously failed to satisfy the Ph.D. Breadth Requirement may be admitted
to the Ph.D. Program. This situation could arise when a promising student is admitted to the Ph.D. program
but then moved into the M.Sc. program to correct background deficiencies that were not spotted during the
admission process. This situation is clearly an anomaly that would require individual consideration. Typi-
cally, such a student would be admitted to the Ph.D. Program with conditions attached.
10.
Supervisory and Examining Committees
The compositions of the M.Sc. and Ph.D. Supervisory and Examining Committees have been changed.
Currently, an M.Sc. Examining Committee must include a member external to the School of Computing Science
and a Ph.D. Examining Committee must include a member from a department or discipline other than the one in
which the candidate is working. These restrictions have proved to be awkward in practice and they do not seem to
be serving any useful purpose. The requirement that a Ph.D. Supervisory Committee consists of at least three
faculty members has been removed for the same reason. The proposed compositions of these committees are
nearly identical to those specified in Sections 1.6 and 1.9 of the
Graduate General Regulations.
The only differ-
ence is that at least one member of an M.Sc. or Ph.D. Supervisory Committee, besides the Senior Supervisory must
be from the School of Computing Science.
11.
House-cleaning
The following courses are being deleted for a variety of reasons such as insufficient enrolments and changes
in the faculty.
CMPT 840-3
Advanced Topics
in Simulation
and Modelling
CMPT 861-3
Biomedical
Computing
CMPT
862-3
Computer
Mapping
CMPT
892-3
Advanced
Seminar II
CMPT
893-3
Advanced
Seminar III
CMPT
895-5
Directed
Reading
II
The following courses have been renumbered
CMPT 710-3 ?
Computational Complexity
(formerly CMPT 810)
CMPT 720-3 ?
Artificial
Intelligence
(formerly CMPT 820)
CMPT
750-3 ?
Computer Architecture
(formerly CMPT 850)
CMPT 815-3 ?
Algorithms of Optimization (formerly CMPT 860)
CMPT 853-3
?
Computer-Aided Design / Design Automation for Digital Systems
(formerly CMPT 863)
The titles and/or calendar abstracts of the following courses have been modified to be more accurate and
descriptive. The numbers of some of the courses have also been changed as noted separately above. There is no
change in the content or other details of these courses.
CMPT 710-3 ?
Computational Complexity (new title, abstract, and number)
CMPT 720-3 ?
Artificial
Intelligence
(new title, abstract, and number)
CMPT 750-3
?
Computer Architecture (new title, abstract, and number)
19
11
.
CMPT 811-3
CMPT 812-3
CMPT 813-3
CMPT 815-3
CMPT 823-3
CMPT 824-3
CMPT 842-3
CMPT 851-3
CMPT 853-3
Distributed Algorithms
(new abstract)
Parallel Computation
(new abstract)
Computational Geometry
(new abstract)
Algorithms of Optimization
(new
number)
Formal Topics in Knowledge Representation
(new abstract)
Issues in Logic Programming
(new title and abstract)
Concurrency Control in Database Systems
(new title and abstract)
Fault-Tolerant Computing and Testing
(new title and abstract)
Computer-Aided Design
/
Design Automation for Digital Systems
(new title and abstract)
12. New Courses
The following new courses are proposed. New Course Proposal forms are included in the Appendix. As pre-
viously discussed, three new 700-level courses (730, 740, 760) are being added to provide a full complement of
M.Sc.-level courses. Two courses (821, 841) are existing courses that have evolved to the point that we think that a
new course proposal should be included for information purposes. Three courses (826, 831, 843) have been offered
previously as special topics courses and are proposed for inclusion because they have been popular. The remaining
three courses (814, 825, 827) have not been previously offered, but substantial parts of these courses have been
taught as sections of other courses.
.
CMPT
730-3
CMPT 740
.
3-
CMPT 760-3
CMPT 821-3
CMPT 841-3
CMPT 826-3
CMPT 831-3
CMPT 843.3
CMPT 814-3
CMPT 825-3
•
?
CMPT 827-3
Foundations of Programming Languages
Database Systems
Operating Systems
Robot Vision
Query Processing in Database Systems
Automated Learning and Reasoning
Functional Programming
Principles of Database and Knowledge-Base Systems
Algorithmic Graph Theory
Natural Language Processing
Expert Systems
The addition of these courses does not imply an increase in the number of graduate courses being offered
each year. They are being added to reflect the current research interests of the faculty more accurately than the pre-
vious courses. During the semesters 86-1 through 89-3,66 graduate courses were offered by the School of Com-
puting Science. Of these 66 courses, only 42 have been courses with descriptive titles. The remaining 24 courses
(more than
35%)
have been CMPT 88X Special Topics courses. The 88X series of courses was intended to be used
for unusual courses that will only be offered once.
.
12
Part LII. Appendix - New Course Proposals.
CMPT 730-3
Foundations of Programming Languages
CMPT 740-3
Database Systems
CMPT 760-3
Operating Systems
CMPT 814-3
Algorithmic Graph Theory
CMPT 821-3
Robot Vision
CMPT 825-3
Natural Language Processing
CMPT 826-3
Automated Learning and Reasoning
CMPT 827-3
Expert Systems
CMPT 831-3
Functional Programming
CMPT 841-3
Query Processing in Database Systems
CMPT 843-3
Principles of Database and Knowledge-Base Systems
.
S
/,,
SiMON FRASER UNIVERSITY
?
MEMORANDUM
0
To ?
Joseph G. Peters, ?
From...Sharon Thomas,
6hbóOicieiic
.
?
?
a' a
?
c ions Management Office
Library
Subject... ?
PROPOSALS
I
Date...
?
. .
?
.
I have examined the following new course proposals:
MPT 730 Foundations of Programming Languages
CMPT 740 Database Systems
CMPT 760 Operating Systems
CMPT 814' Algorithmic Graph Theory
cMPT 821 Robot Vision
MPT 825 Natural Language Processing
CNPT 826 Automated Learning and Reasoning
C1PT 827 Expert Systems
CMPT 831 Functional Programming
CT 841 Query Processing in Database Systems
GMPT 843 Principles of Database and Knowledge-Base Systems
Several of these courses have already been taught as special
topics offerings and, in any case, all of them fall well within the
Library's normal collecting scope.
Our journal collection, although relatively modest, appears
to be adequate for immediate purposes. In fact, the one journal
requested for CMPT 843, IEEE Transactions on Knowledge and Data
Engineering, is already arriving in the Library as part of our
comprehensive subscription to IEEE publications.
Nevertheless it is almost inevitable that expansion of the
graduate programme into these areas will generate future demands
for a more comprehensive collection than we now possess and the
Library may well be faced with demands that it cannot meet.
ST: is
0
SIMONr FRASER UNIVERSITY
NEW GRADUATE COURSE PROPOSAL
CALENDAR INFORMATION:
?
.
Department:
Computing Science ?
Course Number: 730
Title:
Foundations of Programming Languages
Description:
This course covers aspects of programming language syntax and semantics.
Formal methods for defining languages and proving programs correct are considered.
Credit Hours:_
?
Vector: _3-00
?
Prerequisite(s) if any:
ENROLLMENT AND SCHEDULING:
Estimated Enrollment: ?
15
?
When will the course first be offered: 90-1
How often will the course be offered:-
Annually
JUSTIFICATION:
ThiscoursewillcovermaterialthateveryComputingSciencePh.D.studentshouldknow.
The course will also be of value to M.Sc. students. Similar
-
courses are taught at othe
universities.Ourstudentshavepreviouslycoveredmuchofthismaterialthrough
independent study, while preparing for the Ph.D. Comprehensive examinations.
RESOURCES:
Which Faculty member will normally teach the course:
F. Warren Burton or Robert Cameron
What are the budgetary implications of mounting the course:
_None
Are there sufficient Library resources (append details):
?
Yes
Appended: a) Outline of the Course
b)
An indication of the competence of the Faculty member to give the course.
c)
Library resources
Approved:
Departmental Graduate Studies Commi
?
Faculty Graduate Studies Committee:___________
?
10/10
/V1
Faculty:
Senate Graduate Studies Committee:___________
Senate: ?
Date:_____________
Ii.
Computing Science
Course Outline
CMPT 730: Foundations of Programming Languages
This course will cover basic concepts in the area of programming
languages, at the graduate level. The course will be largely of a
theoretical nature and will concentrate on fundamental concepts of
lasting importance, rather than topics of current interest. The course
is intended to cover material that every Ph.D. student in Computing
Science should know.
Outline:
weeks)
an
1.
introduction
Programming
to
language
attribute
syntax,
grammers
including
and their
a review
use.
?
of
(Approximately
BNF notation and
2
?
statement2.
(Approximately
Axiomatic
?
sequences,2
semantics,
weeks)
?
conditionals
including
?
axioms
and
for
?
assignment
looping ?
statements,
constructs. ?
. 3. ?
The lambda calculus,
including
lambda notation, reduction rules,
equality, representation of numbers and lists,
?
conditionals
?
and
recursion. If time permits, combinators may be covered. (Approximately
3 weeks)
functions,
4.
Introduction
environments
to denotational
and detailed
semantics,
examples.
including
(Approximately
domains,
4
semantic
weeks)
5. ?
Type
?
theory, ?
including
(Approximately 2 weeks)
U
Grading Scheme:
Midterm Examination: 40%
Final
Examination:
60%
Polymorphism and type hierarchies.
Required Texts:
Initially, the following two texts will be used in the course.
1.
Pagan, Frank G., Formal Specification of Programming Languages,
Prentice-Hall, 1981.
2.
Gordon, Michael J. C., Programming Language Theory and its
Implementation, Prentice-Hall, 1988.
17
ci
Computing Science
Library Requirements
CMPT 730: Foundations of Programming Languages
This
However,
journals.
course
students
will be
.
may
taught
occasionally
from the required
refer to
texts
papers
for
in
the
the
most
following
part..
1.
ACM Transactions on Programming Languages and Systems.
2.
IEEE Transactions on Software Engineering.
3.
ACM Computing Surveys.
4.
IEEE Computer.
5.
Computer Languages.
The library current subscribes to all of these journals.
S
.
/0
Computing Science
Faculty Qualifications
CMPT 730: Foundations of Programming Languages
Both Dr. Burton and Dr. Cameron are actively doing research in the area
of programming languages. Both have published a number of papers on a
variety of topics in the area of programming languages, including
several in leading journals specializing in programming languages (e.g.
ACM Translations on Programming Languages and Systems) . Both have
previously taught other courses in the area of programming languages,
including both undergaduate courses on programming languages generally
and graduate courses on special topics in programming languages.
.
I',
SIMON
FRASER, UNIVERSITY
NEW GRADUATE COURSE PROPOSAL
CALENDAR INFORMATION:
Department: ?
Computing. Science
?
Course. Number: T40
TitlePatase Systems
Description: Review of introductory database concepts; Query optimization; Concurrency
control; Reliability
of
crash recovery; Distributed databases; Object-oriented databases,
knowledge base management systems.
Credit Hours: ?
3 ?
Vector: 3-0-0
?
Prerequisite(s)
if
any:
ENROLLMENT AND SCHEDULING:
Estimated Enrollment: ?
15
When will the course first be offered- 90-3
How often will the course be
offered .
Annually
JUSTIFICATION:
Thiscourse is introducedasone
of
the six
foundationcourses
in
thegraduate
curriculum of this School.
RESOURCES:
Which Faculty member will normally teach the course: Wo-Shun
Luk,
Jia-WelHan
What are the budgetary implications of mounting the course: None
Are there sufficient Library resources (append details):
?
Yes
Appended: a) Outline of the Course
b)
An indication of the competence of the Faculty member to give the course.
c)
Library resources
Approved:
Departmental Graduate Studies Committee:
Faculty Graduate Studies Committee:_______________________________
Date: (O/(O78
________ _
Faculty:
Senate Graduate Studies
Date: iI,//0/719
•
Senate:
?
Date:
.,
Course Number: CMPT 740
Title: Database Systems
Description: Review of introductory database concepts; Query optimization; concurrency
control; Reliability and crash recovery; Distributed databases; Object-oriented databases,
knowledge base management systems.
Justification:
curriculum of this
This
School.
course is introduced as one of the
six
foundation
courses in the graduate
Course Outline:
1.
Review of the following introductory topics:
• Why Database?
• En
tit
y-Relationship
Model
• ?
• Relational, Network and Hierarchical Models
• Query Languages (SQL as example)
• Relational Databases
• Physical database structures (files, indices, hashing etc.)
2.
Relational Query Processing & Optimization
3.
Concurrency Control and Database Recovery
4.
Distributed Databases
5.
Object-oriented Database Systems
6.
Knowledge-base Management Systems
Expertise of the Instructor: Dr. Luk has had over ten years of research experience in
databases and has published widely in database journals and conference proceedings. He
has also served as a referee for numerous database journals and research grant agencies.
Library resources: The journals that are required for this course are already being
subscribed to by our library (e.g. ACM Transactions of Database Systems, Information
Systems, Journals for Distributed and Parallel Processing and IEEE Transactions on
Software Engineering).
School of Computing Science
?
Course Proposal ?
September 26, 1989
Graduate Course Proposal (Appendix)
Competence of Faculty to Teach CMPT-843
Jiawei Han - Ph.D. in Computer Sciences, has taught the course as a special topics
course for two years and publishes research in the areas covered by the course.
Wo-shun Luk - Ph.D. in Computer Science, has taught the material related to the course
topics and publishes research in the areas covered by the course.
Library Resources
During the terms in which this curse was taught as a special topics course, the
library acquired textbooks and relevant supplemental information at the instructor's
request. In addition, current research papers and class notes are placed on reserve.
.
0
.
SIMON
FRASER UNIVERSITY
NEW GRADUATE COURSE PROPOSAL
CALENDAR INFORMATION:
Department- ?
Computing Science
?
Course Number:
760
Title:
Operating Systems
Description:
This course will cover the pragmatic aspects of modern operating systems
design and implementation.
Credit Hours:
?
_Vector:
_3-0
?
Prerequisite(s)
if
any:
ENROLLME
q
j' AND SCHEDULING:
Estimated Enrollment:
?
15
?
When will the course first be offered:
90-1
How often will the course be offered:
Annually
JUSTIFICATION:
This course Is being Introduced as one of the foundation courses in the graduate
curriculum.
RESOURCES:
Which Faculty member will normally teach the course:
?
Stella Atkins
What are the budgetary implications of mounting the course:
?
None
Are there sufficient Library resources (append details):
?
Ye
Appended: a) Outline of the Course
b)
An indication of the competence of the Faculty member to give the course.
c)
Library resources
Approved:
Departmental Graduate Studies
Faculty Graduate Studies
Date:_
/ i7/O/;f
•_Date:lJ_
C)_.I-/
Senate:
_Date:_____________
Faculty:_
Senate Graduate Studies
CMPT 760
?
GRADUATE COURSE ON OPERATING SYSTEMS DESIGN.
?
Instructor: Dr. M. Stella Atkins.
This course will cover the pragmatic aspects of modem operating systems design and implements-
lion. No single reference textbook is available to cover this rapidly changing area, so most material will be
taken from recent journal and conference papers, which
will
be handed out on each topic.
dents.
The course will include topics chosen from the following areas, depending on the interests of the stu-
(1)
Review: single machine implementation of UNIX: file system, etc.
(2)
The SUN NFS distributed file-sharing system.
(3)
Other Distributed operating systems: ANDREW, AMOEBA, V-system and SPRITE.
(4)
Distributed data structures -- Linda.
(5)
Concurrent Languages for distributed systems - Synchronising Resources (SR).
(6)
Broadcast and multicast primitives and applications.
(7)
Multiprocessor systems: Tr nsputers, COSMIC CUBE, HYPERCUBE, DEC's FIREFLY.
(8)
Load sharing and performance measurement.
(9)
Real-time systems.
(10)
Discussion of the newly-released NeXT computer system -- its operating system (based on
Carnegie-Mellon University's MACH operating system), its networking capabilities (based on
SUN's Network File System), its virtual memory management and its window display mechanisms.
Grading and Assignments
Students will be expected to complete an assignment, such as measuring some aspect of the system,
using either our distributed system of SUN-2 workstations running the message-based V-system, or using
the SUNs connected to the department's research Ethernet. Students will also be expected to complete a
project on a chosen practical subject, probably using either the language SR on our research Ethernet, or
the V-system on our Distributed Computing Laboratory Ethernet, or the Next computers, or, if available,
the new Transputers. In addition, each student will be expected to make a 3040 minute presentation on a
recent research paper. There will also be an examination contributing 40% of the grade.
0
Competence of Faculty Member to teach CMPT 760 -- Operating Systems
Dr. Stella Atkins has studied Operating Systems since 1979, when she started her PhD thesis on the
role of exception mechanisms in software systems design, working under Dr. David Cheriton at UBC. Dr.
Cheriton subsequently moved to Stanford University where he developed the well-known message-based
distributed operating system, known as the V-system. Since completing her PhD in 1985, Dr. Atkins has
completed many research projects in operating systems design, ranging from low-level modifications to the
V-kernel, to high-level concurrency control studies. More recently, since 1988, Dr. Atkins has become
involved in the SFU/Triumf positron emission tomography (PET) design group, and has research publica-
tions on real-time systems and applications.
Six Masters students have completed research theses under her supervision in operating systems dur-
ing 1987 to 1989, and two more are currently working on their theses.
Dr. Atkins has taught operating systems to undergraduates and graduates at Simon Fraser University
since
1985,
and continues to revise the curriculum to reflect recent hardware and software changes.
• ?
Library Resources for CMPT 760
No single reference textbook can keep up-to-date with this rapidly changing area, so most material is
taken from recent journal and conference papers, which will be handed out on each topic.
The ACM journals are heavily used, and copies are available in the Computer Science Library, as
well as the main library. The IEEE journals are available in the main library. IEEE Conference proceed..
ings are usually (but not always) available in the main library; Dr. Atkins subscribes to some of these for
department
more timely
has
access.
sufficient
Other
copies.
documents are taken from the computer manufacturer's manuals, of which the
One
book may be used as a reference:
Distributed Systems: Concepts and Design
by Coulouris and
Dollimore. Addison Wesley, publisher 1988.
.
.76
1,24
SIMON FRASER UNIVERSITY
NEW GRADUATE COURSE PROPOSAL
CALENDAR INFORMATION:
Department ?
Computing Science
?
Course
Number: 814
Title:
Algorithmic Graph Theory.
Description:
The course will examine graphs and networks with special structure which
allows the development of efficient algorithms for hard computational problems.
Credit Hours:
?
3
?
Vector: _3-0-0
?
Prerequisite(s) if any:
ENROLLMENT AND SCHEDULING:
Estimated Enrollment: ?
10 ?
When will the course first be offered:
?
90-3
?
How often will the course be offered: _
depending on demand.
JUSTIFICATION:
Our courses on algorithm design stress universal approaches for the most general problem
instances. If the instances posses a special structure, more efficient algorithms are
possible. Specially structured graphs (chordal, interval, permutation) admit such
algorithmandour students should see how to design them.
RESOURCES:
Which Faculty member will normally teach the course:
?
Pavol Hell
What are the budgetary implications of mounting the course:
_None
Are there sufficient Library resources (append details):
?
Yes
Appended: a) Outline of the Course
b)
An indication of the competence of the Faculty member to give the course.
c)
Library resources
Approved:
Departmental Graduate Studies
Faculty Graduate Studies Committee:
Faculty: ?
(
1,1
1
Senate GraduateStudiesCommittee:_
Date:_
if/I
/ic_7 •
________
Senate: ?
S ?
Date:_____________
a) Proposed Course Outline - CMPT 814
?
ALGORITHMIC GRAPH THEORY
Text: M. Golumbic, Algorithmic Graph Theory and Perfect Graphs, Academic
Press, 1980.
Algorithms, Data Structures, Complexity.
• ?
Classical optimization problems.
• ?
Perfect graphs, chordal and strongly chordal graphs.
•
?
Algorithms based on linear programming duality.
•
?
Algorithms for interval graphs, p-q trees.
• ?
•
?
Permutation graphs and transitively orientable graphs.
•
?
Other algorithms using special structure.
b)
Competence
M.Sc.
P. Hell
in
is
the
active
same
in
area.
research in the area and has supervised one
.
Ph.D. and one
c) Library Resources
The library already has a copy of the text. There are a few papers as well,
but they are to be found in the standard journals.
0
SIMON
FRASER UNIVERSITY
NEW GRADUATE COURSE PROPOSAL
CALENDAR INFORMATION:
Department:
School of Computing Science
?
-Course Number:
CMPT 821
Title:
?
Robot Vision
This course discusses issues and research results pertinent to robot vision. Topics include depth
Description: recovery for robot navigation, three-dimensional object recognition and scene analysis, model-
based approach, parallel vision machines and algorithms, and case study of contemporary robot
vision systems.
Credit Hours: _3
?
Vector:_3-0-0
?
Prerequisite(s) if any:
ENROLLMENT AND SCHEDULING:
Estimated Enrollment:
?
10
?
When will the course first be offered:
?
90-3
How often will the course be offered:
_Once. every two years.
JUSTIFICATION:
Robot Vision is a young and rapidly growing field. ?
It is necessary
to have a
graduate course to concentrate on the fundamental issues and contemporary research
results in this area.
?
This is a modernization of the existing course CHPT 821,
"Pattern Recognition and Image Processing".
RESOURCES:
Which Faculty member will normally teach the course:
?
Dr. Ze-Nian Li
What are the budgetary implications of mounting the course:
?
None
Are there sufficient Library resources (append details): ?
Yes ?
(see attached)
Appended: a) Outline of the Course
b) An indication of the competence of the Faculty member to give the
course.
c) Library resources
Approved:
Departmental Graduate Studies Comnlittee.
?
5L
f'
L
cPk_
Date:
ZZ
/9 /
Faculty Graduate Studies Committee:
Date:
1
0
/1
0
A9
Faculty:
Date:_
__
Senate Graduate Studies Committee:
Date: ?i
Senate:
Date:_____________
.
School of Computing Science
CMPT 821 Robot Vision
Instructor: Dr. Z.N.
Li
This seminar course is intended as a survey of issues and research results pertinent to Robot
Vision. No textbook is required. A listing of reading materials from recent journals and
conference proceedings will be provided.
Tentative Outline
(1)
Depth Recovery for Robot Navigation
(2)
3-D Object Recognition and Scene Analysis
(3)
Model-based Robot Vision Systems
(4)
Parallel Vision Machines and Algorithms
(5)
Hierarchical Representations and Systems
(6)
Case Study
Recommended Preparation:
Knowledge of calculus, computer algorithms, and artificial
intelligence
Grading: Class presentation and participation 20%, two small programming assignments 20%,
term project 60%.
30
CMPT 821
?
0
Dr. Ze-Nian Li obtained his doctorate degree in the area of computer vision, and is
acth'ely engaged in research in robot vision. He has taught this graduate course in US. Dr. Li
has supervised one M.S. thesis and is presently supervising one Ph.D. and two'MS. theses in this
area.
MPT
' 821
Library
Requirements
The SFU library has all the following journals and
corifer
enceproceedingsmeeded for this course.
Related Journals:
(1)
Artificial Intelligence
(2)
Computer Vision, Graphics and Image Processing
(3)
IEEE Journal of Robotics and Automation
(4)
IEEE Transactions on Pattern Aiiálysis'and Machine 'Intelligence
(5)
IEEE Transactions on Systems Man, and Cybernetics
(6)
International Journal of Robotics Research
(7)
Pattern Recognition
Related Conference Proceedings:
(1)
Proceedings of the IEEE International Conference
,
onRàbotjcs and Automation
(2)
Proceedings of the IEEE Computer Society Conference on tom
Recognition
?
puter Vision and Pattern
(3)
Proceedings of the International ConferencdonpatternRecogjtion
LIJ
31
i1
Faculty:
Faculty Graduate Studies
10/loAl
i
is
SIMON FRASER UNIVERSITY
NEW GRADUATE COURSE PROPOSAL
CALENDAR INFORMATION:
Department:
Com p
uting Science ?
Course Number:
8
25
Title:-
Natural Language Processjg
Description:
In this course, theoretical and applied issues related to the development of
natural language processing systems are examined. Investigations into parsing issues,
different computational linguistic formalisms, natural language semantics, and discourse
Credit
related
Hours:
phenomena
?
will be considered
Vector:
and an
3-0-0actual
?
natural
Prerequisite(s)
language
if any:
processor
?
-
W -
developed.
131 be
ENROLLMENT AND SCHEDULING:
Estimated Enrollment:
?
10 ?
When will the course first be offered:
91-1
How often will the course be offered:
_Every second yeaf
JUSTIFICATION:
Seenextpage.
RESOURCES:
Which Faculty member will normally teach the course:
Dr.FredPopowich,Dr.NickCerconeor
What are the budgetary implications of mounting the course:
None ?
Dr. Veronica Dahl
Are there sufficient Library resources (append details):
Yes,, (see attached)
Appended: a) Outline of the Course
b)
An indication of the competence of the Faculty member to give the course.
c)
Library resources
Approved:
Departmental Graduate Studies
,
Senate Graduate Studies Committee:
?
_C1I_,___ -_Date:/ /
?
/4 '
Senate: ?
Date:_____________
3.
SchooF'of Computing Science
CMI'T 825 Natural Language Processing
?
.
Description
In this course, theoretical and applied issues related to the development of natural language processing
systems are examined. Investigations into parsing issues, different computational linguistic formalisms,
natural language semantics, and discourse related phenomena will be considered and an actual natural
language processor will be developed.
Justification
There is currently a great deal of research 'into natural language processing, particularly as applied to 'the
machine translation of natural language and to the development of natural language interfaces to
databases and expert systems. This course is iequired to familiarize Students with the large amount of
contemporary research on issues fundamental to natural language processing and to provide hands-on
experience with the problems associated with the actual development of natural language processors.
.
0
•
??
School of Computing Science
CMPT 825 Natural Language Processing
Instructors:
?
Dr. Fred Popowich
?
Dr. Nick Cercone
?
Dr. Veronica Dahl
In this course, theoretical and applied issues related to the development of natural language processing
systems are examined. Aside from surveying some traditional approaches to natural language
processing, recent research in the field will also be examined. A list of readings from recent journals
along with a list of reference texts will be provided. Students will develop a natural language processor
during the course.
9
?
TENTATIVE OUTLINE:
Linguistic Formalisms for Natural Language Processors:
A survey of specific linguistic formalisms which have formed the basis for natural language
processors. This survey will include formalisms selected from but not restricted to the
following set.
transformational grammar, lexical functional grammar, government and binding theory,
generalized phrase structure grammar, head-driven phrase structure grammar.
Parsing Natural Language:
Examination of different computational tools and parsing strategies for dealing with the
formalisms introduced in the first part of the course. Will also consider the use of non-
syntactic (e.g. semantic and discourse-related information) both during and after the parse.
GRADING:
Will be based on a term project, assignments, and an examination.
RECOMMENDED
Prolog.
PREPARATION:
Knowledge of artificial intelligence, parsing, and either LISP or
.
3S
CMPT 825 - Library Requirements
The SFU library has copies of all of the following books which will be required as references for the
course.
• Lectures on Contemporary Syntactic Theories by
Peter Sells.
• Information-based Syntax and Semantics
by Carl Pollard and Ivan Sag.
• Computational Linguistics
by Ralph Grishnian.
• Prolog and Natural Language Analysis
by Fernando Pereira and Stuart Shieber..
• Natural Language Understanding by
James Allen.
• Logic Grammars
by Harvey Abramson and Veronica Dahl.
The
Proceedings from the Annual Meetings
of
the Association for Computational Linguistics, the Proceedings
of
the International Conferences on Computational Linkuis tics,
along
with
the journal
Computational Linguistics
all contain articles which are relevant to the: course. Although these publications are not currently
available in the SRi library, Drs. Cercone, Dahl and Popowich have copies which can be put on reserve
in the library. Copies of other relevant papers will also be put on reserve.
.
3',
. ?
CMPT 825 - Instructors
Dr. Fred Popowich obtained both his M.Sc. (Computing Science) and his Ph.D. (Artificial Intelligence
I
Cognitive Science) in the area of computational linguistics. He has published papers describing different
natural language processors that he has developed. Currently, he is actively involved in several natural
language processing projects. These projects are supported by a fellowship by the Advanced Systems
Institute of British Columbia and by the Natural Sciences and Engineering Research Council of Canada.
Dr. Veronica Dahi obtained a Doctorat de Specialite en Intelligence
Artificiele at the Aix-Marseille II University. She is internationally
known for having pioneered the introduction of logic programming
into the fields of deductive databases, expert systems and natural
language processing. She has also done extensive research on
building logic programming tools for processing language, and on
producing useful syntheses between linguistic theory and logic
grammars.
Competence of Faculty - CMPT 825
Dr. N. Cercone obtained his M.Sc. (Computer & Information Science) and his
Ph.D. (Computing Science) in artificial intelligence specializing in computational
linguistics and practical Al systems. He is an active researcher in computational
linguistics and expert systems and has taught material of this nature for 13
years. He is currently funded by NSERC with operating, intrastructure, strategic,
and equipment grants. He is also editor of Computational Intelligence.
37
SIMON
FRASER UNIVERSITY
NEW GRADUATE COURSE PROPOSAL
CALENDAR INFORMATION:
Department:
?
Computing Science
?
Course Number: 826
Title ?
Automated Learning and Reasoning
Description: ?
See attached page
Credit Hours:
_3
?
Vector:_
3-0-0
?
Prerequisite(s) if any:
At least one graduate or undergraduate Al course, or instructor's permission.
ENROLLMENT AND SCHEDULING:
Estimated Enrollment:
10 ?
When will the course first be offered:
?
90-2
?
How often will the course be offered: _(
ince every two years.
JUSTIFICATION:
This course has been offered 2 times in the last 2 years, as,a special topics course.
It covers current research not addressed in our other courses.. It had an average
enrollment of 9 students, indicating-there is demand for the course.
RESOURCES:
Which Faculty member will normally teach the
C ourse:
_Dr. R. Hadley, Dr. F. 'Popowich
What are the budgetary implications of mounting the course
Are there sufficient Library resources (append details):
Yes
Appended: a) Outline of the course
b)
An indication of the competence of the Faculty member to give the course.
c)
Library resources
Approved:
Departmental Graduate Studies
Faculty Graduate Studies Committee:
?
c 7(0
Senate
Faculty:
Graduate
?
Studies Committee:—
?
'-,_.
?
'. ?
' Date:
Date:_)
31
( /'
t',/.7
•
Senate: ?
•'
?
Date:_____________
Course description for CMPT 826
This course covers topics which are shared both by Al and Cognitive Science. Current Al research
papers are examined from t
h
e perspective of Cognitive Science, and vice-versa. Topics covered in a
given semester will vary, depending upon the instructor, but
most
of the following topics will be
addressed in any given semester: Connectiomst models of intelligence; "human-like" automated
deduction; reasoning by analogy; topics in natural language; automated concept learning; and
computational approaches to semantics.
:IJ
.
39
CMPT - 826?
AUTOMATED LEARNING AND REASONING
In this course we concentrate upon a number of topics which fall into
the "intersection area" between Al and Cognitive Science. We will be
reading papers on the topics listed below. Not all topics will receive
equal attention. Depending upon student interest, we may focus more on
some topics than others.
Topics
- The connectionjat (Or neural net) model of cognition - massive
parallel processing, and Boltzman machines for concept
learning.
?
The prospects for using neural nets for higher
level reasoning.
- The Frame Problem - its relation to Default Reasoning.
- Automated Reasoning - resolution methods vs. natural deduction
methods (including semantic tableaus).
- Cognitive approaches towards automated theorem proving,
discovering heuristics automatically, and simultaneous forward
and backward chaining.
- Reasoning by analogy - adapting plans by analogy.
- Computational theory of human cognition - Pylyshyn.
- Semantics, and Truth. Can these concepts be explained in
computational terms? Implications for the computational model
of the mind.
- Al approaches to Concept learning and language acquisition -
Michalski
- Philosophical aspects of the computational theory of
cognition, Searle, relevance of,Godel's theorem.
I
Competence of Faculty to teach CMPT 826
R. F. Hadley - Ph.D. has taught the course as a special topics course for 2 years.
Publishes research in the areas covered by the course.
F. Popowich - Ph.D. is in Cognitive Science. Teaches and publishes in Artificial
Intelligence, especially Computational Linguistics.
Library Resources
During the terms in which this course was taught as a special topics course,
the library acquired relevant supplemental texts at the instructor's request. In
addition, current research papers are placed on reserve.
S
S
Senate Graduate Studies
Faculty:
Faculty Graduate Studies
(c,J(o
/8
'''•
SIMON
FRASER UNIVERSITY
?
NEW GRADUATE COURSE PROPOSAL
CALENDAR INFORMATION:
Department ?
Computing Science
?
Course Number.
Title:-
Tit1e ?
Expert Systems
Description:
See attached page.
Credit Hours:
?
3 ?
Vector:
3-0-0 ?
Prerequisite(s) if any:
ENROLLMENT AND SCHEDULING:
Estimated Enrollment:
10 ?
—When will the course first be offered:_______________________
How often will the course be offered:
Onceevery two years.
JUSTIFICATION:
See attached page.
RESOURCES:
Which Faculty member will normally teach the course:
_Dr. Bill Havens, Dr. Nick Cer cone
What are the budgetary implications of mounting the course:_________________________________
Are there sufficient Library resources (append details):
?
Yes(seeattached)
Appended: a) Outline of the Course
b)
An indication of the competence of the Faculty member to give the course.
c)
Library resources
Approved:
Departmental Graduate Studies
Senate: ?
Date:_____________
'I-,
SCHOOL OF COMPUTING SCIENCE
?
CMP'I' 827 Expert Systems
Description
This course will analyse the Artificial Intelligence theory and practice
underlying Expert Systems and survey a number of the pioneering Expert
System applications. Topics will include reasoning engines, the rule-based
approach, search, model-based representations, constraint propagation,
reasoning maintenance, uncertainty, knowledge acquisition, plus practical
issues in Expert System development. The course assumes a reasonable
background in A.I. fundamentals and programming. The course will make
use of the facilities in, the new Expert Systems Laboratory within the Centre
for Systems Science.
Justification
Expert Systems are knowledge-based computer programs which emulate
the reasoning abilities of human experts in some limited field of expertise.
Expert System technology is rapidly advancing and being successfully applied
in a wide variety of real applications. Although within the field of Artificial
Intelligence, Expert System methodology has matured into an important
discipline in its own right. This course will focus on presenting a coherent'
survey of Expert System methodology, including both theoretical issues in
automated reasoning and practical issues in knowledge engineering,
knowledge acquisition and human interfaces.
0
q3
SCHOOL OF COMPUTING SCIENCE
?
CMPT 827 Expert Systems
Instructor:
Bill Havens
Syllabus
Expert Systems are knowledge-based computer programs which emulate
the reasoning abilities of human experts in some limited field of expertise.
Expert System technology is rapidly advancing and being successfully applied
in a wide variety of real applications. This course will analyse the underlying
Artificial Intelligence methodology and survey a number of the pioneering
Expert Systems. The course is intended for graduate students with a
reasonable background in A.I. fundamentals and programming. The course
will make use of the facilities in the new Expert Systems Laboratory within
the Centre for Systems Science.
0 ?
Topics:
• Introduction - What are Expert Systems?; Synthesis and diagnosis tasks,
generative paradigm, recognition systems, general / specialized
knowledge.
• Historical Development - Pioneering systems and their contributions:
MYCThJ, PROSPECTOR, RI, INTERNIST,
etcetera.
• Logical Foundations - Abductive reasoning, inference engines and logic
programming, search.
• Expert System Architectures - Rule-based systems, schema and frame
representations, forward and backward reasoning, blackboard
architectures, explanations, Expert System Shells and programming
environments.
• Constraint Propagation - Domain constraints, local / global consistency,
k-consistency, practical consistency techniques, relationship to
constraint logic programming, use in real-time systems.
• Reasoning Maintenance Systems - Hypothetical worlds, logical support,
chronological
/
dependency bactracking, justification / assumption-
based systems, integration with constraint propagation.
• Uncertainty Calculi - Reasoning with uncertainty, abductive reasoning,
belief measures.
• Knowledge Engineering - Ontological analysis, knowledge acquisition,
learning techniques, explanation capabilities.
• Building Expert Systems - Prototype selection, initial knowledge
acquisition, system design, implementation and critical evaluation.
Prerequisite:
Introductory Artificial Intelligence (e.g. CMPT 410) or consent
of the instructor
Text: F. Hayes-Roth
et.aI.,
Building Expert Systems,
Addison-Wesley, 1983.
Grading:
a term project, occasional homework assignments and a final
examination will be required Of each student.
?
0
U-5
fl
?
SCHOOL OF COMPUTING SCIENCE
CMPT 827 Expert Systems
Library Requirements
The following books will be useful (but not strictly necessary) as reference materials
for this course:
• M. R. Genesereth & N. J. Nilsson (1987)
The Logical Foundations of?
Artificial Intelligence,
Morgan Kaufmann, Los Altos, CA.
• P. Harmon
et.al .
(1988)
Expert System Tools and Applications, John
Wiley, New York.
B. G.
Buchanan & E. H. ShortLiffe (1984)
Rule-Based Expert Systems,
Addison-Wesley, Reading, Mass.
• ?
• E. Charniak & D. McDermott (1984)
Artificial Intelligence,
Addison-Wesley,
Reading, Mass.
The following conference proceedings will be useful research materials:
Proceedings of the International Joint Conferences on Artificial Intelligence
• Proceedings of the biennial conferences of the American Association for Artificial
Intelligence
• Proceedings
of the biennial conferences of the Canadian Society for the
Computational Studies of Intelligence
The following journals will also be useful research materials:
•
IEEE Expert
•
Computational Intelligence
.
?
?
Note: All
of
these references are either already in the Library or are available from the
instructor for library reserve.
Competence of Faculty - CMPT 827
Dr. N. Cercone obtained his M.Sc. (Computer & Information Science) and his
Ph.D. (Computing Science) in artificial intelligence specializing in computational
linguistics and practical A! systems. He is an active researcher in computational
linguistics and expert systems and has taught material of this nature for 13
years. He is currently funded by NSERC with operating intrastructure, strategic,
and equipment grants. He is also editor of Computational Intelligence.
William S. Havens
Bill Havens is an Associate Professor of Computing and Engineering
Science and Director of the SFU Expert Systems Laboratory. He has
taught numerous courses in Artificial Intelligence and Expert
Systems at both the undergraduate and graduate level. His research
program involves developing constraint-based reasoning systems
and applying these techniques to industrial Expert Systems'
problems.
.
0
SIMON
MASER UNIVERSITY
?
NEW GRADUATE COURSE PROPOSAL
CALENDAR INFORMATION:
Department ?
Computing Science ?
Course Number:
831
Title: ?
Functional Programming
Description:
This course will cover functional programming, including programming
techniques, implementation methods and other selected topics.
Credit Hours:
?
3 ?
Vector:
3-0-0 ?
Prerequisite(s) if any:
ENROLLMENT AND SCHEDULING:
Estimated Enrollment-
_
10
—When will the course first be offered:
?
903 ?
How often will the course be offered:
Once: every two years.
JUSTIFICATION:
Functionprogrammingisbecominganincreasinglyimportanttopicincomputing
science.Thiscoursewillpreparesomestudents for M.Sc. or Ph.D. research
in functional programming and give _otherstudentsasolidintroductiontothistopic.
RESOURCES:
Which Faculty member will normally teach the course:
_F. Warren Burton
What are the budgetary implications of mounting the course-
_None
Are there sufficient Library resources (appen4 details):
Yes
Appended: a) Outline of the Course
b)
An indication of the competence of the Faculty member to give the course.
c)
Library resources
Approved:
Departmental Graduate Studies
?
Z2-
.
.
• ?
Faculty Graduate
,
Studies
Faculty:
Senate Graduate Studies
Senate: ?
Date:_____________
.
Computing Science
Course Outline
CMPT 831: Functional Programming
This course will introduce functional programming at the graduate level
programming.
to students who have only a limited knowledge of the area of functional
Outline:
1.
Introduction to a functional programming language (probably Miranda
or Haskell). This part of the course will cover the syntax and
semantics of a modern functional language. At the some time, functional
programming techniques, including use of lazy evaluation and higher
order functions will be taught.
A
pproximately half the course will be
spent on this material. (Note: Miranda is currently available on
several School of Computing machines that graduate students may use.)
2.
Program transformation and verification. This part of the course
will cover equational reasoning and inductive reasoning about functional
programs,
A
pproximately
including
two weeks
reasoning
will be spent
about
on these
infinite
topics.
data structures.
topic.
will
environment
3.
be
Implementation
considered.
based implementations
?
of
Approximately
functional
and comb
three
programming
j
nator
weeks
based
will
languages.
implementations
be spent
?
on this
Both
?
4.
Other aspects of functional programming. In the remaining portion
of the course (a week or two) other topics will be considered. These
may include parallel evaluation of functional programs, analysis of
performance, advanced applications or other topics.
Grading Scheme:
Midterm Examination: 20%
Class Project: 40%
Final Examination: 40%
r
1A
I
Computing Science
Library Requirements
CMPT 831: Functional Programming
Students may occasionally refer to papers in the following journals:
1.
ACM Transactions on Programming Languages and Systems.
2.
IEEE Transactions on Software Engineering.
3.
ACM Computing Surveys.
4.
IEEE Computer.
5.
Computer Languages.
The library current subscribes to all of these journals.
In addition, students may need to refer to the following books:
1.
Peyton Jones, Simon L., The Implementation of Functional Programming
Languages, Prentice-Hall, 1987.
2.
Field, Anthony J.,, and Harrison, Peter C., Functional Programming,
Addison-Wesley, 1988.
3.
Bird, Richard, and Wadler, Philip, Introduction to Functional
Programming, Prentice-Hall, 1988.
The library currently has these books.
Computing Science
Faculty Qualifications
CMPT 831: Functional Programming
Dr. Burton is an active researcher in the area
?
of
?
functional
programming. He has published over a dozen papers on functional
programming, including several in leading journals. He has previously
taught other courses on functional programming, both at the graduate
level and the undergraduate level, both at Simon Fraser University and
the University of Utah.
.
SIMON
FRASER UNIVERSITY
NEW GRADUATE COURSE PROPOSAL
CALENDAR INFORMATION:
Department
?
Computing Science
?
Course Number:
841
Title:
?
Query Processing In Database Systems
Description:
?
See attached page.
Credit Hours:
?
3 ?
Vector:
300 ?
Prerequisite(s)
-
.if any:
ENROLLMENT AND SCHEDULING:
Estimated Enrollment:
_
10
—When will the course first be offered:
?
911 ?
How often will the course be offered:
Once every two years.
JUSTIFICATION:
See attached page.
.
RESOURCES:
Which Faculty member will normally teach the course:
Dr. W. S. Luk
What are the budgetary implications of mounting the course:
None
Are there sufficient Library resources (append details):
Yes
Appended: a) Outline of the Course
b)
An indication of the competence of the Faculty member to give the course.
c)
Library resources
Approved: Departmental Graduate Studies
Faculty Graduate Studies
?
(0
Senate Graduate
?
______
/
I ?
•
Senate:
?
Date:______________
Course Number: CMPT 841
Title: Query Processing in Database Systems
Description: Algorithms for data-intensive operations for disk-based, main-memory-based,
loosely distributed and tightly coupled databases; Analytical and empirical performance
studies of database systems.
Justification: Database Systems, as a research ares, has become one of mainstream areas
in computing science. Over the years, existing database faculty have changed the focus of
their research, new researchers have joined the faculty, old topics have evolved and new
topics have sprung up. As a result, many database related courses have been offered as
special topic courses. This is the time to re-organize our database curriculum and formally
introduce new course titles to prevent proliferation of special topic courses.
Outline:
1.
Query processing algorithms for different database systems (e.g. relational
databases, deductive databases and object oriented databases)
2.
Characteristics of different hardware platforms (e.g. distributed databases,
main memory databases and database machines) and their implications in
• ?
performance of query processing algorithms
3.
Performance evaluation criteria: disk accesses, page accesses, computation
cost and communications cost.
4.
Performance evaluation methodologies: analytic and empirical techniques.
Expertise of the Instructor: Dr. Luk has had over ten years of research experience in
databases and has published widely in database journals and conference proceedings. He
has also served as a referee for numerous database journals and research grant agencies.
Library resources: The journals that are required for this course are already being
subscribed to by. our library (e.g. ACM Transactions of Database Systems, Information
Systems, Journals for Distributed and Parallel Processing and IEEE Transactions on
Software Engineering).
.,
53
SIMON .FRASER
UNWERSFY ?
NEW GRADUATE COURSE PROPOSAL
CALENDAR INFORMATION:
Department: ?
Computing Science ?
Course Number:
?
843
Tide e
Princi p
les of Database and Knowledge-Base Systems.
Description:
Deductive Databases, Semantic Data Modeling, Object-Oriented Databases,
Expert Database Systems, New D.B. applications in Al and Engineering.
Credit Hours: ?
3
?
Vector:
3-0-0 ?
Prerequisite(s) if any:
ENROLLMENT AND SCHEDULING:
Estimated Enrollment-
_
10
—When will the course first be offered:
903
How often will the course be offered:
?
Once every two years.
JUSTIFICATION:
New applications of Database and Artificial Intelligence requires an integration of
database and Al technology. The fast expansion of research and development in this
field proves its necessity and importance.
RESOURCES:
Which Faculty member will normally teach the course:
_Jiawei Han, Wo
. -Shun Luk
What are the budgetary implications of mounting the course:
_None
Are there sufficient Library resources (append details):
New journal, IEEE transactions on Knowledge
and Data Engineering is preferred.
Appended: a) Outline of the Course
?
(Vol. 1, 1989).
b)
An indication of the competence of the Faculty member to give the course.
c)
Library resources
Approved:
Departmental Graduate Studies
Faculty Graduate Studies
?
IM
Faculty: ?
Date:
_ij/' 0/,)•
Senate Graduate Studies Committee:
?
Date: :?'
Senate:
?
Date:_____________
School of Computing Science
?
Course Proposal
?
August 29, 1989
0 ?
Graduate Course Proposal
Department: Computing Science
?
Course Number:
CMPT-843
Title: Principles of Database and
Knowledge-Base
Systems
Course Description:
An advanced course on database systems which covers the following topics: seman-
tic data modeling, engineering databases and spatial databases, object-oriented data
models and systems, deductive database systems, semantic query optimization, learning
intensive
and induction
knowledge-base
in database
systems.
and knowledge-base systems, and architectures of data-
Credit Hours: 3
?
Prerequisite:
CMPT-740 or equivalent
Justification:
An
i
mportant recent development in Computing Science is the application of data-
such
tion
base
and
and
topics
kn
artificial
owledge-based
in many
intelligence
i
nternational
systems.
technologies
conferences,
A good indication
in the
the
development
publications
of such a
of
trend
of
many
engineering,
is the
new
emphasis
books
informa-and
of
• dedicated journals. We have offered such a course as a special topics course in Al
(CMPT-882) in the academic year of 1987 - 1988 and as a special topics course in Data-
bases (CMPT-884) in the academic year-of 1988 - 1989. Students love the contents of
the course and feel the importance of the field. We would like to promote the course to a
graduate-level regular course to enrich our curriculum.
Faculty members teaching the course:
Jiawei Han and Wo-shun Luk
Both have teaching experience in the area, have publications in journals and confer-
area.
ence proceedings, and are referees for some research journals and funding agencies in the
No new budget is required to set up the course.
Outline of the Course:
(1)
Semantic data modeling,
(2) Engineering
databases and spatial databases,
(3) Object-oriented
data models and systems,
(4) Deductive
database systems,
(5)
Compilation and optimization of recursive rules and queries
(6)
Semantic query Optimization
(7)
Learning and induction in database and knowledge-base systems, and
(8) A
rchitectures of data-intensive knowledge-base systems.
