CSC 8860 Seminar in Pattern Recognition
Course #: CSC 8860
Prerequisite: CSC7860 or approval of the instructor.
Day: T-Th
Room: 333 State Hall
Hours: 3.00 p.m. – 4.20.
Instructor: Sorin Draghici
Office: 408 State Hall
Office hours: Tue: 6.00pm – 7.00pm or by appointment.
Telephone: 577-5484
Email: sod@cs.wayne.edu
Web page: http://www.cs.wayne.edu/~sod
On this web page you can find the syllabus, and announcements regarding the course if any.
No textbook. The seminar will focus on the study of a number of selected research papers to be assigned individually by the instructor to each student.
The goal of this course is to present the main research topics in the field of pattern recognition with applications in bioinformatics. Each student will be assigned a number of research papers to study. A project will be undertaken on an individual basis. At the end of the semester, the students will submit a report describing their work. The grading will be based on this report. Each student will give two presentations during the semester. One presentation will be given in early November and the second presentation will be given at the end of the semester.
No examinations are scheduled for this course.
CSC 7991 Introduction to Data Mining
Course #: CSC 7991
Prerequisite: CSC 7850 or approval of the instructor. .
Day: T-Th
Room: 0129 State Hall
Hours: 4.30pm- 5.50pm
Instructor: Sorin Draghici
Office: 408 State Hall
Office hours: Tue: 6.00pm – 7.00pm or by appointment.
Telephone: 577-5484
Email: sod@cs.wayne.edu
Web page: http://www.cs.wayne.edu/~sod
On this web page you can find the syllabus, and announcements regarding the course if any.
Required: Lecture notes
Recommended: Data Mining: Practical Machine Learning Tools and Techniques with Java Implementations - Ian H. Witten, Eibe Frank
The goal of this course is to present the main data mining techniques. The intended audience includes as a central figure the researcher, in particular life scientist, that needs to use computational tools in order to analyze data. At the same time, the course is intended for the computer scientists who would like to use their background in order to solve problems at the border of biology and medicine. The course explains the nature of the specific challenges that such problems pose as well as various adaptations that classical algorithms need to undergo in order to provide good results in this particular field.
Final exam: Friday Dec 13
Midterm exam: Tue, Nov 5
Data mining and machine learning; simple examples; input: concepts, instances, attributes; output: decision tables, decision trees, classification rules, association rules; algorithms: divide and conquer, covering algorithms, mining association rules, linear models, instance based learning; Evaluating the learning: training and testing, predicting performance, cross-validation, leave-one-out, bootstrap.
CSC 6800 - Introduction to Artificial Intelligence
Please use these
practice exercises
. Being able to solve these exercise is a necessary but
not sufficient condition for your success. More exercises can be
found on Lila Kari's home
page
.
Homeworks:
Homework1 assigned 25 Sept, due 9 Oct, 2001.
Important observation regarding the homework submission:
You will need to have a CS account in order to submit your homework. If you do not have a CS account yet, you need to apply for one in the main office, room 431 State Hall. You will need to email me your login name.
All files regarding homework number n must be kept in a
directory called homeworkn in your home directory.
Change the permissions in your home directory as follows
(the R is capital R, not r):
chmod -R o+rwx homework3
The homeworks will be collected by a program that will be launched at 12midnight the day of the deadline. For instance, homework 1 will be collected at 12.01am 10 October. The maximum grade for late homeworks will be reduced by 10% per day. A perfect homework submitted one day late (e.g. between 12.01am 10 Oct and 11.59pm 10 Oct) will get a score of 90% and so on.
Other notes:
If Adobe Acrobat is already installed on your system, you can just click on the links below in order to view the transparencies. You can also print them from Adobe Acrobat/Reader by selecting Print from the File menu.
You may want to print several pages on one sheet of paper. In order to do this, proceed as follows:
> 1. Open the file with Adobe Reader.
> 2. Print it to a postscript file.
> 3. Get hold of a program called psnup (if you are
using a Unix/Linux
> machine it should be already there - you just run
it; if you are using
> a Windows machine, you need to look it up and download
it from
> somewhere).
>
> The conversion is done with:
>
> psnup -4 -b1mm -m1mm -d yourfilename.ps anotherfilename.ps
>
> This would put 4 pages on a single sheet. You can
use (almost) any value there.
CSC 7991 -
Data Analysis of Microarray Data
Homeworks
Which files do I need to process for Homework 1?
Projects
Various
examples and further explanations
of the statistical concepts discussed.
CSC 7991 - Seminar in Neural Networks
CSC 7850 - Introduction to Neural Networks
The course in
Data Mining and seminar in Vision and Pattern
Recognition offered in the Fall might be of interest to the students
who have already taken Neural Networks..
Homework 1:
CSC 1050 - Introduction to C and Unix
The ASCII version of the transparencies used during
the second part of the course (C Programming):
The final grades for both sections (1 May 1998).
CSC 4420 - Computer Operating Systems
The syllabusThe syllabus of the course is available here (last update 1 September 1998).
The ASCII version of the transparencies used is available for download in pdf, postscript or powerpoint format. You will need Adobe Acrobat Reader in order to view or print pdf files. You can download Adobe Acrobat Reader here .
If Adobe Acrobat is already installed on your system, you can just click on the links below in order to view the transparencies. You can also print them from Adobe Acrobat/Reader by selecting Print from the File menu.
You may want to print several pages on one sheet of paper. In order to do this, proceed as follows:
> 1. Open the file with Adobe Reader.
> 2. Print it to a postscript file.
> 3. Get hold of a program called psnup (if you are
using a Unix/Linux
> machine it should be already there - you just run
it; if you are using
> a Windows machine, you need to look it up and download
it from
> somewhere).
>
> The conversion is done with:
>
> psnup -4 -b1mm -m1mm -d yourfilename.ps anotherfilename.ps
>
> This would put 4 pages on a single sheet. You can
use (almost) any value there.
Exams
Midterm exam
Sample exam questions are available here .
Sample answers for some of the questions are available. These answers show you the level of detail and the style you should use when answering an exam question.
The correct answers for the multiple choice questions are available here . Also, you can have a look at the complete solutions of the exam questions.
Final exam
Sample exam questions
are available
here
.
Grades to date
All files regarding homework number n must be kept in a directory called homeworkn in your home directory
Change the permissions in your home directory as follows (the R is capital R, not r):
chmod -R g-rwx homework3
chmod -R o+rwx homework3
Check the result of these commands by doing:
ls -al
should have no permissions for the group and all permissions for others:
-rwx---rwx yourname homework3
The homework is considered submitted only if
the permissions are set appropriately. If the permissions do not
allow me to inspect the files, you will not get any credit for this
homework.
CSC 1010 - Introduction to Computer Science