Objective: The objective of this course is to give an advanced introduction to the concepts for modeling, designing, querying and managing large databases. The course covers a spectrum of topics involved with current approaches to modeling and design of databases and the design of DBMSs to manage databases. The relational model is emphasized and relational database management systems are addressed from the standpoint of query optimization, database security, transaction management, concurrency control, and recovery. Other topics introduced include object-oriented databases, object-relational database management, data warehousing and mining. This course is an overview course which is followed by specialized database courses: CS6411: O-O database models and systems, CS 6421: active and dynamic DBMSs, CS6430: parallel and distributed database systems and applications. A basic knowledge of programming languages, files, and the application development process is assumed. A first course in database management covering introduction to relational, hierarchical and network databases is useful but not a must. Those with sufficient undergraduate database coursework are advised to go straight to one of the advanced courses.

Instructor: Prof. Sham Navathe Office: CoC 139

Phone: 894-0537 Office Hours: to be announced

e-mail: sham@cc and by appointment.

Class time: Tues-Thurs. 4:30 -6:00. Secretary: Gwen Baker: 4-8358,

Class room: CoC. 102 Rm. 264.

Teaching Assistant: Waigen Yee (waigen@cc)

Phone: 894-7008 Office Hours: M2:45-3:45, Tu2:30-3:30, or by appointment

Textbook:

[EN] R. Elmasri and S.B. Navathe, Fundamentals of Database Systems , Benjamin Cummings, Edition 3, 2000.

Note: Edition 3 is substantially revised from edition 2 with new chapters on Oracle/Access (Ch.10), a full section on object-oriented and object relational technology (Chaps. 11-13), enhanced data models (Ch. 23), data warehousing and mining (Ch. 26), emerging technologies and applications (Ch.27). Most chapters including the SQL chapter (Ch.8), and the section on system implementation techniques (Chapters 17-21) have been revised.

Some papers may be assigned as additional readings.

Students will work on one of the two different deliverables: (a) a project, involving either the use of a DBMS, or an implementation of some aspect of a DBMS, or some innovative application of database concepts, to be done in teams of two or three, and (b) a research term paper to be done individually which goes beyond simply summarizing the published work in some area.

Students are welcome to work in teams to propose a project of their own to develop some aspect of a database system, create some innovative application, or investigate or evaluate a database management system or tool. I am completely open ended about this and would like students to think of applications in their own areas of interest and see how current database technology-based solutions can be developed or suggest techniques and approaches that merit further investigation. This is your opportunity to investigate how DB technology may be applied to graphics, robotics, virtual reality, network management or any such applications. Students should hand in a proposal by September 23rd stating (i) the goal, (ii) problem definition (iii) description of design, approach, or unique features (iv) what will be implemented, what will be written up and what will be demonstrated. The platform, the system, language etc. is totally flexible. Students are encouraged to interact with the other Database faculty and Ph.D. students so as to define a "meaningful yet small" project that integrates with any ongoing research in the database group. You may also consult the web pages for the database group under the category of research from the college of computing home page. Those with interest in pursuing the database field further are strongly encouraged to try to interact with on-going research activities and do a project that will lead to a further independent study (CS 8503).

Examples under A:

i) implementation of a specific file access or indexing scheme for a temporal database

ii) use of a DBMS to build an image database

iii) implementing ideas from one of the papers under Storage Management in [S]

iv) implementing a DBMS that simulates storing, accessing, and a limited querying of data in the ER model.)

v) investigating the performance of data mining algorithms

vi) a medical or bioinformatics application using a DBMS

vii) an engineering information system

viii) databases for virtual reality

ix) implementing some aspect of a genome or geographic data management system (see Ch 27).

x) Development of a web-based querying engine for unstructured data.

B. RESEARCH TERM PAPER: Hand in the proposed topic, the scope and focus of the paper

and a preliminary set of references by September 23rd. The instructor and the TA will try to help you with further search for material. The term paper serves two goals: one- it allows "non-computer science" majors to apply database technology to their field of interest, two- it allows one to explore topics in which one could do further independent study, do a masters' project, or take an advanced DB seminar later. Please refrain from writing a general paper - say, on multimedia database management. That topic could be broken into: content-based retrieval from multimedia data, factors affecting quality of service in multimedia databases, efficient handing of stream data, applying data mining to images, etc.

A possible list of topics. (This is only a suggested list and you will have to refine each topic.):

1. Distributed Databases. (Take some aspect: query processing, concurrency control, recovery, distribution design. see Ch. 21 in [EN]. Also see books by Ceri/Pelagatti and Ozsu and Valduriez.)

2. Database Performance Measurement Techniques

3. Knowledge Management (Knowledge representation, recursive query processing, rule processing and optimization, etc.)

4. User Interfaces and data visualization.

5. Concurrency Control and/or recovery algorithms for specific applications - e.g., mobile databases, engineering design databases. (see a book by Bernstein et al.).

6. Active Databases (see book by Widom and Ceri).

7. Different aspects of Object-oriented Database Management (e. g., query languages, theoretical models, storage organization). (Look at a readings book by Zdonik and Maier or books on reserve.).

8. Database management for CAD/CAM and manufacturing applications

9. Geographic Information Systems- database issues

10. Office Information Systems- database issues. (Look at issues of ACM Trans. of Office Information Systems. The word "office" has been recently dropped from the title.)

11. Database Security - security models, security implementation, relationship to web databases.

12. Parallel Databases: architecture, query processing, join algorithms, performance. (consult Prof. Omiecinski for references.)

13. Temporal Databases - language issues, storage and transaction management (consult Prof. Mark).

14. Multimedia Database Management

15. Distributed Database Design, Redesign, Reorganization.

16. Database management issues on the web.

17. Workflow modeling and process modeling - techniques and tools.

Students are welcome to propose any interesting topics. Papers should be 10-12 pages (double spaced, 12 font) and will be graded on the basis of :(i) content- amount of effort and breadth, (ii) understanding and synthesis of the topic shown by the author, (iii) analysis and depth that goes beyond just copying parts of references, and (iv) organization and presentation. Creative papers exploring new ideas or techniques coupled with some experimental proposals are most welcome.

Due date for term paper/project: December 9th (a hard deadline.) Papers must be submitted in duplicate. (One copy will be returned). Project deliverables will be announced later. Each paper must have a detailed bibliography in the following style (a paper must consult at least 5 outside references).:

1. T. Wakayama, S. Kannapan, C. M. Khoong, S.B. Navathe, J. Yates (Eds.),Information and Process Integration in Enterprises: Rethinking Documents, Kluwer Academic Publishers, 1998. [ An edited book].

2. S.B. Navathe and R. Ahmed, "Temporal Extensions to the Relational Model and SQL," Chapter 4 in "Temporal Database Management," (A. Tansel, et al., eds.), Benjamin Cummings, 1993. [ A chapter from a book].

3. H. Beck, T.Anwar, and S.B. Navathe, "A Conceptual Clustering Algorithm for Database Schema Design," in IEEE Transactions on Knowledge and Data Engineering,

Vol. 6, No. 3, June 1994. [ A paper in a journal].

4. A. Savasere, E. Omiecinski and S. B. Navathe, "Discovery of Multiple-Level Association Rules from Large Databases," Proc. 21st Int. Conf. on Very Large Databases, Zurich, Switzerland, September 1995. [ A paper in a conference].

Two closed book exams : 40% (breakdown among two to be announced)

Final (cumulative addressing selected topics): 25%.

Quizzes (in class) and graded homework: 15%.

Project or paper: 20%.

Exams will be curved independently and assigned a letter grade (on a continuous 0.0-4.0 scale). The term paper will also be given a letter grade. The final grade will reflect a weighted average of these letter grades. (E.g., B is 3.0, B+ is 3.33, A- is 3.67, A is 4.0). Those with weighted total above 3.5 receive an A. The instructor may use subjective judgment to adjust "border-line" cases up or down.

Wai Gen Yee (waigen@cc), a Ph.D. student, will be the T.A. for this course. His phone number: 894-8791/894-7008. His office location : CoC 104A (database lab). Office hours: To be announced. He will maintain a web page for this class at www.cc.gatech.edu/classes/cs6450

http://dblp.uni-trier.de maintained by Dr. Michael Lay at Univ of Trier, Germany).