EECS 584 will cover a number of advanced data mangement topics, including issues in relational database management systems, data-centric applications, and Web systems. The specific topics include advanced concurrency control techniques, query processing and optimization strategies for relational database systems, advanced indexing methods, parallel and distributed database systems, next-generation data models, data mining on large databases, data on the web, and topics in data security and privacy.
In addition to learning about advanced topics in databases, this course will also give
you the opportunity to practice important research skills:
- You will gain experience reading and critically evaluating original research papers.
- You will practice communicating complex technical material, both orally and in written form.
- You will complete a small-scale original research project of your own choosing.
The prerequisite for this course is EECS 484, equivalent coursework, or permission from the instructor.
- Location: MW 1:30-3 PM, 2166 DOW
- Instructor: Michael Cafarella
- Instructor's office: 4709 CSE
- Office Hours: Wednesdays 3-4pm, and by appointment
- Presenter preparation meeting:For paper presenters for the coming week: Fridays 1:30-2:30pm
There is no official textbook for this course. The reading list is a collection of papers, which will be posted on the course web page.
You should be familiar with the introductory-level material covered by EECS 484,
and I will occasionally suggest review readings from the textbook used in that course. The text is:
Database Management Systems (3rd edition) - by Raghu Ramakrishnan and Johannes Gehrke, McGraw Hill, 2003.
Copies of the book are on reserve at the library, so you do not need to purchase it unless you really want to. You may also refer to the course material from EECS 484.
Most class periods will consist of a student-run presentation (typically 45 minutes), followed by class discussion led by the instructor. Each student will be assigned to present one paper during the semester. The presenter should be prepared to answer questions from the class.
Everyone in the class, whether a presenter or not, is expected to attend class and contribute to the discussion. The instructor is proud of the 584 tradition of class discussions that are lively, interesting, and involve as much of the class as possible. You can help keep this tradition going by reading papers carefully and actively participating in class.
When you present a paper, your goals should be as follows:
- Motivate the paper, and provide background. Why is this paper important? What problem does it solve? In some cases, you may
find it useful to refer to related material (e.g., from the textbook), and to explain how the concepts
in the assigned paper advance the state of the art.
- Provide an overview of the paper's key contributions.
- Illustrate key technical points. (Examples are a great way to do this.)
- How does it differ from work that came before or after?
- Finally: what should we remember about this paper?
The paper presentations form the basis of class discussion, so it is important that we have good ones. Giving a good paper presentation is difficult, requiring both understanding the paper and lots of preparation. There will not be time to cover all
of the technical details, so you will need to decide which details are most relevant. I am always available to talk about papers ahead of time if you have questions about your presentation.
You can find lots of places online for hints on how to give a talk (these pointers are pretty good). One point that is often underestimated is the usefulness of repeatedly practicing and refining your talk. You will find errors and organizational problems that you will not find if you simply write up the slides. Make sure you leave yourself lots of time to understand the paper and to go through several drafts of your presentation. Inexperienced presenters often start preparing too late and find themselves pinched on time.
Don't worry too much if you feel you lack traditional "public speaking skills." Good organization, clear descriptions, and a distillation of the most important points from the paper are the most important aspects of your talk. Your grade will be based on these factors. Intellectual preparation, not slickness, will make your presentation a great one.
Student presenters are required to meet me in my office each Friday during the time allotted for the discussion section (1:30 PM for the student
presenting the following Monday, and 2:00 PM for the student presenting the following Wednesday). You should come to this
meeting having read the paper and prepared a draft version of your presentation (in the form of PowerPoint slides).
Please create your slides in PowerPoint using this template.
After your presentation, I will post your slides on the course website.
After each presentation I will moderate class discussion of the paper. The important word here is moderate
: I expect people in the class will do most of the talking. Typical topics will include questioning the problem's motivation, key algorithms, and the experimental methodology.
All students are expected to read the assigned paper before class.
By 8 PM the day before each class, you must write a brief reaction
to the assigned paper.
This serves two purposes: (1) It gets you thinking about the paper before class,
and (2) It lets me know what people think about the paper and how to best use discussion time.
Your reaction should be between 1/2 and a full page: say, 300-400 words. Be sure to answer the following questions:
- What problem is addressed by the paper, and why is this problem important?
- What are the 1-2 main technical contributions of the paper? Describe them.
- What are 1-2 weaknesses, or open questions, from the paper? Describe and discuss them.
You should post your reaction to the Discussion forum on
. You should be able to sign into CTools using your UM uniquename.
Please try this well ahead of time, and contact the instructor if you have problems.
You are on your honor not to read or use other students' posts until you have submitted
your own. However, once you have submitted your reaction, you are encouraged to read other
students' reactions before class.
The paper reactions make up a small part of your grade. These are graded on a spot-check basis and assessed only as satisfactory/unsatisfactory. Assuming you write a plausible reaction, the only way to do "poorly" on this part of your grade is not to hand them in. (Reactions handed in after the deadline will not be accepted and for grading purposes will not be counted.)
A big component of this course is a class project. For this project, you will choose a research topic in the area of data management, and explore it in detail. Projects can range from quite theoretical to implementation-heavy. They should include some component of original work; in other words, survey articles are not permitted. You may choose to implement an existing algorithm or technique, but this should be done in order to conduct a unique experiment, or to test a novel hypothesis.
If you have questions about what constitutes a research project, please see me. I will provide a list of topics that you may find interesting. You are free to choose a project from the list or to be creative in developing your own ideas. You are encouraged to choose a project that intersects with your own research interests. For example, if you are interested in networking topics, you may want to pursue something in the area of streaming or distributed databases.
Projects should be done in groups of two. Each person in a group will receive the same grade.
To get you started on the right track, you will be required to submit a project proposal early in the semester, and I will schedule a mid-semester status meeting with each student/group. I am always happy to discuss projects during office hours or by appointment. The result of a research project is often unpredictable, so it is extremely important to take the initiative early in the semester to explore your topic well ahead of the deadline.
|Midterm Exam: 30%
||The exam will be in-class and open-book.
||Your project grade will be based primarily on your final paper and presentation.
Both students in a group will receive the same grade.
|Paper Reactions: 10%
||Each student is expected to read the assigned paper(s) and to submit a reaction for each paper by 8 PM the day before class.
You will receive full credit if you submit by the deadline a reaction that follows the guidelines and indicates that you have read and understood the paper.
I will spot-check the discussion posts on a random basis.
|Class Presentation: 15%
||Each student will present a paper in class.|
Presentations will be graded on a scale of 0-3 as follows:
- 0 - Student fails to complete assigned presentation
- 1 - Presentation is disorganized, or lacking in important details
- 2 - Presentation is solid, and covers important points with reasonable clarity (typical score)
- 3 - Presentation is exceptionally insightful (given infrequently)
|Class Discussion: 5%
A qualitative assessment of your contributions in class discussion.
Students with documented disabilities (including invisible disabilities) are encouraged to contact the instructor during the first two weeks of class. The instructor will attempt to make any necessary arrangements. All conversations will be kept confidential.
Collaboration and Honor Code Policies
All students in the class are presumed to be decent and honorable, and all students in the class are bound by the College of Engineering Honor Code. You may not seek to gain an unfair advantage over your fellow students; you may not consult, look at, or possess the unpublished work of another without their permission; and you must appropriately acknowledge your use of another's work. Any violation of the honor policies appropriate to each piece of course work will be reported to the Honor Council, and if guilt is established penalties may be imposed by the Honor Council and Faculty Committee on Discipline. Such penalties can include, but are not limited to, letter grade deductions or expulsion from the University. If you have any questions about this course policy, please consult the course instructor.
In general, collaboration in this class is allowed and encouraged except during the midterm exam. Here is a bit more discussion of how collaboration applies to each of the assignments in this class:
Paper Reading and Paper Reactions
You are encouraged to discuss papers with your fellow students; some can be difficult to read through, so it may help to go through them in a small group. However, you should write your reactions separately without group discussion or collaboration. The contents of a paper reaction should be exclusively your own work.
You are expected to prepare PowerPoint slides for your in-class presentation(s). You may find relevant diagrams and other material online, and in some rare cases you may be able to find many appropriate slides, perhaps written by the author of the paper. You may use any material, as long as you cite the original author on your slides
. The primary goal is to give a good presentation, not to evaluate your PowerPoint skills.
Regardless of the source of your material, you are responsible for everything in the presentation. Whether or not you wrote each slide by hand, it is expected that you will know and will be able to intelligently discuss everything in the presentation. It should be clean, clear, and relevant to our class. If you use material from other sources, you will almost certainly still need to perform substantial work before you understand it, can explain it, etc.
Final projects are done in groups of two. Although you are encouraged to choose topics that are relevant to your interests, the final project must be a piece of original work that is not submitted to any other class for credit.
All group work is to be completed only within your own group. You may receive help from the course instructor and you may consult with others, but you must complete your group's research, project write-up, and final project presentation on your own.
Each student must complete the exam solely by her or his own efforts. Questions can be asked only of the course instructor. The exam must be completed within the specified time.