Course Project
Project demo on December 11 (Wednesday) and December 12 (Thursday).
Project proposal due on November 21 (Thursday).
You will perform a final course project on a big data-related topic.
You can either work individually or form a group of two.
You are free to select your own project topic, as long as it is
substantially related to big data computer systems. You may also leverage your past
work (on another course project or a research study), in which case
you must disclose such relevance and explain the new work you plan
to do for this project. To help you select a project topic, we
provide below a list of possible ideas for you to consider.
Again, you are welcome to search beyond these suggested topics for
your project.
- Data collection and analysis: You can work on a project
concerning big data collection and analysis. For instance, if
you are interested in image analysis, you may consider designing
a web crawler that collects images on the web that are related to
a particular topic (e.g. flowers, animals, or waterfalls).
Your project can then perform some image analysis to extract
interesting patterns. Parallel collection and analysis may be
desirable to operate on large datasets.
- Analyzing your big data: You may have access to some
valuable dataset(s) that haven't been sufficiently analyzed (or
not yet made public). It can be beneficial to perform analysis
on the dataset(s) to identify new characteristics. It may also be
useful to validate/compare data analysis approaches (e.g. clustering).
- Parallelizing your big data application: You may have
particular experience with some data-intensive application(s) and
find it desirable to parallelize it/them for high scalability.
You want to find out whether the MapReduce/Hadoop paradigm fits your
application scenario and understand why. Unlike the classic
parallel computing, your work should focus on the data-intensive
aspect of the parallelization.
- Big data computer systems: We learned several open-source
big data computer systems in class (Hadoop, memcache, etc.). You
may also know some other open-source systems supporting big data
applications. You can perform a project to dive deeply into one
of these systems. You may install and run them to understand
their behaviors, identify potential pitfalls, and possibly try to
develop some enhancements.
- Public eavesdropping: Wireless communications carry
a large amount of data in the open environment. For instance, it
is possible to use your WiFi device to listen to the WiFi signals
in some public place. You can analyze the statistics on data
encryption uses in public WiFi data. Note that such a project
requires substantial expertise on computer networks and you should
be confident on your possession of relevant skills before choosing
a project of this kind.
Please speak to the instructor for any questions or consultation on
the topic and/or scope of your project. The best time for such talks
are the instructor's office hours and the time right after the classes.
Meetings at other time can also be accommodated.
Turn-in:
You must email a project proposal to the instructor by November 21 (Thursday).
In the proposal, you must specify your team members. You should then
describe concrete topics and goals (including the scope of expected work) of
your project. It'd also be helpful to identify the datasets and tools to be
used in your work.
Please use "CSC296/576 Project proposal" in the subject line of your project
proposal email. Failure to email the project proposal on time will result in
a 20% loss of your project grade.
The instructor will try to provide some feedback to your project proposal. Note that
your proposal doesn't need to be perfect. In fact you won't lose any point
on it as long as you email it by the deadline. We are using this mechanism
to make sure you start the work early and to assess the direction of your
project work.
The final completion of your project will be primarily examined by a
20-minute demo. The TA will send out information on how to set up a demo
slot.
You are also asked to prepare a project report which should explain your
work and findings. Please make sure that you have
a hard copy of the project report ready at the beginning of your demo.
You should electronically turn in your source files, a makefile if needed,
and the project report before or right after your demo.
Late turn-in policy:
At the absence of any emergency, you must make your arranged demo time. Otherwise
you will get no points for the course project.