Past Project Topics
Here
are some ideas for possible course projects. Keep in mind that we are
not interested in merely understanding a specific paper, implementing
the presented algorithms and demonstrating that you understand their
contributions well enough to reproduce it yourself. The watchwords are innovation, creativity, and novelty. As always, depth will
always bring its rewards (just as it does in paper presentations, in
discussions led, and in the subsequent summaries prepared).
In
addition to these suggestions, I am happy to listen to other ideas for
projects, if you have any of your own. I am also happy to work with you
on creating other project ideas. No matter which project you choose, I
will be available to discuss with you any technical issues related to
your project.
Each project can either take the shape of a paper or an implementation. It’s your choice. Each team may contain up to two people. (If the course enrollment changes significantly, we will revisit this rule.) Regardless of whether you do a project alone or as a team of two, my expectation will be that every individual does the same amount of work. Teams must be formed by Friday, January 15. Project choices should be made by Wednesday, January 20. Here, then are the project ideas. (More details on the project schedule and milestones will be posted soon.)
1. Social Network Computing: Social nets such as facebook and del.ico.us have
facilitated massive data, resource and opinion sharing, and in general
online discourse. The resources can range from online resources like
videos, photos, and blogs to material objects like lawnmowers and
bikes. (If you are wondering about what material objects are doing in a
social net, talk to me!) The opinions range from political opinions to
impressions about the job done by the recently hired handyman.
Additionally, the blogosphere is
awash with information. On the one hand, the information in social nets
is mostly free text and is thus largely unstructured. Thus, traditional
IR techniques as well as techniques based on centrality notions such as pagerank can
be effectively used to facilitate search over social nets. On the other
hand, though, social nets have their own structure, bringing the notion
of a seeker’s social neighborhood to the context of search. Just
focusing on search alone, there are interesting questions that need to
be studied:
- What is the right granularity of search? One could be searching for blogs, photos, resources, etc., or simply for opinions. Or one could be analyzing opinions.
- Suppose
we call the granules above “infons”. What techniques can we employ for
effectively searching such a collection of infons? What kind of
analyses are enabled by the combination of text data and the social net structure, together withchronology of opinions?
- How can we carry out these tasks efficiently?
Note:
This is not one project. By focusing on different angles, more than one
project can be formulated around the above description.
2. Link Prediction and Network Evolution: Substantial
work has been done on link prediction in social networks and on
modeling the evolution of the network. Two possible project ideas
follow:
(a) Monitoring and Tracking of underlying network in newsgroups and/or blogs: Analyze
the communication structure in newsgroups or blogs and discover the
underlying social network present there. The framework should be
parameterizable (e.g., strength of links, period of activity, etc.) and
should be queriable (e.g., topic of conversation, who started a
particular discussion, etc.). Visualize the network in creative and
interactive ways.
(b) Predict links and build models of network evolution: Given
a social network together with properties of nodes and/or edges (in the
form of keywords or more structured attributes), what is the state of
the art in link prediction that takes all available information into
account? How can we go beyond it? Can we incorporate time in modeling?
3. Viral Marketing:
4. Trust & Reputation:
5. Recommender Systems:
6. Top-k Query/Search:
7. Opinion Mining & Social Networks: