 |
Steven P. CrainHi. I'm a PhD student in Computational Science and Engineering at Georgia Institute of Technology. I am interested in the automatic discovery of important patterns, and believe that hierarchical and cascaded machine learning algorithms are key to reaching this goal. |
I am interested in helping people use Web search to solve problems. Without doubt, modern Web search is very useful for many activities. However, new developments in Web search will greatly increase the class of problems that it benefits. My research will enable users to better leverage Web search to solve problems.
Current Web search technology uses feature-based machine learning techniques. Two kinds of features are used: document features reflect the quality of a document independent of the search task; query-document features attempt to capture the relevance of each document to a particular query. Consider the document features (for example, the media type, and Page Rank)|their applicability changes if the user is looking for other people's opinions of a product (which tend to be short and have negligible Page Rank) instead of looking for factual information. Likewise, the significance of query-document features varies for different queries and search tasks. In traditional Web search, the query serves as proxy for the user and the search context (the role the search plays in problem solving). More formally, the relevance of a document is assumed to be independent of the user and the search context when the query is known. I am interested in more complex models that allow the search engine to incorporate its knowledge of the query, the user and the search context when ranking documents.
Also see my curriculum vitae.
Burning Your Security At Three Ends: Security, Trust and Privacy in the Age of Disclosure. This is a talk delivered at the Web Science Tea 4/18/2008.
I looked at the issues relating to security, trust and privacy. We discussed how security, trust and privacy are being melted from three directions: the human drive for community; the very attempts to improve security, trust and privacy; and, corporate profit.
Our discussion was be guided by several different security models. The economic model, which considers that we "buy" trust with "payments" of private information is currently very popular. We also used models from information security (for example, the characteristics of a good password).