Research

Metamorphic Petrology

Mon, 15 Jan 2007 11:22:15 -0500

Metamorphic petrology is the study of the changes (metamorphosis) in rocks that are due to heat and pressure. We are in the process of building a database (MetPetDB) that will store and make available public data sets on metamorphic petrology. The data sets will contain information on metamorphic samples and subsamples, chemical analyses and images of samples, subsamples and other analyses. Furthermore, the database will contain information linking the data to existing publications, people and institutions. This data will be uploaded by the users of the system through an upload facility. The database will make it possible to search and browse the available data and run different daa analysis tools on top of data.

Related Information:
Bouchra Bouqata, Adam Marcus, Frank Spear, and Boleslaw Szymanski, ``A day in the life of a metamorphic petrologist'', in Proceedings of the Semantic Web and Databases Workshop, 2006. [swdb2006.pdf][swdb_Presentation.pdf]
Wiki for our development page.

Managing Activities

Tue, 14 Mar 2006 23:08:46 -0500

Users create and modify data as a function of activities they are involved in. However, most desktop management systems provide an object oriented view of a computer, where the emphasis is on describing the information relevant to an object. However, objects may be related to each other in different ways and in different contexts. We develop methods to explore objects on a desktop through activities. Our methods integrate discovered relationships between data objects based on their participation in different activities as well as other properties. Our activity model is well suited to support a query language that is able to alter the context and the definition of an activity to easily visualize complex relationships in data. We show that this new organization makes many new interesting desktop functionalities a reality.

Rank Aggregation vs. Ranker Quality

Tue, 14 Mar 2006 22:25:55 -0500

The rank aggregation problem has been studied extensively in recent years with a focus on how to combine several different rankers to obtain a consensus aggregate ranker. We study the rank aggregation problem from a different perspective: how the individual input rankers impact the performance of the aggregate ranker. We develop a general statistical framework based on a model of how the individual rankers depend on the ground truth ranker. Within this framework, one can study the performance of different aggregation methods. The individual rankers, which are the inputs to the rank aggregation algorithm, are statistical perturbations of the ground truth ranker. With rigorous experimental evaluation, we study how noise level and the misinformation of the rankers affect the performance of the aggregate ranker. We introduce and study a novel Kendall-tau rank aggregator and a simple aggregator called PrOpt, which we compare to some other well known rank aggregation algorithms such as average, median and Markov chain aggregators. Our results show that the relative performance of aggregators varies considerably depending on how the input rankers relate to the ground truth.

Link Bombing

Tue, 14 Mar 2006 21:50:19 -0500

We analyze the recent phenomenon termed a Link Bomb, and investigate the optimal attack pattern for a group of web pages attempting to link bomb a specific web page. The typical modus operandi of a link bomb is to associate a particular page with a search text and then boost that page's pagerank. In an attack, the attacking pages can only control their own content and outgoing links. Thus, when a search is initiated with the text, a high prominence will be given to the attacked page. We show that the best organization of links among the attacking group to maximize the increase in rank of the attacked node is the direct individual attack, where every attacker points directly to the victim and nowhere else. We also discuss optimal attack patterns for a group that wants to hide itself by not pointing directly to the victim. We quantify our results with experiments on a variety of random graph models.

Professional Activities

Tue, 14 Mar 2006 19:17:56 -0500

Sigmod Undergraduate Scholarship Program (SIGMOD-2006)
ACM Conference on Information and Knowledge Management (CIKM-2006), Program Committee Member
The 5th International Conference on Image and Video Retrieval (CIVR-2006), Program Committee Member
Ontologies, Databases and Semantics (ODBASE-2006), Program Committee Member
3rd International Conference on Data Engineering (ICDE-2007), Program Committee Member
Advanced In Information Systems, Fourth Biennial & International Conference, Izmir, Turkey, (ADVIS 2006), Program Committee Member
21st International Symposium on Computer and & Information Sciences, Istanbul, Turkey (ISCIS 2006), Program Committee Member
ACM Sigmod Undergraduate Scholarship Program, Chair
ACM SIGMM Multimedia Information Retrieval Workshop, Steering Commitee Member
Second Computer Vision meets Databases (CV/DB) & Workshop, Baltimore, MD 2005 (in conjunction with SIGMOD/PODS 2005), Program Committee Member
Eleventh Workshop on Multimedia Information Systems, 2005, Program Committee Member
20th International Symposium on Computer and & Information Sciences, Istanbul, Turkey (ISCIS 2005), Program Committee Member
2005 Int. Conf. on Mobile Data Management, & Larnaca, Cyprus (MDM'2005), Program Committee Member