What we do at IDEAL
- We are primarily concerned with analyzing data in various forms - web logs, hypertext, XML, databases, signals and images - in order to characterize and understand the underlying phenomena. Some of our approaches use neurobiologically inspired models while others are more closely related to statistical methods or machine learning and AI techniques.
Yahoo!-UT Austin Data Mining Seminar Series
News
- IDEAL member Meghana Deodhar is runner-up for Best Student Paper at ICML09
- "A Scalable Framework for Discovering Coherent Co-clusters in Noisy Data" by Meghana Deodhar, Joydeep Ghosh, Gunjan Gupta, Hyuk Cho and Inderjit Dhillon
- IDEAL member Alex Liu wins the Best Student Paper Award at DMIN07
- "Generative Oversampling for Mining Imbalanced Datasets" by Alex Liu, Joydeep Ghosh, and Cheryl Martin.
- IDEAL alumna Srujana Merugu wins the Best Research Paper Award at KDD07
Deepak Agarwal and Srujana Merugu at Yahoo! Research captured the Best Research Paper Award for their work on "Predictive Discrete Latent Factor Models For Large Scale Dyadic Data."
Watch video: Srujana getting her award.
- IDEAL alumni Tumer and Agogino win Best Paper Award at AAMAS07

ECE alumni, Drs. Kagan Tumer and Adrian Agogino, received the Best Paper Award at the 2007 International Conference on Autonomous Agents and Multiagent Systems (AAMAS).
Drs. Tumer and Agogino, both PhDs from IDEAL, are researching ways to accommodate the anticipated tripling of air traffic over the next decade.
Read more
The current system routes flights on the basis of projections that do not take into account rapidly changing conditions in weather or airport congestion. Drs. Tumer and Agogino's approach is to create agents to govern specific air locations and use reinforcement learning to keep them informed of changing conditions. This approach shows a 67% reduction in congestion over current air traffic flow management.
- IDEAL members are runner-up for best research paper at Int'l Conf. on Data Mining (ICDM), Dec 2006.
- "Bregman Bubble Clustering: A Robust, Scalable Framework for Locating Multiple, Dense Regions in Data"
by Gunjan Gupta and Joydeep Ghosh
- Click here to download the latest version of Gene DIVER (Gene Density Interactive Visual ExplorER)
- a tool for clustering and visualizing gene datasets. By Gunjan Gupta
- Dr. Ghosh and Dr. Dhillon(CS) receive the 2005 Best Research Paper Award from the UT Co-op Society
-
for "Clustering with Bregman Divergence" (joint work with Arindam Banerjee and Srujana Merugu)
Award pictures with UT President Faulkner and with President and Deans.
Visitor Information
IDEAL internal resources
IDEAL meeting schedule and topics
- IDEAL member Meghana Deodhar is runner-up for Best Student Paper at ICML09
- "A Scalable Framework for Discovering Coherent Co-clusters in Noisy Data" by Meghana Deodhar, Joydeep Ghosh, Gunjan Gupta, Hyuk Cho and Inderjit Dhillon
- IDEAL member Alex Liu wins the Best Student Paper Award at DMIN07
- "Generative Oversampling for Mining Imbalanced Datasets" by Alex Liu, Joydeep Ghosh, and Cheryl Martin.
- IDEAL alumna Srujana Merugu wins the Best Research Paper Award at KDD07
Deepak Agarwal and Srujana Merugu at Yahoo! Research captured the Best Research Paper Award for their work on "Predictive Discrete Latent Factor Models For Large Scale Dyadic Data."
Watch video: Srujana getting her award.- IDEAL alumni Tumer and Agogino win Best Paper Award at AAMAS07

ECE alumni, Drs. Kagan Tumer and Adrian Agogino, received the Best Paper Award at the 2007 International Conference on Autonomous Agents and Multiagent Systems (AAMAS).
Drs. Tumer and Agogino, both PhDs from IDEAL, are researching ways to accommodate the anticipated tripling of air traffic over the next decade.
Read more
The current system routes flights on the basis of projections that do not take into account rapidly changing conditions in weather or airport congestion. Drs. Tumer and Agogino's approach is to create agents to govern specific air locations and use reinforcement learning to keep them informed of changing conditions. This approach shows a 67% reduction in congestion over current air traffic flow management.- IDEAL members are runner-up for best research paper at Int'l Conf. on Data Mining (ICDM), Dec 2006.
- "Bregman Bubble Clustering: A Robust, Scalable Framework for Locating Multiple, Dense Regions in Data"
by Gunjan Gupta and Joydeep Ghosh - Click here to download the latest version of Gene DIVER (Gene Density Interactive Visual ExplorER)
- a tool for clustering and visualizing gene datasets. By Gunjan Gupta
- Dr. Ghosh and Dr. Dhillon(CS) receive the 2005 Best Research Paper Award from the UT Co-op Society
-
for "Clustering with Bregman Divergence" (joint work with Arindam Banerjee and Srujana Merugu)
Award pictures with UT President Faulkner and with President and Deans.