|
|
Address:
|
Contact
Details: |
IAN DAVIDSON
|
|
Address:
|
Contact
Details: |
Hello, I'm an Associate Professor in the computer science department working on mainly learning and data mining algorithm development.
Research Areas
I have several research projects that I work on, but the common connection is novel problems, rigorous formulations and practical applications. I typically work with other faculty/researchers to apply my work so if you can see some potential application then please let me email me.
I have several research assistantship slots available for students who are motivated to do research in data mining. My research is a mix of rigorous algorithm design, implementation and application to novel areas of social importance. Please read some of my work and contact me if your genuinely interested. Do NOT send form letters.
I bought the constrained-clustering.org domain name and its being set up as a central repository for things related to clustering with constraints. So far it only contains my work but it will be expanded. Feel free to email me for suggestions.
Current Grants and Projects
Thanks to the NSF (CAREER Award Project Webpage), Office of Naval Research, DoD and Google (Research Award) for supporting my research projects and my wife for these two projects.
Constrained Clustering
In this work we are exploring adding constraints and background knowledge into clustering in the form of constraints. We begun with complexity results, then moved onto more complexity results and hierarchial algorithms and then spectral clustering. Lately, we have explored encoding more complex constraints in SAT solvers for hierarchies and non-hierarchical clustering.
Behavioral Activity Analysis Using Network Analysis and Tensor Decomposition
Here we model complex activities in a geo-spatial region as both complex networks and tensors for analysis.
Transfer Learning
We explore semi-supservised and unsupervised models of transfer learning where the mechanism for transfer is explicitly modeled and we can examine it to determine sufficient conditions when transfer is successful. We apply this work to fMRI data.
Building Idealized Hierarchies
We show how hierarchical algorithms can be modified to accept guidance, particulary in in the form of an idealized hierarchy.
Current Service
Editorial Boards: IEEE Transactions of Knowledge Discovery and Data Mining
Knowledge Discovery and Data Mining
Knowledge and Information Systems
Conferences: PC Chair – SIAM Data Mining Conference 2012
Vice/area Chair – IEEE ICDM 2011, ACM KDD 2012
Recent Papers
KDD 2011 Paper on Constrained Hierarchies PDF
KDD 2011 Paper on Multi source learning PDF
KDD 2010 Paper on Constrained Spectral Clustering PDF
KDD 2010 Paper on Disparate Clustering PDF
ICDM 2010 Paper on Contextual Outliers PDF
ICDM 2009 Paper on Active Spectral Clustering PDF
AAAI 2010 Paper on Multi-label, Semi-Supervised Learning PDF
SDM 2010 Paper on SAT Solvers and Clustering PDF
New DMKD Journal Paper on Constraints and Hierarchical Clustering PDF
The Book has been sent to the publishers, check it out! Click here for book details, Buy from Amazon
New IJCAI Paper on Dimension Reduction Using Weighted Graphs PDF
New KDD Paper on Finding Alternative Clusterings PDF
New ICDM Paper on Finding Alternative Clusterings PDF
New Journal of DMKD Paper on Constraints and Agglomerative clustering: PDF
The Book has been sent to the publishers, check it out! Click here for book details, Buy from Amazon
Older Papers
Constrained Clustering (In collaboration with S. Basu (SRI), S.S. Ravi (SUNY) and K. Wagstaff (NASA-JPL))
Davidson I. Ester M. and Ravi, S.S., Efficient Incremental Clustering with Constraints, 13th ACM Knowledge Discovery and Data Mining Conference 2007 PDF
Ge, R., Ester, M., Jin W., and Davidson I., Constraint Driven Clustering, 13th ACM Knowledge Discovery and Data Mining Conference 2007
Davidson I., Ravi, S.S., Intractability and Clustering with Constraints, To Appear in the Proceeding of ICML 2007 PDF
Davidson I., Basu S.S., Clustering with Constraints Bibliography, To Appear, PDF
Davidson
I., Wagstaff, K., Basu, Sugato., Measuring Constraint-Set Utility for
Partitional Clustering Algorithms , To Appear in the Proceeding of
ECML/PKDD 2006 (acceptance rate 9%) PDF,
Davidson I., Ravi S.S., Identifying and Generating Easy Sets of
Constraints For Clustering, To
Appear 21st
AAAI
Conference, 2006.
(acceptance rate 21%) PDF
Davidson I. and Ravi, S. S. Hierarchical Clustering with Constraints: Theory and Practice, 9th European Principles and Practice of KDD, PKDD 2005. (acceptance rate 11%) (Email me for Journal/TR version) PDF Extended technical report with all proofs PDF
Davidson I., Ravi, S.S., Clustering under Constraints: Feasibility Issues and the $K$-Means Algorithm, 5th SIAM Data Mining Conference, (acceptance rate 14%). Winner, Best Paper Award, PDF,
Learning With Biased Training Sets (In collaboration with Wei Fan (IBM Watson))
Wei Fan and Ian Davidson "On Sample Selection Bias and Its Efficient Correction via Model Averaging and Unlabeled Examples", SIAM Data Mining Conference 2007 (acceptance rate 9%).
Kun
Zhang, Wei Fan, Xiaojing Yuan, Ian Davidson, and Xiangshang Li,
"Forecasting Skewed Biased Stochastic Ozone Days", IEEE
ICDM 2006 (acceptance rate 9%). ,
Winner
Best Paper Award, PDF,
Davidson I., Fan, Wei., When Efficient Model Averaging
Out-Performs Boosting and Bagging , To Appear in the Proceeding of
ECML/PKDD 2006 (acceptance rate 18%) PDF,
Fan, W. and Davidson I., ReverseTesting: An Efficient Framework
to Select Amongst Classifiers under Sample Selection Bias, To Appear
12th
ACM
KDD Conference,
Philadelphia, 2006. (acceptance rate 11%) PDF
Fan W., Davidson I., Zadrozny B. and Yu P., An Improved Categorization of Classifier's Sensitivity on Sample Selection Bias, 5th IEEE International Conference on Data Mining, ICDM 2005. (acceptance rate 18%) PDF
Applications
Davidson
I., Paul, Goutam, Locating Secret Messages in Images, Research Track:
10th
ACM KDD Conference 2004: Research Track,
Seattle (acceptance rate 29%) PDF
Davidson I. et al, A General Approach to Incorporate Data
Quality Matrices into Data Mining Algorithms, 10th
ACM KDD Conference 2004: Industrial Track,
2004. (Email me for Journal/TR version). Seattle. PDF
Yin, K. and Davidson I., Visually Comparing Clustering Algorithms, 8th PKDD Conference 2004 (acceptance rate 27%), Sydney PDF
Berg G., Davidson I., Paul G., Duan M., Searching For Hidden Messages: Automatic Detection of Steganography, AAAI - IAAI 2003 Pdf
Davidson, I., "Visualizing Clustering Results", SIAM International Conference on Data Mining, 2002 Color PDF
Foundations of Mining and Learning Using Information Theory, Probability and Statistics
Davidson I., Ensemble Approaches for Stable Learners with Convergence Bounds, 19th AAAI Conference 2004, San Jose (acceptance rate 26%) PDF.
Davidson, I., Yin, K., Message Length Estimators, Model Averaging and Optimal Prediction, Dimacs Workshop on Complexity and Inference (Yu, Hansen and Vitanyi), 2003. pdf
Davidson, I., and Aminian, M., Using The Central Limit Theorem for Belief Network Learning, Ian Davidson, Minoo Aminian, The 8th International Symposium on A.I. and Math pdf
Yin, K., Davidson I., Bayesian Model Averaging Across Model Spaces via Compact Encoding, The 8th International Symposium on A.I. and Math pdf
Davidson, I., "Minimum Message Length Clustering Using Gibbs Sampling", 16th International Conference on Uncertainty in A.I., 2000 PDF
In my spare time I enjoy history, am an avid movie buff, playing golf and have just begun learning the piano!
If you would like to read about me and my personal interests then click here.
Best research paper award at SIAM Data Mining Conference 2005
Best application paper award IEEE Data Mining Conference 2006.
Best paper nominations: ACM KDD 2007, 2011
I like working with undergraduates and graduate students, so email me if your interested in my research areas.
Xiang
Wang (2008-)
Leo Shamis (2008-)
Sean Gilpin (2009-)
Kristin
Liu (2008-)
Buyue Qian (2009-)
Constrained
Clustering: Advances in Algorithms, Applications and Theory,
In Preparation Due out 2008 co-edited with Sugato Basu and Kiri
Wagstaff. CRC Press. Click
here for book details, Buy
from Amazon
Knowledge
Discovery and Data Mining: Challenges and Realities of Mining Real
World Data with
co-edited with Hill Zhu, Ideal Press 2007. Click
here for ToC
IEEE ICDM 2005 Clustering with Constraints – Theory and Practice (with S.Basu SRI)
ACM KDD 2006 Clustering with Constraints – Theory and Practice (with S.Basu SRI) PDF of Slides
I mainly teach in the sub-fields of data mining and machine learning which are a very active area of research used extensively in the high tech industry though often under different titles. Data mining, speech recognition, natural language processing, machine learning and expert systems are all sub-fields of A.I. In addition applications of A.I. to specific domains are also popular such as bio-informatics, data mining and machine vision.
For courses at U.C. Davis please see myucdavis course web pages.
This page is always under construction
davidson at cs dot ucdavis dot edu
