CoMMA- Combining Multi-relation Multimedia for Associations

Ankur Teredesai, Juveria Kanodia, Muhammad Ahmad

This eventual goal of the project is to do Association Rule Mining on Multimedia Data across multiple table, particularly pictures and text. Although ample literature exists on mining Multi-relational data mining, however mining multi-modal data across multiple tables has hardly been addressed. The goal of the project is to address this problem and employ the current application as a test bed. Our picture corpus is a group of 798 pictures we found on the web, they are pictures of different kinds such as art, landscape, astronomy etc. The text corpus is the corresponding picture descriptions. We generate association rules on image data (the RGBY values), and on text data separately. Then, we propose an algorithm to link these two different domains together. Our goal is to be able to return words that will describe a given unknown picture.

Help us annotate

CoMMA paper (version: submitted to MDMKDD. request: amt At c s. r i t. e d u)

Runs:

	Preprocessing
	Images.zip
	Documentation.zip

Presentations:

	CoMMA
	CoMMA Update
	AutoAnnotate
	on 2/19 (final)
	on 2/10

References:

	Qiang Ding et.al. Integrating Query Processing and DataMining in Relational RDBMSs.
	William Perrizo et. al. MultiMedia Datamining using Ptrees.
	S. Dzeroski. 2003. Multi-Relational Data Mining: An Introduction.
	P. Domingos. 2003. Prospects and Challenges for Multi-Relational Data Mining.
	B. Goethals. Survey on Frequent Pattern Mining.
	R. Agarwal, C. Aggarwal, and V.V.V. Prasad. A Tree Projection algorithm for generation for frequent itemsets. In J. Parallel and Distributed Computing, 2000.
	Jochen Hipp, Ulrich Güntzer, Gholamreza Nakhaeizadeh. 2000. Algorithms for Association Rule Mining - A General Survey and Comparison. SIGKDD Explorations.
	Rakesh Agrawal, Ramakrishnan Srikant. 1994. Fast Algorithms for Mining Association Rules in Large Databases. In Proc. of 20th Int. Conf. on Very Large Data Bases.
	Alex Freitas. 2000. Understanding the Crucial Differences Between Classification and Discovery of Association Rules - A Position Paper. SIGKDD Explorations.
	Bing Liu. 1998. Integrating Classification and Association Rule Mining. Proceedings of the KDD, New York.
	William Perrizo, Qin Ding, Qiang Ding, Amalendu Roy. On Mining Satellite and Other Remotely Sensed Images
	Qin Ding, Maleq Khan, Amalendu Roy, William Perrizo,. The P - tree Algebra Proceedings of the 2002 ACM symposium on Applied Computing.
	P. Kotala, A. Perera, J Kai Zhou, S. Mudivarthy, W. Perrizo, E. Deckard Data Mining of Micro-array Data using Association Rule Mining
	E.H. Han, G. Karypis, and V. Kumar. Min-apriori: An algorithm for finding association rules in data with continuous attributes. Technical Report TR-97-068, Department of Computer Science, University of Minnesota,Minneapolis, 1997.
	Hector Ariel Leiva MRDTL: A multi-relational decision tree learning algorithm
	Jan Struyf and Hendrik Blockeel. Efficient Multi-Relational Data Mining
	Nico Brandt and Peter Brockhausen and Marc de Haas and Jorg-Uwe Kietz and et al. Mining Multi-Relational Data
	V. Crestana-Jensen and N. Soparkar. Frequent Itemset Counting across Multiple Tables. In Proceedings of 4th Pacific-Asia Conference on Knowledge Discovery and Data Mining, 2000.
	Y. Fu and J. Han. Meta-rule-guided mining of association rules in relational databases. In Proc. 1st Int'l Workshop on Integration of Knowledge Discovery with Deductive and Object-Oriented Databases (KDOOD'95), pages 39--46, Singapore, Dec. 1995.
	Luc Dehaspe and Hannu Toivonen Discovery of Relational Association Rules. Relational Data Mining, Springer-Verlag. ed. Saso Dzeroski and Nada Lavrac, 2001
	Dehaspe, L., and De Raedt, L. 1997. Mining association rules in multiple relations. In Proceedings of the 7th International Workshop on Inductive Logic Programming, volume 1297 of Lecture Notes in Artificial Intelligence, 125--132. Springer-Verlag.
	Peter Brockhausen and Marc de Haas and Jorg-Uwe Kietz and Arno Knobbe and Olaf Rem and Regina Zucker and Nico Brandt (2001). Mining Multi-Relational Data. Technical report, IST Project MiningMart, IST-11993