Distributed cooperative mining for information consortia

ABSTRACT

We consider the situation where a number of agents are distributed and each of them collects a data sequence generated according to an unknown probability distribution. Here each of the distributions is specified by common parameters and individual parameters e.g., a normal distribution with an identical mean and a different variance. Here we introduce a notion of an information consortium, which is a framework where the agents cannot show raw data to one another, but they like to enjoy significant information gain for estimating the respective distributions. Such an information consortium has recently received much interest in a broad range of areas including financial risk management, ubiquitous network mining, etc. In this paper we are concerned with the following three issues: 1) how to design a collaborative strategy for agents to estimate the respective distributions in the information consortium, 2) characterizing when each agent has a benefit in terms of information gain for estimating its distribution or information loss for predicting future data, and 3) charracterizing how much benefit each agent obtains. In this paper we yield a statistical formulation of information consortia and solve all of the above three problems for a general form of probability distributions. Specifically we propose a basic strategy for cooperative estimation and derive a necessary and sufficient condition for each agent to have a significant benefit.

REFERENCES

• 1: P. Chan and S. Stolfo, "Toward Parallel and Distributed Learning By Meta-Learning", In Working Notes AAAI Work. Knowledge Discovery in Databases, pp. 227--240, AAAI, 1993.
• 2: R. Chen , K. Sivakumar , H. Kargupta, Collective Mining of Bayesian Networks from Distributed Heterogeneous Data, Knowledge and Information Systems, v.6 n.2, p.164-187, March 2004 [doi>10.1007/s10115-003-0107-8]
• 3: David W. Cheung , Vincent T. Ng , Ada W. Fu , Yongjian Fu, Efficient Mining of Association Rules in Distributed Databases, IEEE Transactions on Knowledge and Data Engineering, v.8 n.6, p.911-922, December 1996 [doi>10.1109/69.553158]
• 4: David W. Cheung , Jiawei Han , Vincent T. Ng , Ada W. Fu , Yongjian Fu, A fast distributed algorithm for mining association rules, Proceedings of the fourth international conference on on Parallel and distributed information systems, p.31-43, December 18-20, 1996, Miami Beach, Florida, United States
• 5: T. Han and S. Amari, "Statistical Inference Under Multiterminal Data Compression", IEEE Tans. on Information Theory, Vol. 44, No. 6, pp. 2300--2324, (1998).
• 6: Hillol Kargupta , Anupam Joshi, Data mining "to go": ubiquitous KDD for mobile and distributed environments, Tutorial notes of the seventh ACM SIGKDD international conference on Knowledge discovery and data mining, August 26-29, 2001, San Francisco, California [doi>10.1145/502786.502791]
• 7: H. Kargupta, B. Park, D. Hershbereger, and E. Johnson, "Collective data mining: A new perspective toward distributed data mining", Advances in Distributed Data Mining, AAAI/MIT Press, (1999).
• 8: Michael Kearns , H. Sebastian Seung, Learning from a population of hypotheses, Proceedings of the sixth annual conference on Computational learning theory, p.101-110, July 26-28, 1993, Santa Cruz, California, United States [doi>10.1145/168304.168317]
• 9: Atsuyoshi Nakamura , Jun-ichi Takeuchi , Naoki Abe, Efficient distribution-free population learning of simple concepts, Annals of Mathematics and Artificial Intelligence, v.23 n.1-2, p.53-82, 1998 [doi>10.1023/A:1018908122958]
• 10: NetRisk (R. Ceske and L. Swann), "Share and Share Alike", http://www.netrisk.com/downloads/publishedarticles/shareandsharealike.PDF, (1999).
• 11: A. Prodromidis, P. Chan, and S. Stolfo, "Meta-learning in distributed data mining systems: Issues and approaches", In Advances in Distributed and Parallel Knowledge Discovery, H. Kargupta and P. Chan (editors), Chapter 3, AAAI/MIT Press, (2000).
• 12: Kenji Yamanishi, Distributed cooperative Bayesian Learning strategies, Information and Computation, v.150 n.1, p.22-56, April 10, 1999 [doi>10.1006/inco.1998.2753]

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.2 DATABASE MANAGEMENT
      H.2.8 Database applications

Keywords:
collective mining, distributed mining, estimation, information consortium, statistical model