Collective mining of Bayesian networks from distributed heterogeneous data

Abstract

Abstract. We present a collective approach to learning a Bayesian network from distributed heterogenous data. In this approach, we first learn a local Bayesian network at each site using the local data. Then each site identifies the observations that are most likely to be evidence of coupling between local and non-local variables and transmits a subset of these observations to a central site. Another Bayesian network is learnt at the central site using the data transmitted from the local site. The local and central Bayesian networks are combined to obtain a collective Bayesian network, that models the entire data. Experimental results and theoretical justification that demonstrate the feasibility of our approach are presented.

Examining the multifactorial nature of a cognitive process using Bayesian brain-behavior modeling

“…The key component of BBM is a BN-induction algorithm; such algorithms scale well with the number of samples, and have been shown to scale to millions of samples 17, 22 . The BNs in Figure 3 include a model with a small number of variables (Figure 3 (a)), one with medium complexity (Figure 3 (b) and Figure 3 (c)), and a model with high complexity (Figure 3 (d)).…”

Privacy preserving sub-feature selection in distributed data mining

“…where the participating parties get exact solution based on combined database [7]. For several computational experiments, the different database are considered to have single solution based on the data collection from different parties using different techniques such as standard algorithm for decision tree [8] in peer-to-peer systems, identification of TOP-l inner products elements in P2P network [9], collective mining of Bayesian networks [10], mining criminal networks [11], incentive compatible privacy preserving of data in distributed classification [12], etc. The techniques of above paper indicate the importance of the role of participating parties in distributed environment where each party never wants to release their private data without protection.…”

Learning distributed bayesian network structure using majority-based method

“…In this case, every distributed node has the same feature but different observations. A collective approach to implement a distributed Bayesian network has been proposed by Chen et al [3]. But this approach is implemented in a client-server framework for heterogeneous databases.…”