Abstract
Discovering the Bayesian network (BN) structure from big datasets containing rich causal relationships is becoming increasingly valuable for modeling and reasoning under uncertainties in many areas with big data gathered from sensors due to high volume and fast veracity. Most of the current BN structure learning algorithms have shortcomings facing big data. First, learning a BN structure from the entire big dataset is an expensive task which often ends in failure due to memory constraints. Second, it is quite difficult to select a learner from numerous BN structure learning algorithms to consistently achieve good learning accuracy. Lastly, there is a lack of an intelligent method that merges separately learned BN structures into a well structured BN network. To address these shortcomings, we introduce a novel parallel learning approach called PEnBayes (Parallel Ensemble-based Bayesian network learning). PEnBayes starts with an adaptive data preprocessing phase that calculates the Appropriate Learning Size and intelligently divides a big dataset for fast distributed local structure learning. Then, PEnBayes learns a collection of local BN Structures in parallel using a two-layered weighted adjacent matrix-based structure ensemble method. Lastly, PEnBayes merges the local BN Structures into a global network structure using the structure ensemble method at the global layer. For the experiment, we generate big data sets by simulating sensor data from patient monitoring, transportation, and disease diagnosis domains. The Experimental results show that PEnBayes achieves a significantly improved execution performance with more consistent and stable results compared with three baseline learning algorithms.
Highlights
A Bayesian network (BN) BBN is a probabilistic graphical model that represents a probability distribution through a directed acyclic graph (DAG) that encodes conditional dependency and independency relationships among variables in the model
We significantly extend our previous work [8,40] in adopting different BN structure learning algorithms in the Local Learner and design a three-layered ensemble approach to ensure learning stability and accuracy
Our goal is to use the big training data to learn an accurate model of the underlying distribution at both data level and algorithm level to achieve better learning accuracy, stability, and usability towards integrating Bayesian network learning as part of the big data modeling and scientific workflow engine
Summary
A Bayesian network (BN) BBN is a probabilistic graphical model that represents a probability distribution through a directed acyclic graph (DAG) that encodes conditional dependency and independency relationships among variables in the model. One solution is to perform the learning task in a distributed data processing. Distributed Data-Parallel Patterns and Supporting Systems for Scalable Big Data Application. DDP patterns enable programs to execute in parallel by splitting data in distributed computing environments. DDP pattern executes user-defined functions (UDF) in parallel over input datasets. Users only need to select the appropriate DDP pattern for their specific data processing tasks, and implement the corresponding UDFs. Due to the increasing popularity and adoption of these DDP patterns, a number of execution engines have been implemented to support one or more of them. Due to the increasing popularity and adoption of these DDP patterns, a number of execution engines have been implemented to support one or more of them These DDP execution engines manage distributed resources, and execute UDF instances in parallel. When running on distributed resources, DDP engines can achieve good scalability and performance acceleration with good fault tolerance
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have