A customizable multi-agent system for distributed data mining

We present a general Multi-Agent System framework for distributed data mining based on a Peer-to-Peer model. Agent protocols are implemented through message-based asynchronous communication. The framework adopts a dynamic load balancing policy that is particularly suitable for irregular search algorithms. A modular design allows a separation of the general-purpose system protocols and software components from the specific data mining algorithm. The experimental evaluation has been carried out on a parallel frequent subgraph mining algorithm, which has shown good scalability performances.

Symmetric Push-Sum Protocol for decentralised aggregation

Gossip (or Epidemic) protocols have emerged as a communication and computation paradigm for large-scale networked systems. These protocols are based on randomised communication, which provides probabilistic guarantees on convergence speed and accuracy. They also provide robustness, scalability, computational and communication efficiency and high stability under disruption. This work presents a novel Gossip protocol named Symmetric Push-Sum Protocol for the computation of global aggregates (e.g., average) in decentralised and asynchronous systems. The proposed approach combines the simplicity of the push-based approach and the efficiency of the push-pull schemes. The push-pull schemes cannot be directly employed in asynchronous systems as they require synchronous paired communication operations to guarantee their accuracy. Although push schemes guarantee accuracy even with asynchronous communication, they suffer from a slower and unstable convergence. Symmetric Push- Sum Protocol does not require synchronous communication and achieves a convergence speed similar to the push-pull schemes, while keeping the accuracy stability of the push scheme. In the experimental analysis, we focus on computing the global average as an important class of node aggregation problems. The results have confirmed that the proposed method inherits the advantages of both other schemes and outperforms well-known state of the art protocols for decentralized Gossip-based aggregation.

Fault tolerant decentralised K-Means clustering for asynchronous large-scale networks

The K-Means algorithm for cluster analysis is one of the most influential and popular data mining methods. Its straightforward parallel formulation is well suited for distributed memory systems with reliable interconnection networks, such as massively parallel processors and clusters of workstations. However, in large-scale geographically distributed systems the straightforward parallel algorithm can be rendered useless by a single communication failure or high latency in communication paths. The lack of scalable and fault tolerant global communication and synchronisation methods in large-scale systems has hindered the adoption of the K-Means algorithm for applications in large networked systems such as wireless sensor networks, peer-to-peer systems and mobile ad hoc networks. This work proposes a fully distributed K-Means algorithm (EpidemicK-Means) which does not require global communication and is intrinsically fault tolerant. The proposed distributed K-Means algorithm provides a clustering solution which can approximate the solution of an ideal centralised algorithm over the aggregated data as closely as desired. A comparative performance analysis is carried out against the state of the art sampling methods and shows that the proposed method overcomes the limitations of the sampling-based approaches for skewed clusters distributions. The experimental analysis confirms that the proposed algorithm is very accurate and fault tolerant under unreliable network conditions (message loss and node failures) and is suitable for asynchronous networks of very large and extreme scale.