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.

Classification and fault detection methods for fuel cell monitoring and quality control

In this paper, various types of fault detection methods for fuel cells are compared. For example, those that use a model based approach or a data driven approach or a combination of the two. The potential advantages and drawbacks of each method are discussed and comparisons between methods are made. In particular, classification algorithms are investigated, which separate a data set into classes or clusters based on some prior knowledge or measure of similarity. In particular, the application of classification methods to vectors of reconstructed currents by magnetic tomography or to vectors of magnetic field measurements directly is explored. Bases are simulated using the finite integration technique (FIT) and regularization techniques are employed to overcome ill-posedness. Fisher's linear discriminant is used to illustrate these concepts. Numerical experiments show that the ill-posedness of the magnetic tomography problem is a part of the classification problem on magnetic field measurements as well. This is independent of the particular working mode of the cell but influenced by the type of faulty behavior that is studied. The numerical results demonstrate the ill-posedness by the exponential decay behavior of the singular values for three examples of fault classes.

Pesticide-handling practices of smallholder coffee farmers in Eastern Jamaica

Pesticide use among smallholder coffee producers in Jamaica has been associated with significant occupational health effects. Research on pesticide handling practices, however, has been scarce, especially in eastern Jamaica. This explorative study aims at filling this gap and provides a first basis to develop effective interventions to promote a safer pesticide use. A random sample of 81 coffee farmers was surveyed. The majority of farmers reported to suffer from at least one health symptom associated with pesticide handling, but safety practices were scarcely adopted. There was also the risk that other household members and the wider local community are exposed to pesticides. The lack of training on pesticide management, the role of health services and the cost for protective equipment seemed to be the most significant factors that influence current pesticide handling practices in eastern Jamaica. Further research is recommended to develop a systemic understanding of farmer’s behaviour to provide a more solid basis for the development of future intervention programmes.

The Copenhagen principles on the handling of detainees: implications for the procedural regulation of internment

The Copenhagen Principles on the Handling of Detainees in International Military Operations were released in October 2012 after a five-year long process involving states and certain organizations. The Principles address a number of issues concerning the handling and transfer of detainees. They apply in military operations conducted by states abroad in the context of non-international armed conflicts and peace operations. This article focuses on those principles that address the procedural regulation of internment (ie preventive, security detention), as it is here that the current law is particularly unclear. On the one hand, the treaty provisions applicable in non-international armed conflicts contain no rules on the procedural regulation of internment, in comparison with the law of international armed conflict. On the other hand, the relevant rules under international human rights law (IHRL) appear derogable in such situations. This article demonstrates that the approach taken to this issue in the Copenhagen Principles is one which essentially draws on the procedural rules applicable to civilian internment in the international armed conflicts. These rules adopt standards that are lower than those under IHRL. Reference is then made to other recent practice, which illustrates that the Copenhagen Principles do not apply in a legal vacuum. In particular, two recent judicial developments highlight the continued relevance of human rights law and domestic law, respectively, in regulating detention operations in the context of international military operations. Compliance with the Copenhagen Principles may not, therefore, be sufficient for detention to be lawful.

‘It Was All My Fault’: negative interpretation bias in depressed adolescents

The extent to which cognitive models of development and maintenance of depression apply to adolescents is largely untested, despite the widespread application of Cognitive Behavior Therapy (CBT) for depressed adolescents. Cognitive models suggest that negative cognitions, including interpretation bias, play a role in etiology and maintenance of depression. Given that cognitive development is incomplete by the teenage years and that CBT is not superior to non-cognitive treatments in the treatment of adolescent depression, it is important to test the underlying model. The primary aim of this study was to test the hypothesis that interpretation biases are exhibited by depressed adolescents. Four groups of adolescents were recruited: clinically-referred depressed (n = 27), clinically-referred non-depressed (n = 24), community with elevated depression symptoms (n = 42) and healthy community (n = 150). Participants completed a 20 item ambiguous scenarios questionnaire. Clinically-referred depressed adolescents made significantly more negative interpretations and rated scenarios as less pleasant than all other groups. The results suggest that this element of the cognitive model of depression is applicable to adolescents. Other aspects of the model should be tested so that cognitive treatment can be modified or adapted if necessary.

Scalable and fault tolerant failure detection and consensus

Future extreme-scale high-performance computing systems will be required to work under frequent component failures. The MPI Forum's User Level Failure Mitigation proposal has introduced an operation, MPI_Comm_shrink, to synchronize the alive processes on the list of failed processes, so that applications can continue to execute even in the presence of failures by adopting algorithm-based fault tolerance techniques. This MPI_Comm_shrink operation requires a fault tolerant failure detection and consensus algorithm. This paper presents and compares two novel failure detection and consensus algorithms. The proposed algorithms are based on Gossip protocols and are inherently fault-tolerant and scalable. The proposed algorithms were implemented and tested using the Extreme-scale Simulator. The results show that in both algorithms the number of Gossip cycles to achieve global consensus scales logarithmically with system size. The second algorithm also shows better scalability in terms of memory and network bandwidth usage and a perfect synchronization in achieving global consensus.