Search results

The purpose of this study is to alleviate the specified issues to a great extent. To promote patients’ health via early prediction of diseases, knowledge extraction using data mining approaches shows an integral part of e-health system. However, medical databases are highly imbalanced, voluminous, conflicting and complex in nature, and these can lead to erroneous diagnosis of diseases (i.e. detecting class-values of diseases). In literature, numerous standard disease decision support system (DDSS) have been proposed, but most of them are disease specific. Also, they usually suffer from several drawbacks like lack of understandability, incapability of operating rare cases, inefficiency in making quick and correct decision, etc.

Addressing the limitations of the existing systems, the present research introduces a two-step framework for designing a DDSS, in which the first step (data-level optimization) deals in identifying an optimal data-partition (Popt) for each disease data set and then the best training set for Popt in parallel manner. On the other hand, the second step explores a generic predictive model (integrating C4.5 and PRISM learners) over the discovered information for effective diagnosis of disease. The designed model is a generic one (i.e. not disease specific).

The empirical results (in terms of top three measures, namely, accuracy, true positive rate and false positive rate) obtained over 14 benchmark medical data sets (collected from https://archive.ics.uci.edu/ml) demonstrate that the hybrid model outperforms the base learners in almost all cases for initial diagnosis of the diseases. After all, the proposed DDSS may work as an e-doctor to detect diseases.

The model designed in this study is original, and the necessary parallelized methods are implemented in C on Cluster HPC machine (FUJITSU) with total 256 cores (under one Master node).

To provide administrators at an Australian university with data on the feasibility of redirecting under‐utilised computer laboratories facilities into a distributed high performance computing facility.

The individual log‐in records for each computer located in the computer laboratories at the university were investigated. The log‐in data were investigated over a 24‐hour/seven day a week period between June 2001 and August 2003. The data were analysed in terms of student access to the computer facilities during “normal” business hours, weekend times, and the semester breaks.

The computer laboratories were hugely under‐utilised, with less than 10 per cent of all log‐ins occurring during off peak times (7 pm‐8 am). Similarly, only weekends were likewise under‐utilised. This strongly suggests that this spare computer capacity could be used for alternate means during these times.

Future research needs to determine whether the needs of the general computer laboratory user who requires a stable and secure system can coexist with the users of a high performance computer facility where different software and differently configured computer systems are required.

This research has the potential for universities to utilise more effectively their computer laboratory resources by allocating under‐utilised resources into other projects, such as to a high performance computing facility (HPCF). The cost of these re‐allocated resources would be a fraction of the cost compared to a scenario in which a separate dedicated HPCF had to be provided.

This paper suggests an alternate utilisation of the spare computing laboratory resources available at many universities.

The motivation for this research work is the need for an efficient software tool for inductance calculation of components in flexible electronics. A software package PROVOD has been developed and it has produced very accurate results but the applied numerical method can lead to a huge amount of calculations. The aim of this research is to apply the parallel computing to this specific computational technique and to investigate the impact of increasing the number of parallel executing threads.

The largest possible amount of operations is put in parallel using the fact that the inductance between two segments is a sum of independent elements. OpenMP and Microsoft's Concurrency Runtime have been tested as parallel programming techniques.

Parallel computing with a different number of threads (up to 24) has been tested with OpenMP. A significant increase in computational speed (up to 21 times) has been obtained.

The research is limited by the available number of parallel processors.

Accurate and fast inductance calculation for flexible electronic components is possible to achieve. The impact of parallel processing is proven.

The proposed method of calculation acceleration of inductances can be helpful in the design and optimization of new flexible devices in electronics.

Parallel computing is applied to the design of flexible electronic components. It is shown that a large number of parallel processors can be efficiently used in this type of calculation. The obtained results are interesting for people involved in the design of flexible components, and generally, for researchers/engineers dealing with similar electromagnetic problems.