An Autoassociator for on-Line Motor Monitoring

Petsche, Thomas; Junior, Angelo Marcantonio; Darken, Christian J.; Hanson, Stephen J.; Kuhn, G.; Santoso, Iwan

doi:10.1142/9789812816955_0012

Cited by 4 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Neural networks have been successfully applied to a variety of real world classification tasks in industry, business and science [186]. Applications include bankruptcy prediction [2], [96], [101], [167], [187], [195], handwriting recognition [61], [92], [98], [100], [113], speech recognition [25], [106], product inspection [97], [130], fault detection [11], [80], medical diagnosis [19], [20], [30], [31], and bond rating [44], [163], [174]. A number of performance comparisons between neural and conventional classifiers have been made by many studies [36], [82], [115].…”

Section: Introductionmentioning

confidence: 99%

Neural networks for classification: a survey

Zhang

2000

IEEE Trans. Syst., Man, Cybern. C

1,449

568

View full text Add to dashboard Cite

Classification is one of the most active research and application areas of neural networks. The literature is vast and growing. This paper summarizes the some of the most important developments in neural network classification research. Specifically, the issues of posterior probability estimation, the link between neural and conventional classifiers, learning and generalization tradeoff in classification, the feature variable selection, as well as the effect of misclassification costs are examined. Our purpose is to provide a synthesis of the published research in this area and stimulate further research interests and efforts in the identified topics.

show abstract

Section: Introductionmentioning

confidence: 99%

Neural networks for classification: a survey

Zhang

2000

IEEE Trans. Syst., Man, Cybern. C

1,449

568

View full text Add to dashboard Cite

show abstract

“…This would optimize the task at hand. However studies have confirmed that back propagation is prone to the following problems -it may get stuck at a local optimum [22] and it may take a very long time to converge [23]. This led researchers to attack the ANN training problem with other methods.…”

Section: [3] [4] [5] [6] [7] [8]mentioning

confidence: 99%

Improving Fault Prediction using ANN-PSO in Object Oriented Systems

Kayarvizhy¹,

Kanmani²,

Uthariaraj³

2013

IJCA

View full text Add to dashboard Cite

Object oriented software metrics are computed and used in predicting software quality attributes of object oriented systems. Mapping software metrics to software quality attributes like fault prediction is a complex process and requires extensive computations. Many models have been proposed for fault prediction. Since accuracy is of prime importance in prediction models they are being constantly improved through various research studies. Artificial Neural network (ANN) has gained immense popularity due to its adaptability to the problem at hand by training with known data. Back propagation is a widely used ANN training technique. However the back propagation technique leads to slow convergence rate and an impending threat of getting caught in local minima. In this paper we explore the Particle Swarm Optimization (PSO) technique as an alternative to optimize the weights of ANN for fault prediction in object oriented systems. We evaluate the effect on prediction accuracy that PSO brings to ANN compared to other techniques like BP and Genetic Algorithm (GA). We also evaluate prediction accuracy improvements by optimizing the various parameters of PSO.

show abstract