An Improved KNN Algorithm of Intelligent Built-in Test

Aimed at the faults of K-nearest neighbor (KNN) algorithm in complex equipment's built-in test (BIT), an improved KNN (IKNN) algorithm is proposed to solve the problem from two aspects. Firstly, the weight of each input feature is learned using neural network to make important features contribute more in the classifications; this improves the precision of classification. Secondly, clustering each sample of the training set to reduce the data volume of training set, this improves the running speed of the algorithm. Simulation experiments prove the effectiveness of the IKNN algorithm with higher precision and less calculation.

[1]  Jin Zhang,et al.  An Evaluation of Engine Faults Diagnostics Using Artificial Neural Networks , 2001 .

[2]  J. M. Anderson,et al.  The enemy is FA, CND, and RTOK (avionics testing) , 1988, AUTOTESTCON '88. Symposium Proceedings IEEE International Automatic Testing Conference, Futuretest..

[3]  Guy Denney,et al.  F16 jet engine trending and diagnostics with neural networks , 1993, Defense, Security, and Sensing.

[4]  David W. Aha,et al.  A Review and Empirical Evaluation of Feature Weighting Methods for a Class of Lazy Learning Algorithms , 1997, Artificial Intelligence Review.

[5]  Andreas Steininger,et al.  Testing and built-in self-test - A survey , 2000, J. Syst. Archit..

[6]  Sudipto Guha,et al.  CURE: an efficient clustering algorithm for large databases , 1998, SIGMOD '98.

[7]  Sang-Chan Park,et al.  A hybrid approach of neural network and memory-based learning to data mining , 2000, IEEE Trans. Neural Networks Learn. Syst..

[8]  Huan Liu,et al.  Neural-network feature selector , 1997, IEEE Trans. Neural Networks.

[9]  Thierry Denoeux,et al.  An evidence-theoretic k-NN rule with parameter optimization , 1998, IEEE Trans. Syst. Man Cybern. Part C.

[10]  Ludmila I. Kuncheva,et al.  Fitness functions in editing k-NN reference set by genetic algorithms , 1997, Pattern Recognit..