350 rub
Journal Radioengineering №2 for 2014 г.
Article in number:
Relevance vector method in pattern recognition signals from seismic detection system
Keywords: classification relevant vector machines (RVM) seismic detection systems
Authors:
L. I. Dvoyris - Dr.Sci. (Eng.), Professor
P. V. Potapov
Abstract:
The task of pattern recognition is one of the most important in the field of artificial intelligence. To this day the solution for each application remains a difficult theoretical and technical problem. In the last decade an alternative method of support vector machine was also intensively developed, and received a similar name - Relevance Vector Machine, RVM. This article deals with the problem of binary classification, the difficulties in solving it, and show the different methods of classification based on relevant vector machines. There was a comparison of the methods based on the data of signal processing of seismic means of detection recorded during the field experiment. The results obtained by the authors suggest that the RVM method has the best accuracy and reliability performance of detection signals in the time domain, and a small number of relevant vectors can greatly simplify the hardware implementation of a subsystem based on the current detection method.
Pages: 61-65
References

  1. Bishop C. Pattern Recognition and Machine Learning // Springer Science+Business Media. LLC. 2006.
  2. Rasmussen C.E., Williams C.K.I. Gaussian Processes for Machine Learning. Massachusetts Institute of Technology. MIT Press. 2006. www.GaussianProcess.org/gpml
  3. Bishop C., Tipping M. Variational relevance vector machine. In UAI. 2000.
  4. Bishop C., Tipping M. Bayesian Regression and Classification // Advances in Learning Theory: Methods, Models and Applications. IOS Press. NATO Science Series III: Computer and Systems Sciences. 2003. V. 190.
  5. Tipping M., Faul A. Fast Marginal Likelihood Maximisation for Sparse Bayesian Models // Proceedings of the Ninth International Workshop on Artifical Intelligence and Statistics. Key West, FL. Jan 3-6. 2006.
  6. Jaakkola T., Jordan M. Bayesian parameter estimation through variational methods // Statistics and Computing. 2000. V. 10. P. 25-37.
  7. Dvojris L.I., Gerashhenkov V.A. Oczenka tochnosti i nadezhnosti raspoznavaniya ob''ektov po signalam sejsmicheskix obnaruzhitelej metodom SVM vo vremennoj i chastotnoj oblastyax // Radiotexnika (Zhurnal v zhurnale). 2012. № 1. S. 12-16.