04578nam 2200733 450 991078554900332120230125232906.01-283-89601-X1-60650-344-810.5643/9781606503447(CKB)2670000000234570(EBL)954641(OCoLC)830170654(SSID)ssj0000738957(PQKBManifestationID)12332618(PQKBTitleCode)TC0000738957(PQKBWorkID)10671549(PQKB)10345448(OCoLC)809681761(CaBNvSL)swl00401235(MiAaPQ)EBC954641(Au-PeEL)EBL954641(CaPaEBR)ebr10588195(CaONFJC)MIL420851(EXLCZ)99267000000023457020190118d2012 uy 0engur|n|---|||||txtccrQuality recognition and prediction smarter pattern technology with the Mahalanobis-Taguchi system /Shoichi Teshima, Yoshiko Hasegawa, Kazuo Tatebayashi1st ed.New York :Momentum Press, LLC,[2012]©20121 online resource (242 p.)Description based upon print version of record.1-60650-342-1 Includes bibliographical references and index.Foreword -- Preface -- Acknowledgments --1. Pattern recognition and the MT system -- 1.1 Overview of pattern recognition and the fields of application -- 1.2 Standard execution procedure for pattern recognition -- 1.3 Fields with substantial experience in the use of MT system applications --2. Merits of the MT system and its computation methods -- 2.1 Characteristics shared by all MT system components -- 2.2 Features of the MT method -- 2.3 Features of the T method -- 2.4 The MT system computation formulas --3. Data handled by the MT system and feature extraction -- 3.1 Use of measured values in an unmodified form -- 3.2 Performing feature extraction -- 3.3 Feature extraction technique from character pattern -- 3.4 Feature extraction technique from waveform pattern -- 3.5 Differences between other waveform features and variation values/abundance values --4. MT method application procedure and important points to heed -- 4.1 Example of character recognition -- 4.2 Example of weather prediction --5. T method application procedures and key points -- 5.1 Yield prediction for manufacturing-production using T method-1 -- 5.2 Character pattern recognition using the RT method --6. Examples of actual applications -- 6.1 Blade wear monitoring via cutting vibration waveform (MT method) -- 6.2 Appearance inspection of a clutch disk -- 6.3 Monitoring of machine conditions (MT method) -- 6.4 Application to medical diagnosis (MT method) -- 6.5 Strength estimation based on raw material mixing (T method-1) -- 6.6. Real estate price prediction by T method-1 --Appendices -- A. Differences between the MT system and artificial intelligence -- B. Difference between the MT system and traditional statistical theory -- C. Supplementary considerations concerning mathematical formulas -- D. Strategy to use when data incorporates unmeasured values -- E. Fusion with artificial intelligence and other resources -- F. Mahalanobis distance computation using Microsoft Excel -- G. Paley's construct for generation of Hadamard matrice --Bibliography and reference sources -- Bibliography (in English) -- Bibliography (in Japanese) -- References -- Glossary: definition of terms -- Index -- About the authors.The MT system is a diagnostic and predictive method for analyzing patterns in multivariate data that has provided benefits in many diverse applications over the past decade or so. It has proven itself superior in many cases to more traditional artificial intelligence applications such as neural nets.Pattern recognition systemsTaguchi methods (Quality control)Pattern recognition systems.Taguchi methods (Quality control)621.3819598Teshima Shoichi1507051Hasegawa YoshikoTatebayashi KazuoMiAaPQMiAaPQMiAaPQBOOK9910785549003321Quality recognition and prediction3737513UNINA