A Novel Ordinal Regression Method with Minimum Class Variance Support Vector Machine
|
|
- Lee Fields
- 5 years ago
- Views:
Transcription
1 Intenatonal Confeence on Mateals Engneeng and Infomaton echnology Applcatons (MEIA 05) A ovel Odnal Regesson Method wth Mnmum Class Vaance Suppot Vecto Machne Jnong Hu,, a, Xaomng Wang and Zengx Huang School of Compute and Soft Engneeng, Xhua Unvesty, Chengdu 60039, Chna Key Laboatoy of Patten Recognton and Intellgent Infomaton Pocessng, Chengdu Unvesty, Chengdu 6006, Chna a dewh@hotmal.com Keywods: Machne leanng, Odnal egesson, Suppot vecto machne, Suppot vecto odnal egesson. Abstact. In the pape, we popose a novel odnal egesson method called mnmum class vaance suppot vecto odnal egesson (MCVSVOR). MCVSVOR s deved fom mnmum class vaance suppot vecto machne (MCVSVM) whch s a vaant of SVM, and so nhets the latte s chaactestcs such as takng the dstbuton of the categoes nto consdeaton and good genealzaton pefomance. Fnally, the expemental esults valdate the effectveness of MCVSVOR and ndcate ts supeo genealzaton pefomance ove SVOR.. Intoducton In the pactcal applcatons of machne leanng, a stuaton s fequently nvolved,.e. exhbtng an ode among the dffeent categoes. hs type of supevsed leanng poblems s efeed to as odnal egesson whch pedcts categoes of odnal scale [-4]. Dffeent fom tadtonal metc egesson poblems, ts gades ae usually dscete and fnte. Also, t dffes fom tadtonal classfcaton poblems n that thee s an odnal elatonshp among dffeent classes. In fact, odnal egesson shows esemblance to both egesson and classfcaton because labels ae dscete and odnal []. In the past decade, many methods have been poposed to deal wth the odnal egesson poblems [, 9, 3]. Suppot vecto odnal egesson (SVOR) s a poweful method whch s desgned to tackle the odnal egesson poblems and ognated n suppot vecto machne (SVM). Howeve, SVM s actually a local method n the sense that soluton s exclusvely detemned by suppot vectos wheeas all othe data ponts ae elevant to the decson hypeplane,.e., the SVM soluton does not take nto consdeaton the dstbuton of the categoes and may esult n a non-obust soluton [6]. In ode to ovecome the dawback of SVM, a modfed class of SVM called mnmum class vaance suppot vecto machne (MCVSVM) was pesented n [6]. hs method s nsped fom the optmzaton of Fshe s dscmnant ato [5]. Smla to SVM, MCVSVM mplements the lage magn pncple [5]. Howeve, unlke SVM, the soluton of MCVSVM takes nto consdeaton both the samples n the boundaes and the dstbuton of the categoes and gves a obust soluton. In ths pape, we popose a novel odnal egesson leanng method called mnmum class vaance odnal egesson (MCVSVOR) n whch the dstbuton of the categoes s explctly consdeed and the lage magn pncple s emboded. Followng the basc dea of SVOR, we defne the MCVSVOR optmzaton poblem. Snce MCVSVOR s deved fom MCVSVM, t nhets the latte s chaactestcs such as takng fully the dstbuton of the categoes nto consdeaton and embodyng the lage magn pncple. At the same tme, we also develop the lnea and nonlnea cases of MCVSVOR and analyze the elatonshp between MCVSVOR and SVR. he elatonshp shows that MCVSVOR can be solved usng the exstng SVOR softwae, whch makes the soluton easy to be computed. Fnally, the expemental esults ndcate that MCVSVOR s effectve and can get supeo genealzaton pefomance ove SVOR. 05. he authos - Publshed by Atlants Pess 894
2 . Related wok In ths pape, we consde an odnal egesson poblem wth odeed categoes whch ae denoted by consecutve nteges Y = {,, } to keep the known ank nfomaton. he tanng d dataset s epesented by D = {( x,y ) x R,y Y}, whee categoy, and y epesents the coespondng ank of the nput data pont dmensonalty of sample vecto. he dataset composed of x efes to the th sample n the -th = x. Hee d efes to the = sample ponts. Hee s the numbe of tanng samples n the -th categoy. And, we set X=[ x,, x, x,, x,, x,, x ]=[ x,, x ].. Suppot vecto odnal egesson he task of odnal egesson s to compute a functon f : R {,, } such that f( x ) = y [0, ]. Moeove, SVOR ams at fndng paallel dscmnant hypeplanes wx b = 0 ( =,, ) that sepaate the data ponts of dffeent anks. So, the followng optmzaton poblem s defned [3,4] mn ww+ C ( ξ + ξ ) wb,, ξ, ξ = = st.. wξ b + ξ, ξ 0,, wξ ξ, ξ 0,, b b,, b Whee =, and =,,. Hee, two auxlay vaables b 0 = and b = + ae ntoduced. ote, SVM mplements the lage magn pncple [4]. So, SVOR also embodes the pncple snce t s deved fom SVM.. Kenel dscmnant leanng fo odnal egesson Fo the above gven tanng dataset, the wthn-class scatte matx SW s defned as [5, ] S = ( x u )( x u ) () W = x X Whee X = { x,,, y = = }, denotes vecto tanspose. Hee, optmzaton [] mn ws w C w W st +.. w ( u u ), =,,, u = x s the mean sample vecto of X, and x X s the cadnalty of () X. KDLOR defnes the followng (3) 3. Mnmum class vaance suppot vecto odnal egesson Followng the dea of SVOR, n the lnea case we defne the pmal optmzaton poblem of MCVSVOR as follows wξ b + ξ, ξ 0,, mn ws Ww+ C ( ξ + ξ ) st.. wξ b ξ, ξ 0,, (4) wb,, ξ, ξ = = b b,, Whee =,, =,,, and S W s the wthn-class scatte matx whch s defned as (). Smla to MCVSVM, by ths way, the dstbuton of the categoes s taken fully nto consdeaton. 895
3 Besdes, the poposed method embodes the lage magn pncple snce t s deved fom MCVSVM whch mplements lage magn pncple [5]. So, t s dffeent fom KDLOR although they both take the dstbuton of the categoes nto consdeaton. Smla to SVOR, the pmal optmzaton poblem of MCVSVOR can be effcently solved by ts dual optmzaton poblem. Obvously, (4) s a quadatc pogammng poblem. he pmal Lagangan (4) s L= wsww+ C ( ξ + ξ ) α ( + ξ wξ + β) = = = = (5) + α ( + ξ + wξ b ) βξ β ξ γ ( b b ) = = = = = = = α = [,, ], α = [,, ], β = [,, ], β = Whee the vectos α α α α β β [ β,, β ] and γ = [ γ,, γ ] ae the Lagangan multples fo the constants of (4). By dffeentatng wth espect to w, ξ, ξ and b and usng the Kaush-Kuhn-ucke (KK) condtons, the followng holds L = SWw ( α α ) ξ = 0 w = = = C α b = 0,, ξ (6) = C α 0,, b = ξ = ( α + γ ) + ( α + γ ) = 0, b = = If the matx S W s nonsngula o nvetble, we have W ( α α ) = = w = S x (7) As n MCVSVM and KDLOR, MCVSVOR may encounte the sngulaty poblem of S W snce ts nvese matx s necessay, whch often occus n the case whee the numbe of samples s smalle than the dmensonalty of the samples. o solve ths sngulaty poblem, smla to KDLOR, we can employ the egulazaton method [5, 6, 7] whch s to add a constant ρ > 0 to the dagonal elements of S W as SW = SW + ρi, whee I s an dentty matx. he optmum value of ρ can be estmated though a coss valdaton method. By eplacng (6) nto (5) and usng the KK condtons, the constant optmzaton poblem (4) s efomulated to the Wolf dual poblem ' ' mn ( α α )( α α )( x ) S x ( α + α ) α, α ' ' W, ' ',, st..0 α C,, + 0 α C,, α γ α γ γ = = + = +, 0, Whee uns ove,,. hs s a convex quadatc pogammng poblem and smla to the dual optmzaton poblem of SVOR. Suppose { α, α, γ } s the soluton of the above optmzaton poblem, w s obtaned fom (7), and so the dscmnant functon value fo a new nput vecto x s (8) 896
4 W W = = = = f ( x ) = wx = S ( α α ) x x = ( α α )( x ) S x (9) hus, the pedctve odnal decson functon s gven by mn ag{ : f( x ) < b} (0) 4. Expements 4. Synthetc dataset As s shown n Fg., the synthetc dataset ncludes thee odnal categoes and each categoy conssts n 00 samples. In ths expement, the kenel functon k( xy, ) = exp( γ x y ) s adopted. he expemental esult s llustated n Fg.. It s can be found that the samples can be aanged odely by the hypeplane geneated by MCVSVOR,.e., the samples wth the same ank s classfed n same bn by MCVSVOR. he expemental esult valdates the effectveness of the poposed method Illustaton of the decson hypeplane geneated by MCVSVOR 4. Benchmak datasets In ode to evaluate the pefomance of the poposed method, n ths secton the expemental esults on seveal benchmak datasets, whch wee used n [4] and [], wll be epoted. A summay of the chaactestcs of the selected datasets ae pesented n able. Fo each dataset, the taget values wee dscetzed nto ten odnal quanttes usng equal-fequency bnnng. Each dataset was andomly pattoned nto tanng/test splts as specfed n able. he pattonng was epeated 0 tmes ndependently. he nput vectos wee nomalzed to zeo mean and unt vaance, coodnate-wse. able Chaactestcs of the selected datasets. Datasets o. of Attbutes o. of anng Samples o. of est Samples Pydmnes Machne CPU Boston Housng Abalone Bank Compute Calfona Census
5 5. Concluson In ths pape, we popose a novel odnal egesson method called MCVSVOR. Dffeent fom tadtonal SVOR whch s obtaned by extendng SVM to tackle the odnal egesson poblems, the poposed method s deved fom MCVSVM and nhets ts mets such as good obustness and genealzaton ablty. he expemental esults ndcate the effectveness of MCVSVOR by compang t wth the tadtonal egesson methods SVOR and KDLOR. Refeences [] J. S. Cadoso, R. Sousa, Classfcaton models wth global constants fo odnal data, n: ICMLA, 00, pp [] W. Chu, Z. Ghahaman, Gaussan pocesses fo odnal egesson, Jounal of Machne Leanng Reseach 6 (005) [3] W. Chu, S. S. Keeth, ew appoaches to suppot vecto odnal egesson, n: Poceedng of Intenatonal Confeence on Machne Leanng (ICML-), 005, pp [4] W. Chu, S. S. Keeth, Suppot vecto odnal egesson, eual Computaton 9 (3) (007) [5] R. O. Duda, P. E. Hat, D. G. Stok, Patten Classfcaton (Second Edton), ew Yok: Wley, 00. [6] Y. Guo,. Haste, R. bshan, Regulazed lnea dscmnant analyss and ts applcaton n mcoaays, Bostatstcs 8 () (007) [7] R. Hebch,. Gaepel, K. Obemaye, Suppot vecto leanng fo odnal egesson, n: Intenatonal Confeence on Atfcal eual etwoks, 999, pp [8] M. Kadzńsk, S. Geco, R. Słowńsk, Robust odnal egesson fo domnance-based ough set appoach to multple ctea sotng, Infomaton Scences 83 () (04) -8. [9] Y. Lu, Y. Lu, K. C. Chan, Odnal egesson va manfold leanng, n: AAAI, 0, pp [0] A. Shashua, A. Levn. Rankng wth lage magn pncple: two appoaches, In: Advances n eual Infomaton Pocessng Systems 5, 003, pp [] S. K. Shevade, W. Chu, Mnmum enclosng sphees fomulatons fo suppot vecto odnal egesson, n: Sxth Intenatonal Confeence on Data Mnng, 006, pp: [] B. Y. Sun, J. L, D. D. Wu, X. M. Zhang, W. B. L, Kenel dscmnant leanng fo odnal egesson, IEEE ansactons on Knowledge and Data Engneeng (6) (00) [3] V. oa, J. Domngo-Fee, J. M. Mateo-Sanz, M. g, Regesson fo odnal vaables wthout undelyng contnuous vaables, Infomaton Scences 76 (4) (006) [4] V. Vapnk, he atue of Statstcal Leanng heoy. ew Yok: Spnge Velag, 995. [5] M. Wang, F. L. Chung, S.. Wang, On mnmum class localty pesevng vaance suppot vecto machne, Patten Recognton 43 (8) (00) [6] S. Zafeou, A. efas, I. Ptas, Mnmum class vaance suppot vecto machnes, IEEE ansanctons on Image Pocessng 6 (0) (007)
Exact Simplification of Support Vector Solutions
Jounal of Machne Leanng Reseach 2 (200) 293-297 Submtted 3/0; Publshed 2/0 Exact Smplfcaton of Suppot Vecto Solutons Tom Downs TD@ITEE.UQ.EDU.AU School of Infomaton Technology and Electcal Engneeng Unvesty
More informationMultistage Median Ranked Set Sampling for Estimating the Population Median
Jounal of Mathematcs and Statstcs 3 (: 58-64 007 ISSN 549-3644 007 Scence Publcatons Multstage Medan Ranked Set Samplng fo Estmatng the Populaton Medan Abdul Azz Jeman Ame Al-Oma and Kamaulzaman Ibahm
More informationExperimental study on parameter choices in norm-r support vector regression machines with noisy input
Soft Comput 006) 0: 9 3 DOI 0.007/s00500-005-0474-z ORIGINAL PAPER S. Wang J. Zhu F. L. Chung Hu Dewen Expemental study on paamete choces n nom- suppot vecto egesson machnes wth nosy nput Publshed onlne:
More informationMachine Learning 4771
Machne Leanng 4771 Instucto: Tony Jebaa Topc 6 Revew: Suppot Vecto Machnes Pmal & Dual Soluton Non-sepaable SVMs Kenels SVM Demo Revew: SVM Suppot vecto machnes ae (n the smplest case) lnea classfes that
More informationDistinct 8-QAM+ Perfect Arrays Fanxin Zeng 1, a, Zhenyu Zhang 2,1, b, Linjie Qian 1, c
nd Intenatonal Confeence on Electcal Compute Engneeng and Electoncs (ICECEE 15) Dstnct 8-QAM+ Pefect Aays Fanxn Zeng 1 a Zhenyu Zhang 1 b Lnje Qan 1 c 1 Chongqng Key Laboatoy of Emegency Communcaton Chongqng
More informationCOMPLEMENTARY ENERGY METHOD FOR CURVED COMPOSITE BEAMS
ultscence - XXX. mcocd Intenatonal ultdscplnay Scentfc Confeence Unvesty of skolc Hungay - pl 06 ISBN 978-963-358-3- COPLEENTRY ENERGY ETHOD FOR CURVED COPOSITE BES Ákos József Lengyel István Ecsed ssstant
More informationOptimization Methods: Linear Programming- Revised Simplex Method. Module 3 Lecture Notes 5. Revised Simplex Method, Duality and Sensitivity analysis
Optmzaton Meods: Lnea Pogammng- Revsed Smple Meod Module Lectue Notes Revsed Smple Meod, Dualty and Senstvty analyss Intoducton In e pevous class, e smple meod was dscussed whee e smple tableau at each
More informationMachine Learning. Spectral Clustering. Lecture 23, April 14, Reading: Eric Xing 1
Machne Leanng -7/5 7/5-78, 78, Spng 8 Spectal Clusteng Ec Xng Lectue 3, pl 4, 8 Readng: Ec Xng Data Clusteng wo dffeent ctea Compactness, e.g., k-means, mxtue models Connectvty, e.g., spectal clusteng
More informationIf there are k binding constraints at x then re-label these constraints so that they are the first k constraints.
Mathematcal Foundatons -1- Constaned Optmzaton Constaned Optmzaton Ma{ f ( ) X} whee X {, h ( ), 1,, m} Necessay condtons fo to be a soluton to ths mamzaton poblem Mathematcally, f ag Ma{ f ( ) X}, then
More informationScalars and Vectors Scalar
Scalas and ectos Scala A phscal quantt that s completel chaacteed b a eal numbe (o b ts numecal value) s called a scala. In othe wods a scala possesses onl a magntude. Mass denst volume tempeatue tme eneg
More informationUNIT10 PLANE OF REGRESSION
UIT0 PLAE OF REGRESSIO Plane of Regesson Stuctue 0. Intoducton Ojectves 0. Yule s otaton 0. Plane of Regesson fo thee Vaales 0.4 Popetes of Resduals 0.5 Vaance of the Resduals 0.6 Summay 0.7 Solutons /
More informationSet of square-integrable function 2 L : function space F
Set of squae-ntegable functon L : functon space F Motvaton: In ou pevous dscussons we have seen that fo fee patcles wave equatons (Helmholt o Schödnge) can be expessed n tems of egenvalue equatons. H E,
More informationA Method of Reliability Target Setting for Electric Power Distribution Systems Using Data Envelopment Analysis
27 กก ก 9 2-3 2554 ก ก ก A Method of Relablty aget Settng fo Electc Powe Dstbuton Systems Usng Data Envelopment Analyss ก 2 ก ก ก ก ก 0900 2 ก ก ก ก ก 0900 E-mal: penjan262@hotmal.com Penjan Sng-o Psut
More informationRobust Feature Induction for Support Vector Machines
Robust Featue Inducton fo Suppot Vecto Machnes Rong Jn Depatment of Compute Scence and Engneeng, Mchgan State Unvesty, East Lansng, MI4884 ROGJI@CSE.MSU.EDU Huan Lu Depatment of Compute Scence and Engneeng,
More informationThermodynamics of solids 4. Statistical thermodynamics and the 3 rd law. Kwangheon Park Kyung Hee University Department of Nuclear Engineering
Themodynamcs of solds 4. Statstcal themodynamcs and the 3 d law Kwangheon Pak Kyung Hee Unvesty Depatment of Nuclea Engneeng 4.1. Intoducton to statstcal themodynamcs Classcal themodynamcs Statstcal themodynamcs
More informationEfficiency of the principal component Liu-type estimator in logistic
Effcency of the pncpal component Lu-type estmato n logstc egesson model Jbo Wu and Yasn Asa 2 School of Mathematcs and Fnance, Chongqng Unvesty of Ats and Scences, Chongqng, Chna 2 Depatment of Mathematcs-Compute
More informationA Brief Guide to Recognizing and Coping With Failures of the Classical Regression Assumptions
A Bef Gude to Recognzng and Copng Wth Falues of the Classcal Regesson Assumptons Model: Y 1 k X 1 X fxed n epeated samples IID 0, I. Specfcaton Poblems A. Unnecessay explanatoy vaables 1. OLS s no longe
More informationP 365. r r r )...(1 365
SCIENCE WORLD JOURNAL VOL (NO4) 008 www.scecncewoldounal.og ISSN 597-64 SHORT COMMUNICATION ANALYSING THE APPROXIMATION MODEL TO BIRTHDAY PROBLEM *CHOJI, D.N. & DEME, A.C. Depatment of Mathematcs Unvesty
More informationOn Maneuvering Target Tracking with Online Observed Colored Glint Noise Parameter Estimation
Wold Academy of Scence, Engneeng and Technology 6 7 On Maneuveng Taget Tacng wth Onlne Obseved Coloed Glnt Nose Paamete Estmaton M. A. Masnad-Sha, and S. A. Banan Abstact In ths pape a compehensve algothm
More informationPHYS 705: Classical Mechanics. Derivation of Lagrange Equations from D Alembert s Principle
1 PHYS 705: Classcal Mechancs Devaton of Lagange Equatons fom D Alembet s Pncple 2 D Alembet s Pncple Followng a smla agument fo the vtual dsplacement to be consstent wth constants,.e, (no vtual wok fo
More informationResearch Article Incremental Tensor Principal Component Analysis for Handwritten Digit Recognition
Hndaw Publshng Copoaton athematcal Poblems n Engneeng, Atcle ID 89758, 0 pages http://dx.do.og/0.55/04/89758 Reseach Atcle Incemental enso Pncpal Component Analyss fo Handwtten Dgt Recognton Chang Lu,,
More information8 Baire Category Theorem and Uniform Boundedness
8 Bae Categoy Theoem and Unfom Boundedness Pncple 8.1 Bae s Categoy Theoem Valdty of many esults n analyss depends on the completeness popety. Ths popety addesses the nadequacy of the system of atonal
More informationEnergy in Closed Systems
Enegy n Closed Systems Anamta Palt palt.anamta@gmal.com Abstact The wtng ndcates a beakdown of the classcal laws. We consde consevaton of enegy wth a many body system n elaton to the nvese squae law and
More informationPhysics 11b Lecture #2. Electric Field Electric Flux Gauss s Law
Physcs 11b Lectue # Electc Feld Electc Flux Gauss s Law What We Dd Last Tme Electc chage = How object esponds to electc foce Comes n postve and negatve flavos Conseved Electc foce Coulomb s Law F Same
More informationAPPLICATIONS OF SEMIGENERALIZED -CLOSED SETS
Intenatonal Jounal of Mathematcal Engneeng Scence ISSN : 22776982 Volume Issue 4 (Apl 202) http://www.mes.com/ https://stes.google.com/ste/mesounal/ APPLICATIONS OF SEMIGENERALIZED CLOSED SETS G.SHANMUGAM,
More information3. A Review of Some Existing AW (BT, CT) Algorithms
3. A Revew of Some Exstng AW (BT, CT) Algothms In ths secton, some typcal ant-wndp algothms wll be descbed. As the soltons fo bmpless and condtoned tansfe ae smla to those fo ant-wndp, the pesented algothms
More informationON THE FRESNEL SINE INTEGRAL AND THE CONVOLUTION
IJMMS 3:37, 37 333 PII. S16117131151 http://jmms.hndaw.com Hndaw Publshng Cop. ON THE FRESNEL SINE INTEGRAL AND THE CONVOLUTION ADEM KILIÇMAN Receved 19 Novembe and n evsed fom 7 Mach 3 The Fesnel sne
More informationDirichlet Mixture Priors: Inference and Adjustment
Dchlet Mxtue Pos: Infeence and Adustment Xugang Ye (Wokng wth Stephen Altschul and Y Kuo Yu) Natonal Cante fo Botechnology Infomaton Motvaton Real-wold obects Independent obsevatons Categocal data () (2)
More informationAn Approach to Inverse Fuzzy Arithmetic
An Appoach to Invese Fuzzy Athmetc Mchael Hanss Insttute A of Mechancs, Unvesty of Stuttgat Stuttgat, Gemany mhanss@mechaun-stuttgatde Abstact A novel appoach of nvese fuzzy athmetc s ntoduced to successfully
More informationPARAMETER ESTIMATION FOR TWO WEIBULL POPULATIONS UNDER JOINT TYPE II CENSORED SCHEME
Sept 04 Vol 5 No 04 Intenatonal Jounal of Engneeng Appled Scences 0-04 EAAS & ARF All ghts eseed wwweaas-ounalog ISSN305-869 PARAMETER ESTIMATION FOR TWO WEIBULL POPULATIONS UNDER JOINT TYPE II CENSORED
More informationSupport Vector Machines. Vibhav Gogate The University of Texas at dallas
Support Vector Machnes Vbhav Gogate he Unversty of exas at dallas What We have Learned So Far? 1. Decson rees. Naïve Bayes 3. Lnear Regresson 4. Logstc Regresson 5. Perceptron 6. Neural networks 7. K-Nearest
More informationCSJM University Class: B.Sc.-II Sub:Physics Paper-II Title: Electromagnetics Unit-1: Electrostatics Lecture: 1 to 4
CSJM Unvesty Class: B.Sc.-II Sub:Physcs Pape-II Ttle: Electomagnetcs Unt-: Electostatcs Lectue: to 4 Electostatcs: It deals the study of behavo of statc o statonay Chages. Electc Chage: It s popety by
More informationObserver Design for Takagi-Sugeno Descriptor System with Lipschitz Constraints
Intenatonal Jounal of Instumentaton and Contol Systems (IJICS) Vol., No., Apl Obseve Desgn fo akag-sugeno Descpto System wth Lpschtz Constants Klan Ilhem,Jab Dalel, Bel Hadj Al Saloua and Abdelkm Mohamed
More informationAN EXACT METHOD FOR BERTH ALLOCATION AT RAW MATERIAL DOCKS
AN EXACT METHOD FOR BERTH ALLOCATION AT RAW MATERIAL DOCKS Shaohua L, a, Lxn Tang b, Jyn Lu c a Key Laboatoy of Pocess Industy Automaton, Mnsty of Educaton, Chna b Depatment of Systems Engneeng, Notheasten
More informationFUZZY CONTROL VIA IMPERFECT PREMISE MATCHING APPROACH FOR DISCRETE TAKAGI-SUGENO FUZZY SYSTEMS WITH MULTIPLICATIVE NOISES
Jounal of Mane Scence echnology Vol. 4 No.5 pp. 949-957 (6) 949 DOI:.69/JMS-6-54- FUZZY CONROL VIA IMPERFEC PREMISE MACHING APPROACH FOR DISCREE AKAGI-SUGENO FUZZY SYSEMS WIH MULIPLICAIVE NOISES Wen-Je
More informationVibration Input Identification using Dynamic Strain Measurement
Vbaton Input Identfcaton usng Dynamc Stan Measuement Takum ITOFUJI 1 ;TakuyaYOSHIMURA ; 1, Tokyo Metopoltan Unvesty, Japan ABSTRACT Tansfe Path Analyss (TPA) has been conducted n ode to mpove the nose
More informationN = N t ; t 0. N is the number of claims paid by the
Iulan MICEA, Ph Mhaela COVIG, Ph Canddate epatment of Mathematcs The Buchaest Academy of Economc Studes an CECHIN-CISTA Uncedt Tac Bank, Lugoj SOME APPOXIMATIONS USE IN THE ISK POCESS OF INSUANCE COMPANY
More informationConstraint Score: A New Filter Method for Feature Selection with Pairwise Constraints
onstant Scoe: A New Flte ethod fo Featue Selecton wth Pawse onstants Daoqang Zhang, Songcan hen and Zh-Hua Zhou Depatment of ompute Scence and Engneeng Nanjng Unvesty of Aeonautcs and Astonautcs, Nanjng
More informationKhintchine-Type Inequalities and Their Applications in Optimization
Khntchne-Type Inequaltes and The Applcatons n Optmzaton Anthony Man-Cho So Depatment of Systems Engneeng & Engneeng Management The Chnese Unvesty of Hong Kong ISDS-Kolloquum Unvestaet Wen 29 June 2009
More informationLearning the structure of Bayesian belief networks
Lectue 17 Leanng the stuctue of Bayesan belef netwoks Mlos Hauskecht mlos@cs.ptt.edu 5329 Sennott Squae Leanng of BBN Leanng. Leanng of paametes of condtonal pobabltes Leanng of the netwok stuctue Vaables:
More informationGENERALIZED MULTIVARIATE EXPONENTIAL TYPE (GMET) ESTIMATOR USING MULTI-AUXILIARY INFORMATION UNDER TWO-PHASE SAMPLING
Pak. J. Statst. 08 Vol. (), 9-6 GENERALIZED MULTIVARIATE EXPONENTIAL TYPE (GMET) ESTIMATOR USING MULTI-AUXILIARY INFORMATION UNDER TWO-PHASE SAMPLING Ayesha Ayaz, Zahoo Ahmad, Aam Sanaullah and Muhammad
More informationVParC: A Compression Scheme for Numeric Data in Column-Oriented Databases
The Intenatonal Aab Jounal of Infomaton Technology VPaC: A Compesson Scheme fo Numec Data n Column-Oented Databases Ke Yan, Hong Zhu, and Kevn Lü School of Compute Scence and Technology, Huazhong Unvesty
More informationThe Greatest Deviation Correlation Coefficient and its Geometrical Interpretation
By Rudy A. Gdeon The Unvesty of Montana The Geatest Devaton Coelaton Coeffcent and ts Geometcal Intepetaton The Geatest Devaton Coelaton Coeffcent (GDCC) was ntoduced by Gdeon and Hollste (987). The GDCC
More informationGENERALIZATION OF AN IDENTITY INVOLVING THE GENERALIZED FIBONACCI NUMBERS AND ITS APPLICATIONS
#A39 INTEGERS 9 (009), 497-513 GENERALIZATION OF AN IDENTITY INVOLVING THE GENERALIZED FIBONACCI NUMBERS AND ITS APPLICATIONS Mohaad Faokh D. G. Depatent of Matheatcs, Fedows Unvesty of Mashhad, Mashhad,
More informationGroupoid and Topological Quotient Group
lobal Jounal of Pue and Appled Mathematcs SSN 0973-768 Volume 3 Numbe 7 07 pp 373-39 Reseach nda Publcatons http://wwwpublcatoncom oupod and Topolocal Quotent oup Mohammad Qasm Manna Depatment of Mathematcs
More informationSOME NEW SELF-DUAL [96, 48, 16] CODES WITH AN AUTOMORPHISM OF ORDER 15. KEYWORDS: automorphisms, construction, self-dual codes
Факултет по математика и информатика, том ХVІ С, 014 SOME NEW SELF-DUAL [96, 48, 16] CODES WITH AN AUTOMORPHISM OF ORDER 15 NIKOLAY I. YANKOV ABSTRACT: A new method fo constuctng bnay self-dual codes wth
More informationRigid Bodies: Equivalent Systems of Forces
Engneeng Statcs, ENGR 2301 Chapte 3 Rgd Bodes: Equvalent Sstems of oces Intoducton Teatment of a bod as a sngle patcle s not alwas possble. In geneal, the se of the bod and the specfc ponts of applcaton
More informationNew Condition of Stabilization of Uncertain Continuous Takagi-Sugeno Fuzzy System based on Fuzzy Lyapunov Function
I.J. Intellgent Systems and Applcatons 4 9-5 Publshed Onlne Apl n MCS (http://www.mecs-pess.og/) DOI:.585/sa..4. New Condton of Stablzaton of Uncetan Contnuous aag-sugeno Fuzzy System based on Fuzzy Lyapunov
More informationEngineering Mechanics. Force resultants, Torques, Scalar Products, Equivalent Force systems
Engneeng echancs oce esultants, Toques, Scala oducts, Equvalent oce sstems Tata cgaw-hll Companes, 008 Resultant of Two oces foce: acton of one bod on anothe; chaacteed b ts pont of applcaton, magntude,
More informationChapter 8. Linear Momentum, Impulse, and Collisions
Chapte 8 Lnea oentu, Ipulse, and Collsons 8. Lnea oentu and Ipulse The lnea oentu p of a patcle of ass ovng wth velocty v s defned as: p " v ote that p s a vecto that ponts n the sae decton as the velocty
More informationA NOVEL DWELLING TIME DESIGN METHOD FOR LOW PROBABILITY OF INTERCEPT IN A COMPLEX RADAR NETWORK
Z. Zhang et al., Int. J. of Desgn & Natue and Ecodynamcs. Vol. 0, No. 4 (205) 30 39 A NOVEL DWELLING TIME DESIGN METHOD FOR LOW PROBABILITY OF INTERCEPT IN A COMPLEX RADAR NETWORK Z. ZHANG,2,3, J. ZHU
More informationTHE REGRESSION MODEL OF TRANSMISSION LINE ICING BASED ON NEURAL NETWORKS
The 4th Intenatonal Wokshop on Atmosphec Icng of Stuctues, Chongqng, Chna, May 8 - May 3, 20 THE REGRESSION MODEL OF TRANSMISSION LINE ICING BASED ON NEURAL NETWORKS Sun Muxa, Da Dong*, Hao Yanpeng, Huang
More informationState Feedback Controller Design via Takagi- Sugeno Fuzzy Model : LMI Approach
State Feedback Contolle Desgn va akag- Sugeno Fuzzy Model : LMI Appoach F. Khabe, K. Zeha, and A. Hamzaou Abstact In ths pape, we ntoduce a obust state feedback contolle desgn usng Lnea Matx Inequaltes
More informationPhysics 2A Chapter 11 - Universal Gravitation Fall 2017
Physcs A Chapte - Unvesal Gavtaton Fall 07 hese notes ae ve pages. A quck summay: he text boxes n the notes contan the esults that wll compse the toolbox o Chapte. hee ae thee sectons: the law o gavtaton,
More informationUnknown Input Based Observer Synthesis for a Polynomial T-S Fuzzy Model System with Uncertainties
Unknown Input Based Obseve Synthess fo a Polynomal -S Fuzzy Model System wth Uncetantes Van-Phong Vu Wen-June Wang Fellow IEEE Hsang-heh hen Jacek M Zuada Lfe Fellow IEEE Abstact hs pape poposes a new
More informationA Queuing Model for an Automated Workstation Receiving Jobs from an Automated Workstation
Intenatonal Jounal of Opeatons Reseach Intenatonal Jounal of Opeatons Reseach Vol. 7, o. 4, 918 (1 A Queung Model fo an Automated Wokstaton Recevng Jobs fom an Automated Wokstaton Davd S. Km School of
More informationAdvanced Robust PDC Fuzzy Control of Nonlinear Systems
Advanced obust PDC Fuzzy Contol of Nonlnea Systems M Polanský Abstact hs pape ntoduces a new method called APDC (Advanced obust Paallel Dstbuted Compensaton) fo automatc contol of nonlnea systems hs method
More informationan application to HRQoL
AlmaMate Studoum Unvesty of Bologna A flexle IRT Model fo health questonnae: an applcaton to HRQoL Seena Boccol Gula Cavn Depatment of Statstcal Scence, Unvesty of Bologna 9 th Intenatonal Confeence on
More informationGenerating Functions, Weighted and Non-Weighted Sums for Powers of Second-Order Recurrence Sequences
Geneatng Functons, Weghted and Non-Weghted Sums fo Powes of Second-Ode Recuence Sequences Pantelmon Stăncă Aubun Unvesty Montgomey, Depatment of Mathematcs Montgomey, AL 3614-403, USA e-mal: stanca@studel.aum.edu
More informationThe Forming Theory and the NC Machining for The Rotary Burs with the Spectral Edge Distribution
oden Appled Scence The Fomn Theoy and the NC achnn fo The Rotay us wth the Spectal Ede Dstbuton Huan Lu Depatment of echancal Enneen, Zhejan Unvesty of Scence and Technoloy Hanzhou, c.y. chan, 310023,
More informationA. Thicknesses and Densities
10 Lab0 The Eath s Shells A. Thcknesses and Denstes Any theoy of the nteo of the Eath must be consstent wth the fact that ts aggegate densty s 5.5 g/cm (ecall we calculated ths densty last tme). In othe
More informationRanks of quotients, remainders and p-adic digits of matrices
axv:1401.6667v2 [math.nt] 31 Jan 2014 Ranks of quotents, emandes and p-adc dgts of matces Mustafa Elshekh Andy Novocn Mak Gesbecht Abstact Fo a pme p and a matx A Z n n, wte A as A = p(a quo p)+ (A em
More informationImplementation in the ANSYS Finite Element Code of the Electric Vector Potential T-Ω,Ω Formulation
Implementaton n the ANSYS Fnte Element Code of the Electc Vecto Potental T-Ω,Ω Fomulaton Peto Teston Dpatmento d Ingegnea Elettca ed Elettonca, Unvestà d Cagla Pazza d Am, 0923 Cagla Pegogo Sonato Dpatmento
More informationCEEP-BIT WORKING PAPER SERIES. Efficiency evaluation of multistage supply chain with data envelopment analysis models
CEEP-BIT WORKING PPER SERIES Effcency evaluaton of multstage supply chan wth data envelopment analyss models Ke Wang Wokng Pape 48 http://ceep.bt.edu.cn/englsh/publcatons/wp/ndex.htm Cente fo Enegy and
More informationMulti-element based on proxy re-encryption scheme for mobile cloud computing
36 11 Vol.36 No.11 015 11 Jounal on Communcatons Novembe 015 do:10.11959/.ssn.1000-436x.01517 1 1 1. 10094. 100070 TP309. A Mult-element based on poxy e-encypton scheme fo moble cloud computng SU Mang
More information4 Recursive Linear Predictor
4 Recusve Lnea Pedcto The man objectve of ths chapte s to desgn a lnea pedcto wthout havng a po knowledge about the coelaton popetes of the nput sgnal. In the conventonal lnea pedcto the known coelaton
More informationState Estimation. Ali Abur Northeastern University, USA. Nov. 01, 2017 Fall 2017 CURENT Course Lecture Notes
State Estmaton Al Abu Notheasten Unvesty, USA Nov. 0, 07 Fall 07 CURENT Couse Lectue Notes Opeatng States of a Powe System Al Abu NORMAL STATE SECURE o INSECURE RESTORATIVE STATE EMERGENCY STATE PARTIAL
More information19 The Born-Oppenheimer Approximation
9 The Bon-Oppenheme Appoxmaton The full nonelatvstc Hamltonan fo a molecule s gven by (n a.u.) Ĥ = A M A A A, Z A + A + >j j (883) Lets ewte the Hamltonan to emphasze the goal as Ĥ = + A A A, >j j M A
More informationPattern Analyses (EOF Analysis) Introduction Definition of EOFs Estimation of EOFs Inference Rotated EOFs
Patten Analyses (EOF Analyss) Intoducton Defnton of EOFs Estmaton of EOFs Infeence Rotated EOFs . Patten Analyses Intoducton: What s t about? Patten analyses ae technques used to dentfy pattens of the
More informationLASER ABLATION ICP-MS: DATA REDUCTION
Lee, C-T A Lase Ablaton Data educton 2006 LASE ABLATON CP-MS: DATA EDUCTON Cn-Ty A. Lee 24 Septembe 2006 Analyss and calculaton of concentatons Lase ablaton analyses ae done n tme-esolved mode. A ~30 s
More informationV. Principles of Irreversible Thermodynamics. s = S - S 0 (7.3) s = = - g i, k. "Flux": = da i. "Force": = -Â g a ik k = X i. Â J i X i (7.
Themodynamcs and Knetcs of Solds 71 V. Pncples of Ievesble Themodynamcs 5. Onsage s Teatment s = S - S 0 = s( a 1, a 2,...) a n = A g - A n (7.6) Equlbum themodynamcs detemnes the paametes of an equlbum
More informationEffective Discriminative Feature Selection with Non-trivial Solutions
Effectve Dscmnatve Featue Selecton wth Non-tval Solutons Hong Tao, Chenpng Hou, Membe, IEEE, Fepng Ne, Yuanyuan Jao, Dongyun Y axv:4.548v [cs.lg] Ap 5 Abstact Featue selecton and featue tansfomaton, the
More informationMechanics Physics 151
Mechancs Physcs 151 Lectue 18 Hamltonan Equatons of Moton (Chapte 8) What s Ahead We ae statng Hamltonan fomalsm Hamltonan equaton Today and 11/6 Canoncal tansfomaton 1/3, 1/5, 1/10 Close lnk to non-elatvstc
More informationUsing DP for hierarchical discretization of continuous attributes. Amit Goyal (31 st March 2008)
Usng DP fo heachcal dscetzaton of contnos attbtes Amt Goyal 31 st Mach 2008 Refeence Chng-Cheng Shen and Yen-Lang Chen. A dynamc-pogammng algothm fo heachcal dscetzaton of contnos attbtes. In Eopean Jonal
More informationTian Zheng Department of Statistics Columbia University
Haplotype Tansmsson Assocaton (HTA) An "Impotance" Measue fo Selectng Genetc Makes Tan Zheng Depatment of Statstcs Columba Unvesty Ths s a jont wok wth Pofesso Shaw-Hwa Lo n the Depatment of Statstcs at
More informationOptimal System for Warm Standby Components in the Presence of Standby Switching Failures, Two Types of Failures and General Repair Time
Intenatonal Jounal of ompute Applcatons (5 ) Volume 44 No, Apl Optmal System fo Wam Standby omponents n the esence of Standby Swtchng Falues, Two Types of Falues and Geneal Repa Tme Mohamed Salah EL-Shebeny
More informationDetection and Estimation Theory
ESE 54 Detecton and Etmaton Theoy Joeph A. O Sullvan Samuel C. Sach Pofeo Electonc Sytem and Sgnal Reeach Laboatoy Electcal and Sytem Engneeng Wahngton Unvety 411 Jolley Hall 314-935-4173 (Lnda anwe) jao@wutl.edu
More informationINTERVAL ESTIMATION FOR THE QUANTILE OF A TWO-PARAMETER EXPONENTIAL DISTRIBUTION
Intenatonal Jounal of Innovatve Management, Infomaton & Poducton ISME Intenatonalc0 ISSN 85-5439 Volume, Numbe, June 0 PP. 78-8 INTERVAL ESTIMATION FOR THE QUANTILE OF A TWO-PARAMETER EXPONENTIAL DISTRIBUTION
More informationAnalytical and Numerical Solutions for a Rotating Annular Disk of Variable Thickness
Appled Mathematcs 00 43-438 do:0.436/am.00.5057 Publshed Onlne Novembe 00 (http://www.scrp.og/jounal/am) Analytcal and Numecal Solutons fo a Rotatng Annula Ds of Vaable Thcness Abstact Ashaf M. Zenou Daoud
More information(8) Gain Stage and Simple Output Stage
EEEB23 Electoncs Analyss & Desgn (8) Gan Stage and Smple Output Stage Leanng Outcome Able to: Analyze an example of a gan stage and output stage of a multstage amplfe. efeence: Neamen, Chapte 11 8.0) ntoducton
More informationMULTIPOLE FIELDS. Multipoles, 2 l poles. Monopoles, dipoles, quadrupoles, octupoles... Electric Dipole R 1 R 2. P(r,θ,φ) e r
MULTIPOLE FIELDS Mutpoes poes. Monopoes dpoes quadupoes octupoes... 4 8 6 Eectc Dpoe +q O θ e R R P(θφ) -q e The potenta at the fed pont P(θφ) s ( θϕ )= q R R Bo E. Seneus : Now R = ( e) = + cosθ R = (
More informationFuzzy Controller Design for Markovian Jump Nonlinear Systems
72 Intenatonal Jounal of Juxang Contol Dong Automaton and Guang-Hong and Systems Yang vol. 5 no. 6 pp. 72-77 Decembe 27 Fuzzy Contolle Desgn fo Maovan Jump Nonlnea Systems Juxang Dong and Guang-Hong Yang*
More informationA Study about One-Dimensional Steady State. Heat Transfer in Cylindrical and. Spherical Coordinates
Appled Mathematcal Scences, Vol. 7, 03, no. 5, 67-633 HIKARI Ltd, www.m-hka.com http://dx.do.og/0.988/ams.03.38448 A Study about One-Dmensonal Steady State Heat ansfe n ylndcal and Sphecal oodnates Lesson
More informationLecture 10 Support Vector Machines II
Lecture 10 Support Vector Machnes II 22 February 2016 Taylor B. Arnold Yale Statstcs STAT 365/665 1/28 Notes: Problem 3 s posted and due ths upcomng Frday There was an early bug n the fake-test data; fxed
More information4 SingularValue Decomposition (SVD)
/6/00 Z:\ jeh\self\boo Kannan\Jan-5-00\4 SVD 4 SngulaValue Decomposton (SVD) Chapte 4 Pat SVD he sngula value decomposton of a matx s the factozaton of nto the poduct of thee matces = UDV whee the columns
More informationA Micro-Doppler Modulation of Spin Projectile on CW Radar
ITM Web of Confeences 11, 08005 (2017) DOI: 10.1051/ tmconf/20171108005 A Mco-Dopple Modulaton of Spn Pojectle on CW Rada Zh-Xue LIU a Bacheng Odnance Test Cente of Chna, Bacheng 137001, P. R. Chna Abstact.
More informationCorrespondence Analysis & Related Methods
Coespondence Analyss & Related Methods Ineta contbutons n weghted PCA PCA s a method of data vsualzaton whch epesents the tue postons of ponts n a map whch comes closest to all the ponts, closest n sense
More informationVISUALIZATION OF THE ABSTRACT THEORIES IN DSP COURSE BASED ON CDIO CONCEPT
VISUALIZATION OF THE ABSTRACT THEORIES IN DSP COURSE BASED ON CDIO CONCEPT Wang L-uan, L Jan, Zhen Xao-qong Chengdu Unvesty of Infomaton Technology ABSTRACT The pape analyzes the chaactestcs of many fomulas
More informationKNAPSACK PROBLEMS WITH SETUP. Yanchun Yang. A Dissertation. Submitted to. the Graduate Faculty of. Auburn University. in Partial Fulfillment of the
KAPSACK PROBLEMS WITH SETUP Yanchun Yang A Dssetaton Submtted to the Gaduate Faculty of Aubun Unvesty n Patal Fulfllment of the Requements fo the Degee of Docto of Phlosophy Aubun, Alabama August 7, 2006
More informationThe Study of Teaching-learning-based Optimization Algorithm
Advanced Scence and Technology Letters Vol. (AST 06), pp.05- http://dx.do.org/0.57/astl.06. The Study of Teachng-learnng-based Optmzaton Algorthm u Sun, Yan fu, Lele Kong, Haolang Q,, Helongang Insttute
More informationStudy on Vibration Response Reduction of Bladed Disk by Use of Asymmetric Vane Spacing (Study on Response Reduction of Mistuned Bladed Disk)
Intenatonal Jounal of Gas ubne, Populson and Powe Systems Febuay 0, Volume 4, Numbe Study on Vbaton Response Reducton of Bladed Dsk by Use of Asymmetc Vane Spacng (Study on Response Reducton of Mstuned
More informationA New Approach for Deriving the Instability Potential for Plates Based on Rigid Body and Force Equilibrium Considerations
Avalable onlne at www.scencedect.com Poceda Engneeng 4 (20) 4 22 The Twelfth East Asa-Pacfc Confeence on Stuctual Engneeng and Constucton A New Appoach fo Devng the Instablty Potental fo Plates Based on
More informationTHE FUZZY MAPPING AGGREGATION OPERATOR BASED ON RIMER AND ITS APPLICATION
Intenatonal Jounal of Computatonal Intellgence Systems, Vol. 7, No. 2 (Apl 2014), 264-271 THE FUZZY MAPPING AGGREGATION OPERATOR BASED ON RIMER AND ITS APPLICATION Xaopng Qu, Mng Jan School of Tanspotaton
More informationKernel Methods and SVMs Extension
Kernel Methods and SVMs Extenson The purpose of ths document s to revew materal covered n Machne Learnng 1 Supervsed Learnng regardng support vector machnes (SVMs). Ths document also provdes a general
More informationPart V: Velocity and Acceleration Analysis of Mechanisms
Pat V: Velocty an Acceleaton Analyss of Mechansms Ths secton wll evew the most common an cuently pactce methos fo completng the knematcs analyss of mechansms; escbng moton though velocty an acceleaton.
More informationOptimization Algorithms for System Integration
Optmzaton Algothms fo System Integaton Costas Papadmtou 1, a and Evaggelos totsos 1,b 1 Unvesty of hessaly, Depatment of Mechancal and Industal Engneeng, Volos 38334, Geece a costasp@uth.g, b entotso@uth.g
More informationAsymptotic Waves for a Non Linear System
Int Jounal of Math Analyss, Vol 3, 9, no 8, 359-367 Asymptotc Waves fo a Non Lnea System Hamlaou Abdelhamd Dépatement de Mathématques, Faculté des Scences Unvesté Bad Mokhta BP,Annaba, Algea hamdhamlaou@yahoof
More informationDISC-GLASSO: DISCRIMINATIVE GRAPH LEARNING WITH SPARSITY REGULARIZATION. 201 Broadway, Cambridge, MA 02139, USA
DISC-GLASSO: DISCRIMINATIVE GRAPH LEARNING WITH SPARSITY REGULARIZATION Jun-Yu Kao,2 Dong Tan Hassan Mansou Antono Otega 2 Anthony Veto Mtsubsh Electc Reseach Labs (MERL), 20 Boadway, Cambdge, MA 0239,
More informationIntegral Vector Operations and Related Theorems Applications in Mechanics and E&M
Dola Bagayoko (0) Integal Vecto Opeatons and elated Theoems Applcatons n Mechancs and E&M Ι Basc Defnton Please efe to you calculus evewed below. Ι, ΙΙ, andιιι notes and textbooks fo detals on the concepts
More informationRotating Variable-Thickness Inhomogeneous Cylinders: Part II Viscoelastic Solutions and Applications
Appled Mathematcs 010 1 489-498 do:10.436/am.010.16064 Publshed Onlne Decembe 010 (http://www.scrp.og/jounal/am) Rotatng Vaable-Thckness Inhomogeneous Cylndes: Pat II Vscoelastc Solutons and Applcatons
More information