APPLICATION OF A SUPPORT VECTOR MACHINE FOR LIQUEFACTION ASSESSMENT

Size: px
Start display at page:

Download "APPLICATION OF A SUPPORT VECTOR MACHINE FOR LIQUEFACTION ASSESSMENT"

Transcription

1 38 Journal of Marne Scence and echnology, Vol., o. 3, pp (03) DOI: 0.69/JMS APPLICAIO OF A SUPPOR VECOR MACHIE FOR LIQUEFACIO ASSESSME Chng-Ynn Lee and Shuh-G Chern Key ords: A, CP, lquefacton, SVM. ABSRAC hs study presents a support vector machne (SVM)-based approach for predctng earthquake lquefacton. he SVM model database ncludes fve ndexes: earthquake magntude, total overburden pressure, effectve overburden pressure, q c values from cone penetraton tests (CP), and peak ground acceleraton. he proposed model as traned and tested on a dataset comprsng 466 feld lquefacton performance records and CP measurements. A grd search method th k-fold cross-valdaton as also used to verfy the feasblty. Compared th an artfcal neural netork (A) based method, the SVM-based method has the advantage of ncreased accuracy and smpler operaton. Expermental results sho that the proposed SVM approach can ncrease the classfcaton accuracy rate to a standard of 98.7%. I. IRODUCIO Lquefacton s one of the most destructve phenomena caused by earthquakes, and often occurs n loose, saturated sol deposts. Examples of lquefacton nclude the earthquakes n gata, 964; Alaska, 964; angshan, 979; Loma Preta, 989; Kobe, 995; urkey, 998; Ch-Ch, aan, 999; and Honshu, Japan, 0. In ve of the serous damage caused by earthquake-nduced lquefacton, geotechncal engneers are actvely engaged n the study of the sol lquefacton caused by earthquakes, and have developed many assessment methods for evaluatng sol lquefacton. Hoever, the hgh uncertanty n earthquake envronments and sol characterstcs makes t dffcult to choose a sutable emprcal equaton for regresson analyss. herefore, many scholars and experts have attempted to develop scentfcally Paper Submtted //; revsed 04/4/; accepted 05/8/. Author for correspondence: Shuh-G Chern (e-mal: sgchern@mal.ntou.edu.t). Department of Harbor and Rver Engneerng, atonal aan Ocean Unversty, Keelung, aan, R.O.C. derved analytcal models that are smpler, easer to mplement, and more accurate than tradtonal emprcal equatons for sol lquefacton analyss. Many of the exstng assessment methods ere developed from observatons of the behavor of stes durng earthquakes. Geotechncal engneers have often used the smple lquefacton analytcal model developed by SP-, because of ts computatonal speed and analytcal ablty. Based on recent mprovements n data processng and analytcal ablty, the cone penetraton test (CP) offers the advantage of fast, contnuous, and accurate sol parameter measurements. Related testng data has also contnued to accumulate. herefore, the potental of applyng CP to lquefacton research has gron sgnfcantly. hs study presents a relatvely ne soft computng method knon as a support vector machne (SVM) [, 5]. SVMs have been dely used n recent years n areas such as mage dentfcaton and facal recognton. An dentfcaton model that adopts SVM analyss s an effectve method for accurately predctng lquefacton, and can be used n practcal applcatons. Prevous studes have shon that the SVM method s a poerful and effectve tool for dealng th lquefacton problems, and s more accurate and relable than conventonal methods [8, 0]. II. OVER VIEW OF SVM hs secton presents the basc SVM concepts for typcal bnary classfcaton problems.. Lnear SVM A support vector machne, as presented by Vapnk [], s a machne learnng algorthm based on the statstcal learnng theory. he dagram n Fg. shos the basc concepts of ths approach. he crcles and the damonds n ths fgure represent to samples, and H s a labelng lne separatng the to samples. he H and H dashed lnes pass through the nearest samples to the labelng lne. he nearest data ponts used to defne the margn are called support vectors (SV), and the dstance beteen H and H s called the margn. he

2 C.-Y. Lee and S.-G. Chern: Applcaton of a Support Vector Machne for Lquefacton Assessment 39 H: x + b = H: x + b = 0 H: x + b = - y = - Support vectors Support vectors y = + Margn = L (, b, ) = y ( x + b) P α α = = yx b y+ α α α = = = (7) here α 0 ( =,,, ) are the Lagrangan multplers. he goal here s to fnd and b hch mnmzes, and the α hch maxmzes Eq. (7). hs can be done by dfferentatng L p th respect to and b and settng the dervatves to zero Fg.. Optmal hyperplane for a lnear SVM. separatng hyperplane H that has the maxmum dstance beteen the nearest data (.e., the maxmum margn) s called the optmal separatng hyperplane. As Fg. shos, the data patterns can be shon as {x, y }, =,,, k, here x R s an -dmensonal data vector th each sample belongng to ether of the to classes labeled as y {-, +}, and the decson functon (hyperplane) can be expressed as x+ b= 0 () here x s an nput vector, s an adaptve eght vector, b s a bas, and x s an nner product of and x. For the lnearly separable class, a separatng hyperplane for the to classes can be defned as x+ b, y=+ () x+ b, y= (3) Eqs. () and (3) can be combned nto y ( x + b) 0 (4) he goal of the SVM s to fnd and b for the optmal separatng hyperplane to maxmze the margn (Fg. ). Hence, the hyperplane that optmally separates the data s the one that mnmzes. he optmal separatng hyperplane can be obtaned by solvng the follong convex quadratc optmzaton problem []: Mnmze = (5) subject to y ( x ) + b, (6) he above equaton can be transformed nto the equvalent Lagrangan dual problem as Lb (,, α) = 0 = α y x (8) = = Lb (,, α) = 0 αy = 0 (9) b Based on Eq. (9), the thrd term on the rght hand sde of Eq. (7) s zero. Multplyng Eq. (8) by leads to = α = αα j j j = = j (0) y x y y x x Eq. (7) can then be reformulated as L subject to () D ( α) = α αα jyyjx xj = = j () = α y = 0 () α 0, () he problem s no re-cast as fndng the optmum Lagrangan multplers that maxmze the objectve functon Eq. () subject to Eq. (). hs s a convex quadratc optmzaton problem, and requres a quadratc program (QP) solver that returns α ι. he soluton α ι for the dual optmzaton problem determnes the parameter * and b * of the optmal hyperplane. hus, the optmal hyperplane decson functon can be rtten as * * * * f( x) = sgn( x + b ) = sgn α y x xj + b (3) = here sgn s the sgnum functon. If the result s postve, then t s classfed x as class, and classfed as class otherse.. Lnearly Inseparable SVM he soft margn method, hch ntroduces an addtonal cost functon assocated th msclassfcaton, s an appro-

3 30 Journal of Marne Scence and echnology, Vol., o. 3 (03) X x φ z 3 F x - Orgn b - ξ z z Fg.. Hyperplane through to lnearly nseparab classes. Fg. 3. Mappng from the data space X to the feature space F. prate ay to extend the SVM methodology to data that s not lnearly separable. Cortes and Vapnk [5] ntroduced postve slack varables ξ and a penalty factor C. As Fg. shos, data ponts on the ncorrect sde of the margn boundary have a penalty that ncreases th dstance. o reduce the number of msclassfcatons, modfy the constrants of Eq. (5) for the non-separable case as follos: mnmze l + C ξ (4) = subject to y ( x ) + b + ξ 0, (5) here ξ s called a slack varable used to account for the effects of msclassfcaton. C s called a penalty factor, a parameter defnes the trade-off beteen the number of msclassfcaton n the tranng data and margn maxmzaton. As before, reformulatng ths as a Lagrangan requres the mnmzaton of, b, and ξ, and the maxmzaton of α (here α 0): Lb (,, ξ, α, β) = + C ξ = y( x b) = = (6) α + + ξ + βξ subject to α, β 0 ( =,,..., ) (7) Dfferentatng L th, b, and ξ, and settng the dervatves to zero leads to Lb (,, ξ, α, β) = 0 = α y x (8) = = Lb (,, ξ, α, β) = 0 αy = 0 (9) b Lb (,, ξ, α, β ) = 0 C α β = 0 ξ (0) After substtutng these values n, L D has the same form as Eq. (0), Eq. (). Agan, maxmze L b yyxx D (,, ξ, α, β) = α αα j j j = = j subject to () () = α y = 0 () 0 α C, () he equatons are almost the same dual problem as before, th a slght dfference beng that the multplers α have an extra constrant. 3. onlnear Separable SVM he concepts can also be extended to the case of a nonlnear separatng hyperplane by mappng the nput space onto a hgh dmensonal space, x φ( x), here the data can be lnearly classfed (Fg. 3). he key property of ths mappng s that the functon φ must be subject to the condton that the dot product of the to functons φ(x ) φ(x j ) can be rtten as a kernel functon K(x, x j ) he decson functon then becomes f( x) = yαk( x, xj) + b (3) = Dfferent kernel functons can construct varous learnng machnes. Some typcal kernel functons are as follos: Lnear kernel: K( x, x ) = x x (4) j j d Polynomal kernel: K( x, x ) = ( γx x + r), γ > 0 (5) Radal bass functon (RBF): j j ( γ ) K( x, x ) = exp x x, γ > 0 (6) j j

4 C.-Y. Lee and S.-G. Chern: Applcaton of a Support Vector Machne for Lquefacton Assessment 3 Randomly Mxed Data Dvson nto k-folds st Learnng est Learnng Learnng Learnng st Learnng Learnng est Learnng Learnng Error Rate Accuracy rate % Accuracy rate % Fg. 4. A k-fold cross-valdaton procedure. Sgmod kernel: K( x, x ) = tan( γ x x + r) (7) j j In the questons above, γ, r and d are kernel parameters. III. CROSS-VALIDAIO Cross-valdaton s a technque for assessng ho the results of statstcal analyss can be generalzed to an ndependent dataset. hs technque s manly used n stuatons here the goal s predcton, and one ants to estmate ho accurately a predctve model ll perform n practce. hs study adopts a k-fold cross-valdaton technque that randomly parttons the orgnal sample nto k subsamples. A sngle subsample s retaned as valdaton data for testng the model, and the remanng k subsamples are used as tranng data. he cross-valdaton process s repeated k tmes (the folds), th each of the k subsamples used exactly once as the valdaton data. he k results from the folds can be averaged (or otherse combned) to produce a sngle estmaton. Fg. 4 provdes an example of a k-fold cross-valdaton procedure. he advantage of ths method over repeated random subsamplng s that all observatons are used for both tranng and valdaton, and each observaton s used for valdaton exactly once. he man draback of ths method s that t requres ntense computaton. Fg. 5 shos the k-fold cross-valdaton error versus k for a bg data set, and ndcates that a k value beteen 4 and 0 s a good trade-off: ncreasng ths value sgnfcantly ncreases computaton tme and does not sgnfcantly mprove results []. hus, ths study adopts 5-fold crossvaldaton. hs approach may not be useful n achevng hgh tranng accuracy, but t can prevent the over-fttng problem. IV. GRID SEARCH he grd search algorthm performs an exhaustve search through the parameter space of a learnng algorthm to solve the problem of model selecton (.e., fndng the optmal parameters for a dataset) k Fg. 5. A plot of k-fold cross-valdaton error vs. k []. Researchers have proposed four basc kernel functons for SVM models. Frst, decde hch one to use, and then choose the penalty C and kernel parameters. For example, there are to parameters for an RBF kernel: C and γ. Varous pars of (C, γ) values are tred th a grd search procedure and the one th the best cross-valdaton accuracy s chosen. estng exponentally grong sequences of C and γ s a practcal method for dentfyng good parameters (e.g., C = -4, -3.5,, 4 ; γ = -4, -3.5,, 4 ). V. APPLICAIOS OF SVM CLASSIFICAIO he case records n ths study ere evaluated usng the MALAB (R00a) program and tool box [, 6]. Fg. 6 shos the flochart of the proposed SVM system. he database ncludes 466 CP-based feld lquefacton records from over major earthquakes beteen 964 and 999. he data conssts of case records from Japan, 85 from Chna, 7 from Canada, 9 from the USA, and 34 from aan. hs represents 50 stes that lquefed and 6 stes that dd not lquefy. Fve parameters that ere recorded n all 466 stes are () earthquake magntude, M; () total overburden pressure, σ 0 ; (3) effectve overburden pressure, σ 0 ; (4) q c values from CPs; and (5) peak acceleraton, a max able summarzes the maxmum and mnmum values of each parameter. he parameter values for all 466 case records are presented n a paper rtten by Chern et al. [3]. he nput representng the lquefacton potental s gven a bnary value of for lquefed stes and a value of for non-lquefed stes. Before the datasets ere used to tran the SVM model, they ere preprocessed usng Eq. (8). Each parameter s normalzed beteen 0 and, th x x y = x max mn xmn (8) n hch y s a normalzed nput parameter, x s the orgnal nput parameter, and x max and x mn are the maxmum and mnmum parameters, respectvely.

5 3 Journal of Marne Scence and echnology, Vol., o. 3 (03) able. he maxmum and mnmum values of the reference datasets. Parameter M σ 0 (kpa) σ 0 (kpa) q c (kpa) a max (kpa) Max Mn ranng Data Decson Kernel k-fold Cross-Valdaton Data Preprocessng ormalzaton estng Data Accuracy (%) Cross-Valdaton Accuracy = 95.79% Best C = Best gamma = 3 Best C, gamma 5 0 logg logc Fg. 7. Parameters C and γ versus the accuracy rate. o mprove SVM classfcaton accuracy, the grd search procedure plays an mportant role n the performance of the SVM. Fg. 7 also shos that the parameters C and γ greatly affect the classfcaton accuracy of the SVM. 5 Learnng set est set VI. RESULS AD DISCUSSIO he procedure for usng the SVM s descrbed belo: Grd Search Best Parameter ranng Data SVM Model Accuracy Rate % Fg. 6. Flochart of the proposed SVM system. he man advantage of scalng s to avod attrbutes th greater numerc ranges domnatng those th smaller numerc ranges. Another advantage of ths method s to avod numercal dffcultes durng calculaton. A normal SVM model randomly selects kernel parameters usng a tral-and-error method [8, 0]. he grd search approach n ths study s an alternatve ay to fnd the best parameters for the SVM classfer. hs approach avods the over-fttng problem of the SVM model occurrng because of the mproper determnaton of these parameters. he RBF kernel s a reasonable frst choce for an SVM model [9]. Hence, the proposed SVM model as frst constructed by a radal bass functon (RBF) kernel. here are to parameters, C and γ, to be determned. After the grd search procedure, the optmal parameters th maxmal classfcaton accuracy ere selected. As shon n Fg. 7, the best (C, γ) s (.5, 5 ) th a cross-valdaton rate of 95.79%. In ths result, the optmal parameters are used to tran the SVM model to generate the fnal classfer.. ransform data to the format of the SVM package.. Conduct smple scalng on the data. 3. Consder the RBF kernel. 4. Use cross-valdaton and grd searchng to fnd the best parameters C and γ. 5. Use the best C and γ to tran the hole tranng set. 6. est. 7. Fnd the best accuracy rate. After the tranng procedure, the best (C, γ) s (.5, 5 ) th a cross-valdaton rate of 95.79%. Out of the 466 datasets used, only 6 cases ere msclassfed, achevng an overall classfcaton accuracy rate of 98.7%. In addton to verfyng the effectveness of the proposed method, ths study compares t th an A method n the reference [3, 4]. he A model proposed n that paper combnes fuzzy theory th a subtractve clusterng algorthm to form a fuzzy-neural netork system. o verfy the feasblty of the A model, ths study compares that A model th the B5 model employed n Goh [7] usng the same 09 data groups, ncludng 74 tranng data groups and 35 test data groups. he results of ths analyss are presented n able. he A-G5 model [3] performs better than Goh s optmal B5 model n both the tranng and testng segments. herefore, the 466 collected CP datasets are used n ths study to compare the SVM model th the A (C4, C4H6, C5, and C5) models [3]. Results are lsted n able 3, t shos that the SVM model acheves better results than the A models because of ts loer total error rate of.9%.

6 C.-Y. Lee and S.-G. Chern: Applcaton of a Support Vector Machne for Lquefacton Assessment 33 able. Result of the G5 model and B5 model [7]. Model Input varables o. of elements n every o. of Error o. of hdden neurons tranng cluster ranng estng otal error rate (%) B5 M, D 50, σ 0, q c, a max 5.75 G5 M, D 50, σ 0, q c, a max 53, 3, able 3. Comparson beteen SVM and A models [3]. Model Input varables o. of elements n every tranng cluster o. of hdden neurons o. of Msclassfed otal error rate (%) C4 M, σ 0, q c, a max 7,6,8, C4H6 M, σ 0, q c, a max 7,6,8, C5 M, σ 0, σ 0, q c, a max 90,93,4, C5 M, σ 0, σ 0, qc, a max 90,9,3, SVM M, σ 0, σ 0, q c, a max 6.9 able 4. he classfcaton accuraces versus C for dfferent kernel functons. C Lnear Poly RBF Sgmod As ndcated prevously, there are four types of basc kernel functons: lnear, RBF, second order polynomal, and sgmod. hs study employs the accuracy rate as a crteron to fnd the optmal kernel functon. able 4 shos the accuracy rate versus the C parameter from 0.0 to 00 for dfferent kernel functons. he RBF kernel functon th parameter C =.5 provdes the best performance for the SVM model. he excellent classfcaton accuracy of an SVM suggests ts practcalty for engneerng applcatons. herefore, ths study develops a lquefacton assessment algorthm based on SVM theory, called LA-SVM. he graphcal user nterface (GUI) of ths algorthm as mplemented n a MALAB/GUI. hs nterface provdes an ntutve and user-frendly means of nteracton. Users do not need any dagrams, formulae, or manuals. By smply usng a mouse cursor to select optons and nput tranng data and parameter ranges, they can receve the classfcaton results and accuraces of the testng data n a short CPU runtme. LA-SVM greatly smplfes the lquefacton assessment process and produces extremely accurate results. he operaton steps are lsted as follos:. Launch LA-SVM program (Fg. 8).. Select the nput button, and nput the tranng data and testng data. 3. Select the data s normalzaton and ts ranges (specfed by users). 4. Select the grd search method and specfy the parameter ranges. Fg. 8. Easy to use LA-SVM/GUI nterface. Fg. 9. Expermental Result of LA-SVM. 5. Specfy the number of folds for cross-valdaton. 6. Select the run button, and start the analyss. 7. When analyss s complete, obtan the optmal kernel functon parameters and the classfcaton accuracy (Fg. 9).

7 34 Journal of Marne Scence and echnology, Vol., o. 3 (03) VII. COCLUSIO SVM has been successfully appled n many applcatons, but t s less dely appled n the geotechncal feld. he results n ths study sho that SVM s a poerful computatonal tool that can be used to analyze the complex relatonshp beteen sol and sesmc parameters n lquefacton assessment. he expermental results n ths study ndcate that an SVM acheves greater classfcaton accuracy than an A. In addton to ts hgher accuracy rate, the SVM model requres only to parameters, as compared to the A, hch requres multple parameters. In concluson, the SVM model s more effectve and feasble than the conventonal A. An SVM not only has a sold foundaton n statstcal learnng theory, but can also effectvely handle nonlnear classfcaton. herefore, t s regarded as one of the most effectve classfcaton methods. he expermental results and dscusson above sho that the proposed LA-SVM can be effectvely appled to lquefacton assessment. he LA-SVM program has an ntutve nterface that s easly understandable. As ne lquefacton assessment cases are collected to expand the database, the classfcaton accuracy of LA-SVM can be further ncreased. hus, LA-SVM s a novel lquefacton assessment tool orthy of promoton and support. REFERECES. Boser, B. E., Guyon, I. M., and Vapnk, V.., A tranng algorthm for optmal margn classers, Proceedngs of the Ffth Annual Workshop on Computatonal Learnng heory, ACM Press, pp (99).. Chang, C. C. and Ln, C. J., LIBSVM: A Lbrary for Support Vector Machnes, (00). 3. Chern, S. G., Lee, C. Y., and Wang, C. C., CP-based lquefacton assessment by usng fuzzy-neural netork, Journal of Marne Scence and echnology, Vol. 6, o., pp (008). 4. Chern, S. G., Lee, C. Y., and Wang, C. C., CP-based smplfed lquefacton assessment by usng fuzzy-neural netork, Journal of Marne Scence and echnology, Vol. 7, o. 4, pp (009). 5. Cortes, C. and Vapnk, V., Support-vector netork, Machne Learnng, Vol. 0, pp (995). 6. Faruto and Lyang, LIBSVM-faruto Ult- mate Verson, a toolbox th mplement for support vector machne based on lbsvm, matlabsky.com (0). 7. Goh, A.. C., eural netork modelng of CP sesmc lquefacton data, Journal of Geotechncal Engneerng, ASCE, Vol. 9, o., pp (996). 8. Goh, A.. C. and Goh, S. H., Support vector machnes: ther use n geotechncal engneerng as llustrated usng sesmc lquefacton data, Computer and Geotechncs, Vol. 34, pp (007). 9. Hsu, C. C., Chang, C. C., and Ln, C. J., A Practcal Gude to Support Vector Classfcaton, ape (00). 0. Samu, P. and Stharam,. G., Machne learnng modelng for predctng sol lquefacton susceptblty, Journal of atural Hazard and Earth System Scence, Vol., pp. -9 (0).. Sterln, P., Overfttng Preventon th Cross-Valdaton, Pars (007).. Vapnk, V.., he ature of Statstcal Learnng heory, Sprnger, e York (995).

Support Vector Machines

Support Vector Machines /9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.

More information

Classification / Regression Support Vector Machines

Classification / Regression Support Vector Machines Classfcaton / Regresson Support Vector Machnes Jeff Howbert Introducton to Machne Learnng Wnter 04 Topcs SVM classfers for lnearly separable classes SVM classfers for non-lnearly separable classes SVM

More information

Support Vector Machines

Support Vector Machines Support Vector Machnes Decson surface s a hyperplane (lne n 2D) n feature space (smlar to the Perceptron) Arguably, the most mportant recent dscovery n machne learnng In a nutshell: map the data to a predetermned

More information

Support Vector Machines. CS534 - Machine Learning

Support Vector Machines. CS534 - Machine Learning Support Vector Machnes CS534 - Machne Learnng Perceptron Revsted: Lnear Separators Bnar classfcaton can be veed as the task of separatng classes n feature space: b > 0 b 0 b < 0 f() sgn( b) Lnear Separators

More information

CS246: Mining Massive Datasets Jure Leskovec, Stanford University

CS246: Mining Massive Datasets Jure Leskovec, Stanford University CS46: Mnng Massve Datasets Jure Leskovec, Stanford Unversty http://cs46.stanford.edu /19/013 Jure Leskovec, Stanford CS46: Mnng Massve Datasets, http://cs46.stanford.edu Perceptron: y = sgn( x Ho to fnd

More information

Learning the Kernel Parameters in Kernel Minimum Distance Classifier

Learning the Kernel Parameters in Kernel Minimum Distance Classifier Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department

More information

12/2/2009. Announcements. Parametric / Non-parametric. Case-Based Reasoning. Nearest-Neighbor on Images. Nearest-Neighbor Classification

12/2/2009. Announcements. Parametric / Non-parametric. Case-Based Reasoning. Nearest-Neighbor on Images. Nearest-Neighbor Classification Introducton to Artfcal Intellgence V22.0472-001 Fall 2009 Lecture 24: Nearest-Neghbors & Support Vector Machnes Rob Fergus Dept of Computer Scence, Courant Insttute, NYU Sldes from Danel Yeung, John DeNero

More information

Machine Learning. Support Vector Machines. (contains material adapted from talks by Constantin F. Aliferis & Ioannis Tsamardinos, and Martin Law)

Machine Learning. Support Vector Machines. (contains material adapted from talks by Constantin F. Aliferis & Ioannis Tsamardinos, and Martin Law) Machne Learnng Support Vector Machnes (contans materal adapted from talks by Constantn F. Alfers & Ioanns Tsamardnos, and Martn Law) Bryan Pardo, Machne Learnng: EECS 349 Fall 2014 Support Vector Machnes

More information

Outline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1

Outline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1 4/14/011 Outlne Dscrmnatve classfers for mage recognton Wednesday, Aprl 13 Krsten Grauman UT-Austn Last tme: wndow-based generc obect detecton basc ppelne face detecton wth boostng as case study Today:

More information

The Research of Support Vector Machine in Agricultural Data Classification

The Research of Support Vector Machine in Agricultural Data Classification The Research of Support Vector Machne n Agrcultural Data Classfcaton Le Sh, Qguo Duan, Xnmng Ma, Me Weng College of Informaton and Management Scence, HeNan Agrcultural Unversty, Zhengzhou 45000 Chna Zhengzhou

More information

Machine Learning 9. week

Machine Learning 9. week Machne Learnng 9. week Mappng Concept Radal Bass Functons (RBF) RBF Networks 1 Mappng It s probably the best scenaro for the classfcaton of two dataset s to separate them lnearly. As you see n the below

More information

BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET

BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET 1 BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET TZU-CHENG CHUANG School of Electrcal and Computer Engneerng, Purdue Unversty, West Lafayette, Indana 47907 SAUL B. GELFAND School

More information

INF 4300 Support Vector Machine Classifiers (SVM) Anne Solberg

INF 4300 Support Vector Machine Classifiers (SVM) Anne Solberg INF 43 Support Vector Machne Classfers (SVM) Anne Solberg (anne@f.uo.no) 9..7 Lnear classfers th mamum margn for toclass problems The kernel trck from lnear to a hghdmensonal generalzaton Generaton from

More information

Edge Detection in Noisy Images Using the Support Vector Machines

Edge Detection in Noisy Images Using the Support Vector Machines Edge Detecton n Nosy Images Usng the Support Vector Machnes Hlaro Gómez-Moreno, Saturnno Maldonado-Bascón, Francsco López-Ferreras Sgnal Theory and Communcatons Department. Unversty of Alcalá Crta. Madrd-Barcelona

More information

Feature Reduction and Selection

Feature Reduction and Selection Feature Reducton and Selecton Dr. Shuang LIANG School of Software Engneerng TongJ Unversty Fall, 2012 Today s Topcs Introducton Problems of Dmensonalty Feature Reducton Statstc methods Prncpal Components

More information

Determining the Optimal Bandwidth Based on Multi-criterion Fusion

Determining the Optimal Bandwidth Based on Multi-criterion Fusion Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn

More information

Incremental Learning with Support Vector Machines and Fuzzy Set Theory

Incremental Learning with Support Vector Machines and Fuzzy Set Theory The 25th Workshop on Combnatoral Mathematcs and Computaton Theory Incremental Learnng wth Support Vector Machnes and Fuzzy Set Theory Yu-Mng Chuang 1 and Cha-Hwa Ln 2* 1 Department of Computer Scence and

More information

Using Neural Networks and Support Vector Machines in Data Mining

Using Neural Networks and Support Vector Machines in Data Mining Usng eural etworks and Support Vector Machnes n Data Mnng RICHARD A. WASIOWSKI Computer Scence Department Calforna State Unversty Domnguez Hlls Carson, CA 90747 USA Abstract: - Multvarate data analyss

More information

CHAPTER 3 SEQUENTIAL MINIMAL OPTIMIZATION TRAINED SUPPORT VECTOR CLASSIFIER FOR CANCER PREDICTION

CHAPTER 3 SEQUENTIAL MINIMAL OPTIMIZATION TRAINED SUPPORT VECTOR CLASSIFIER FOR CANCER PREDICTION 48 CHAPTER 3 SEQUENTIAL MINIMAL OPTIMIZATION TRAINED SUPPORT VECTOR CLASSIFIER FOR CANCER PREDICTION 3.1 INTRODUCTION The raw mcroarray data s bascally an mage wth dfferent colors ndcatng hybrdzaton (Xue

More information

S1 Note. Basis functions.

S1 Note. Basis functions. S1 Note. Bass functons. Contents Types of bass functons...1 The Fourer bass...2 B-splne bass...3 Power and type I error rates wth dfferent numbers of bass functons...4 Table S1. Smulaton results of type

More information

Discriminative classifiers for object classification. Last time

Discriminative classifiers for object classification. Last time Dscrmnatve classfers for object classfcaton Thursday, Nov 12 Krsten Grauman UT Austn Last tme Supervsed classfcaton Loss and rsk, kbayes rule Skn color detecton example Sldng ndo detecton Classfers, boostng

More information

Smoothing Spline ANOVA for variable screening

Smoothing Spline ANOVA for variable screening Smoothng Splne ANOVA for varable screenng a useful tool for metamodels tranng and mult-objectve optmzaton L. Rcco, E. Rgon, A. Turco Outlne RSM Introducton Possble couplng Test case MOO MOO wth Game Theory

More information

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,

More information

Face Recognition Based on SVM and 2DPCA

Face Recognition Based on SVM and 2DPCA Vol. 4, o. 3, September, 2011 Face Recognton Based on SVM and 2DPCA Tha Hoang Le, Len Bu Faculty of Informaton Technology, HCMC Unversty of Scence Faculty of Informaton Scences and Engneerng, Unversty

More information

Parallelism for Nested Loops with Non-uniform and Flow Dependences

Parallelism for Nested Loops with Non-uniform and Flow Dependences Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr

More information

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana

More information

Classifier Selection Based on Data Complexity Measures *

Classifier Selection Based on Data Complexity Measures * Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.

More information

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth

More information

Discrimination of Faulted Transmission Lines Using Multi Class Support Vector Machines

Discrimination of Faulted Transmission Lines Using Multi Class Support Vector Machines 16th NAIONAL POWER SYSEMS CONFERENCE, 15th-17th DECEMBER, 2010 497 Dscrmnaton of Faulted ransmsson Lnes Usng Mult Class Support Vector Machnes D.hukaram, Senor Member IEEE, and Rmjhm Agrawal Abstract hs

More information

Announcements. Supervised Learning

Announcements. Supervised Learning Announcements See Chapter 5 of Duda, Hart, and Stork. Tutoral by Burge lnked to on web page. Supervsed Learnng Classfcaton wth labeled eamples. Images vectors n hgh-d space. Supervsed Learnng Labeled eamples

More information

The Study of Remote Sensing Image Classification Based on Support Vector Machine

The Study of Remote Sensing Image Classification Based on Support Vector Machine Sensors & Transducers 03 by IFSA http://www.sensorsportal.com The Study of Remote Sensng Image Classfcaton Based on Support Vector Machne, ZHANG Jan-Hua Key Research Insttute of Yellow Rver Cvlzaton and

More information

SVM-based Learning for Multiple Model Estimation

SVM-based Learning for Multiple Model Estimation SVM-based Learnng for Multple Model Estmaton Vladmr Cherkassky and Yunqan Ma Department of Electrcal and Computer Engneerng Unversty of Mnnesota Mnneapols, MN 55455 {cherkass,myq}@ece.umn.edu Abstract:

More information

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task Proceedngs of NTCIR-6 Workshop Meetng, May 15-18, 2007, Tokyo, Japan Term Weghtng Classfcaton System Usng the Ch-square Statstc for the Classfcaton Subtask at NTCIR-6 Patent Retreval Task Kotaro Hashmoto

More information

Proper Choice of Data Used for the Estimation of Datum Transformation Parameters

Proper Choice of Data Used for the Estimation of Datum Transformation Parameters Proper Choce of Data Used for the Estmaton of Datum Transformaton Parameters Hakan S. KUTOGLU, Turkey Key words: Coordnate systems; transformaton; estmaton, relablty. SUMMARY Advances n technologes and

More information

Lecture 5: Multilayer Perceptrons

Lecture 5: Multilayer Perceptrons Lecture 5: Multlayer Perceptrons Roger Grosse 1 Introducton So far, we ve only talked about lnear models: lnear regresson and lnear bnary classfers. We noted that there are functons that can t be represented

More information

Optimal Design of Nonlinear Fuzzy Model by Means of Independent Fuzzy Scatter Partition

Optimal Design of Nonlinear Fuzzy Model by Means of Independent Fuzzy Scatter Partition Optmal Desgn of onlnear Fuzzy Model by Means of Independent Fuzzy Scatter Partton Keon-Jun Park, Hyung-Kl Kang and Yong-Kab Km *, Department of Informaton and Communcaton Engneerng, Wonkwang Unversty,

More information

2x x l. Module 3: Element Properties Lecture 4: Lagrange and Serendipity Elements

2x x l. Module 3: Element Properties Lecture 4: Lagrange and Serendipity Elements Module 3: Element Propertes Lecture : Lagrange and Serendpty Elements 5 In last lecture note, the nterpolaton functons are derved on the bass of assumed polynomal from Pascal s trangle for the fled varable.

More information

Biostatistics 615/815

Biostatistics 615/815 The E-M Algorthm Bostatstcs 615/815 Lecture 17 Last Lecture: The Smplex Method General method for optmzaton Makes few assumptons about functon Crawls towards mnmum Some recommendatons Multple startng ponts

More information

Spam Filtering Based on Support Vector Machines with Taguchi Method for Parameter Selection

Spam Filtering Based on Support Vector Machines with Taguchi Method for Parameter Selection E-mal Spam Flterng Based on Support Vector Machnes wth Taguch Method for Parameter Selecton We-Chh Hsu, Tsan-Yng Yu E-mal Spam Flterng Based on Support Vector Machnes wth Taguch Method for Parameter Selecton

More information

A Modified Median Filter for the Removal of Impulse Noise Based on the Support Vector Machines

A Modified Median Filter for the Removal of Impulse Noise Based on the Support Vector Machines A Modfed Medan Flter for the Removal of Impulse Nose Based on the Support Vector Machnes H. GOMEZ-MORENO, S. MALDONADO-BASCON, F. LOPEZ-FERRERAS, M. UTRILLA- MANSO AND P. GIL-JIMENEZ Departamento de Teoría

More information

Fault Diagnosis of Sucker-Rod Pumping System Using Support Vector Machine

Fault Diagnosis of Sucker-Rod Pumping System Using Support Vector Machine Fault Dagnoss of Sucker-Rod Pumpng System Usng Support Vector Machne Jln Feng, Maofa Wang,, Yheng Yang 2, Fangpng Gao, Zhan Pan, Wefeng Shan, Qngje Lu, Quge Yang, and Jng Yuan Department of Informaton

More information

A Binarization Algorithm specialized on Document Images and Photos

A Binarization Algorithm specialized on Document Images and Photos A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a

More information

Face Recognition Method Based on Within-class Clustering SVM

Face Recognition Method Based on Within-class Clustering SVM Face Recognton Method Based on Wthn-class Clusterng SVM Yan Wu, Xao Yao and Yng Xa Department of Computer Scence and Engneerng Tong Unversty Shangha, Chna Abstract - A face recognton method based on Wthn-class

More information

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 A mathematcal programmng approach to the analyss, desgn and

More information

NAG Fortran Library Chapter Introduction. G10 Smoothing in Statistics

NAG Fortran Library Chapter Introduction. G10 Smoothing in Statistics Introducton G10 NAG Fortran Lbrary Chapter Introducton G10 Smoothng n Statstcs Contents 1 Scope of the Chapter... 2 2 Background to the Problems... 2 2.1 Smoothng Methods... 2 2.2 Smoothng Splnes and Regresson

More information

Training of Kernel Fuzzy Classifiers by Dynamic Cluster Generation

Training of Kernel Fuzzy Classifiers by Dynamic Cluster Generation Tranng of Kernel Fuzzy Classfers by Dynamc Cluster Generaton Shgeo Abe Graduate School of Scence and Technology Kobe Unversty Nada, Kobe, Japan abe@eedept.kobe-u.ac.jp Abstract We dscuss kernel fuzzy classfers

More information

Collaboratively Regularized Nearest Points for Set Based Recognition

Collaboratively Regularized Nearest Points for Set Based Recognition Academc Center for Computng and Meda Studes, Kyoto Unversty Collaboratvely Regularzed Nearest Ponts for Set Based Recognton Yang Wu, Mchhko Mnoh, Masayuk Mukunok Kyoto Unversty 9/1/013 BMVC 013 @ Brstol,

More information

Face Recognition University at Buffalo CSE666 Lecture Slides Resources:

Face Recognition University at Buffalo CSE666 Lecture Slides Resources: Face Recognton Unversty at Buffalo CSE666 Lecture Sldes Resources: http://www.face-rec.org/algorthms/ Overvew of face recognton algorthms Correlaton - Pxel based correspondence between two face mages Structural

More information

Efficient Text Classification by Weighted Proximal SVM *

Efficient Text Classification by Weighted Proximal SVM * Effcent ext Classfcaton by Weghted Proxmal SVM * Dong Zhuang 1, Benyu Zhang, Qang Yang 3, Jun Yan 4, Zheng Chen, Yng Chen 1 1 Computer Scence and Engneerng, Bejng Insttute of echnology, Bejng 100081, Chna

More information

Classifying Acoustic Transient Signals Using Artificial Intelligence

Classifying Acoustic Transient Signals Using Artificial Intelligence Classfyng Acoustc Transent Sgnals Usng Artfcal Intellgence Steve Sutton, Unversty of North Carolna At Wlmngton (suttons@charter.net) Greg Huff, Unversty of North Carolna At Wlmngton (jgh7476@uncwl.edu)

More information

Novel Fuzzy logic Based Edge Detection Technique

Novel Fuzzy logic Based Edge Detection Technique Novel Fuzzy logc Based Edge Detecton Technque Aborsade, D.O Department of Electroncs Engneerng, adoke Akntola Unversty of Tech., Ogbomoso. Oyo-state. doaborsade@yahoo.com Abstract Ths paper s based on

More information

X- Chart Using ANOM Approach

X- Chart Using ANOM Approach ISSN 1684-8403 Journal of Statstcs Volume 17, 010, pp. 3-3 Abstract X- Chart Usng ANOM Approach Gullapall Chakravarth 1 and Chaluvad Venkateswara Rao Control lmts for ndvdual measurements (X) chart are

More information

Meta-heuristics for Multidimensional Knapsack Problems

Meta-heuristics for Multidimensional Knapsack Problems 2012 4th Internatonal Conference on Computer Research and Development IPCSIT vol.39 (2012) (2012) IACSIT Press, Sngapore Meta-heurstcs for Multdmensonal Knapsack Problems Zhbao Man + Computer Scence Department,

More information

Support Vector Machines for Business Applications

Support Vector Machines for Business Applications Support Vector Machnes for Busness Applcatons Bran C. Lovell and Chrstan J Walder The Unversty of Queensland and Max Planck Insttute, Tübngen {lovell, walder}@tee.uq.edu.au Introducton Recent years have

More information

Sum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints

Sum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints Australan Journal of Basc and Appled Scences, 2(4): 1204-1208, 2008 ISSN 1991-8178 Sum of Lnear and Fractonal Multobjectve Programmng Problem under Fuzzy Rules Constrants 1 2 Sanjay Jan and Kalash Lachhwan

More information

y and the total sum of

y and the total sum of Lnear regresson Testng for non-lnearty In analytcal chemstry, lnear regresson s commonly used n the constructon of calbraton functons requred for analytcal technques such as gas chromatography, atomc absorpton

More information

An Improved Spectral Clustering Algorithm Based on Local Neighbors in Kernel Space 1

An Improved Spectral Clustering Algorithm Based on Local Neighbors in Kernel Space 1 DOI: 10.98/CSIS110415064L An Improved Spectral Clusterng Algorthm Based on Local Neghbors n Kernel Space 1 Xnyue Lu 1,, Xng Yong and Hongfe Ln 1 1 School of Computer Scence and Technology, Dalan Unversty

More information

CLASSIFICATION OF ULTRASONIC SIGNALS

CLASSIFICATION OF ULTRASONIC SIGNALS The 8 th Internatonal Conference of the Slovenan Socety for Non-Destructve Testng»Applcaton of Contemporary Non-Destructve Testng n Engneerng«September -3, 5, Portorož, Slovena, pp. 7-33 CLASSIFICATION

More information

Simulation: Solving Dynamic Models ABE 5646 Week 11 Chapter 2, Spring 2010

Simulation: Solving Dynamic Models ABE 5646 Week 11 Chapter 2, Spring 2010 Smulaton: Solvng Dynamc Models ABE 5646 Week Chapter 2, Sprng 200 Week Descrpton Readng Materal Mar 5- Mar 9 Evaluatng [Crop] Models Comparng a model wth data - Graphcal, errors - Measures of agreement

More information

Solving two-person zero-sum game by Matlab

Solving two-person zero-sum game by Matlab Appled Mechancs and Materals Onlne: 2011-02-02 ISSN: 1662-7482, Vols. 50-51, pp 262-265 do:10.4028/www.scentfc.net/amm.50-51.262 2011 Trans Tech Publcatons, Swtzerland Solvng two-person zero-sum game by

More information

Relevance Feedback Document Retrieval using Non-Relevant Documents

Relevance Feedback Document Retrieval using Non-Relevant Documents Relevance Feedback Document Retreval usng Non-Relevant Documents TAKASHI ONODA, HIROSHI MURATA and SEIJI YAMADA Ths paper reports a new document retreval method usng non-relevant documents. From a large

More information

Lobachevsky State University of Nizhni Novgorod. Polyhedron. Quick Start Guide

Lobachevsky State University of Nizhni Novgorod. Polyhedron. Quick Start Guide Lobachevsky State Unversty of Nzhn Novgorod Polyhedron Quck Start Gude Nzhn Novgorod 2016 Contents Specfcaton of Polyhedron software... 3 Theoretcal background... 4 1. Interface of Polyhedron... 6 1.1.

More information

Data Mining: Model Evaluation

Data Mining: Model Evaluation Data Mnng: Model Evaluaton Aprl 16, 2013 1 Issues: Evaluatng Classfcaton Methods Accurac classfer accurac: predctng class label predctor accurac: guessng value of predcted attrbutes Speed tme to construct

More information

Categories and Subject Descriptors B.7.2 [Integrated Circuits]: Design Aids Verification. General Terms Algorithms

Categories and Subject Descriptors B.7.2 [Integrated Circuits]: Design Aids Verification. General Terms Algorithms 3. Fndng Determnstc Soluton from Underdetermned Equaton: Large-Scale Performance Modelng by Least Angle Regresson Xn L ECE Department, Carnege Mellon Unversty Forbs Avenue, Pttsburgh, PA 3 xnl@ece.cmu.edu

More information

An Optimal Algorithm for Prufer Codes *

An Optimal Algorithm for Prufer Codes * J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,

More information

SUMMARY... I TABLE OF CONTENTS...II INTRODUCTION...

SUMMARY... I TABLE OF CONTENTS...II INTRODUCTION... Summary A follow-the-leader robot system s mplemented usng Dscrete-Event Supervsory Control methods. The system conssts of three robots, a leader and two followers. The dea s to get the two followers to

More information

Relevance Assignment and Fusion of Multiple Learning Methods Applied to Remote Sensing Image Analysis

Relevance Assignment and Fusion of Multiple Learning Methods Applied to Remote Sensing Image Analysis Assgnment and Fuson of Multple Learnng Methods Appled to Remote Sensng Image Analyss Peter Bajcsy, We-Wen Feng and Praveen Kumar Natonal Center for Supercomputng Applcaton (NCSA), Unversty of Illnos at

More information

Complex System Reliability Evaluation using Support Vector Machine for Incomplete Data-set

Complex System Reliability Evaluation using Support Vector Machine for Incomplete Data-set Internatonal Journal of Performablty Engneerng, Vol. 7, No. 1, January 2010, pp.32-42. RAMS Consultants Prnted n Inda Complex System Relablty Evaluaton usng Support Vector Machne for Incomplete Data-set

More information

A New Approach For the Ranking of Fuzzy Sets With Different Heights

A New Approach For the Ranking of Fuzzy Sets With Different Heights New pproach For the ankng of Fuzzy Sets Wth Dfferent Heghts Pushpnder Sngh School of Mathematcs Computer pplcatons Thapar Unversty, Patala-7 00 Inda pushpndersnl@gmalcom STCT ankng of fuzzy sets plays

More information

Taxonomy of Large Margin Principle Algorithms for Ordinal Regression Problems

Taxonomy of Large Margin Principle Algorithms for Ordinal Regression Problems Taxonomy of Large Margn Prncple Algorthms for Ordnal Regresson Problems Amnon Shashua Computer Scence Department Stanford Unversty Stanford, CA 94305 emal: shashua@cs.stanford.edu Anat Levn School of Computer

More information

Classification of Face Images Based on Gender using Dimensionality Reduction Techniques and SVM

Classification of Face Images Based on Gender using Dimensionality Reduction Techniques and SVM Classfcaton of Face Images Based on Gender usng Dmensonalty Reducton Technques and SVM Fahm Mannan 260 266 294 School of Computer Scence McGll Unversty Abstract Ths report presents gender classfcaton based

More information

Open Access Recognition of Oil Shale Based on LIBSVM Optimized by Modified Genetic Algorithm

Open Access Recognition of Oil Shale Based on LIBSVM Optimized by Modified Genetic Algorithm Send Orders for Reprnts to reprnts@benthamscence.ae The Open Petroleum Engneerng Journal, 05, 8, 363-367 363 Open Access Recognton of Ol Shale Based on LIBSVM Optmzed by Modfed Genetc Algorthm Qhua Hu,*,

More information

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 An Iteratve Soluton Approach to Process Plant Layout usng Mxed

More information

Machine Learning. Topic 6: Clustering

Machine Learning. Topic 6: Clustering Machne Learnng Topc 6: lusterng lusterng Groupng data nto (hopefully useful) sets. Thngs on the left Thngs on the rght Applcatons of lusterng Hypothess Generaton lusters mght suggest natural groups. Hypothess

More information

A Background Subtraction for a Vision-based User Interface *

A Background Subtraction for a Vision-based User Interface * A Background Subtracton for a Vson-based User Interface * Dongpyo Hong and Woontack Woo KJIST U-VR Lab. {dhon wwoo}@kjst.ac.kr Abstract In ths paper, we propose a robust and effcent background subtracton

More information

Support Vector classifiers for Land Cover Classification

Support Vector classifiers for Land Cover Classification Map Inda 2003 Image Processng & Interpretaton Support Vector classfers for Land Cover Classfcaton Mahesh Pal Paul M. Mather Lecturer, department of Cvl engneerng Prof., School of geography Natonal Insttute

More information

The Comparison of Calibration Method of Binocular Stereo Vision System Ke Zhang a *, Zhao Gao b

The Comparison of Calibration Method of Binocular Stereo Vision System Ke Zhang a *, Zhao Gao b 3rd Internatonal Conference on Materal, Mechancal and Manufacturng Engneerng (IC3ME 2015) The Comparson of Calbraton Method of Bnocular Stereo Vson System Ke Zhang a *, Zhao Gao b College of Engneerng,

More information

General Vector Machine. Hong Zhao Department of Physics, Xiamen University

General Vector Machine. Hong Zhao Department of Physics, Xiamen University General Vector Machne Hong Zhao (zhaoh@xmu.edu.cn) Department of Physcs, Xamen Unversty The support vector machne (SVM) s an mportant class of learnng machnes for functon approach, pattern recognton, and

More information

Learning a Class-Specific Dictionary for Facial Expression Recognition

Learning a Class-Specific Dictionary for Facial Expression Recognition BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 16, No 4 Sofa 016 Prnt ISSN: 1311-970; Onlne ISSN: 1314-4081 DOI: 10.1515/cat-016-0067 Learnng a Class-Specfc Dctonary for

More information

INF Repetition Anne Solberg INF

INF Repetition Anne Solberg INF INF 43 7..7 Repetton Anne Solberg anne@f.uo.no INF 43 Classfers covered Gaussan classfer k =I k = k arbtrary Knn-classfer Support Vector Machnes Recommendaton: lnear or Radal Bass Functon kernels INF 43

More information

Combination of Local Multiple Patterns and Exponential Discriminant Analysis for Facial Recognition

Combination of Local Multiple Patterns and Exponential Discriminant Analysis for Facial Recognition Sensors & ransducers 203 by IFSA http://.sensorsportal.com Combnaton of Local Multple Patterns and Exponental Dscrmnant Analyss for Facal Recognton, 2 Lfang Zhou, 2 Bn Fang, 3 Wesheng L, 3 Ldou Wang College

More information

Wishing you all a Total Quality New Year!

Wishing you all a Total Quality New Year! Total Qualty Management and Sx Sgma Post Graduate Program 214-15 Sesson 4 Vnay Kumar Kalakband Assstant Professor Operatons & Systems Area 1 Wshng you all a Total Qualty New Year! Hope you acheve Sx sgma

More information

Pattern classification of cotton yarn neps

Pattern classification of cotton yarn neps Indan Journal of Fbre & extle Research Vol. 41, September 016, pp. 70-77 Pattern classfcaton of cotton yarn neps Abul Hasnat, Anndya Ghosh a, Azzul Hoque b & Santanu Halder c Government College of Engneerng

More information

CS 534: Computer Vision Model Fitting

CS 534: Computer Vision Model Fitting CS 534: Computer Vson Model Fttng Sprng 004 Ahmed Elgammal Dept of Computer Scence CS 534 Model Fttng - 1 Outlnes Model fttng s mportant Least-squares fttng Maxmum lkelhood estmaton MAP estmaton Robust

More information

Japanese Dependency Analysis Based on Improved SVM and KNN

Japanese Dependency Analysis Based on Improved SVM and KNN Proceedngs of the 7th WSEAS Internatonal Conference on Smulaton, Modellng and Optmzaton, Bejng, Chna, September 15-17, 2007 140 Japanese Dependency Analyss Based on Improved SVM and KNN ZHOU HUIWEI and

More information

An Improved Image Segmentation Algorithm Based on the Otsu Method

An Improved Image Segmentation Algorithm Based on the Otsu Method 3th ACIS Internatonal Conference on Software Engneerng, Artfcal Intellgence, Networkng arallel/dstrbuted Computng An Improved Image Segmentaton Algorthm Based on the Otsu Method Mengxng Huang, enjao Yu,

More information

A Simple and Efficient Goal Programming Model for Computing of Fuzzy Linear Regression Parameters with Considering Outliers

A Simple and Efficient Goal Programming Model for Computing of Fuzzy Linear Regression Parameters with Considering Outliers 62626262621 Journal of Uncertan Systems Vol.5, No.1, pp.62-71, 211 Onlne at: www.us.org.u A Smple and Effcent Goal Programmng Model for Computng of Fuzzy Lnear Regresson Parameters wth Consderng Outlers

More information

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching A Fast Vsual Trackng Algorthm Based on Crcle Pxels Matchng Zhqang Hou hou_zhq@sohu.com Chongzhao Han czhan@mal.xjtu.edu.cn Ln Zheng Abstract: A fast vsual trackng algorthm based on crcle pxels matchng

More information

CS246: Mining Massive Datasets Jure Leskovec, Stanford University

CS246: Mining Massive Datasets Jure Leskovec, Stanford University CS246: Mnng Massve Datasets Jure Leskovec, Stanford Unversty http://cs246.stanford.edu 2/17/2015 Jure Leskovec, Stanford CS246: Mnng Massve Datasets, http://cs246.stanford.edu 2 Hgh dm. data Graph data

More information

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization Problem efntons and Evaluaton Crtera for Computatonal Expensve Optmzaton B. Lu 1, Q. Chen and Q. Zhang 3, J. J. Lang 4, P. N. Suganthan, B. Y. Qu 6 1 epartment of Computng, Glyndwr Unversty, UK Faclty

More information

Human Face Recognition Using Generalized. Kernel Fisher Discriminant

Human Face Recognition Using Generalized. Kernel Fisher Discriminant Human Face Recognton Usng Generalzed Kernel Fsher Dscrmnant ng-yu Sun,2 De-Shuang Huang Ln Guo. Insttute of Intellgent Machnes, Chnese Academy of Scences, P.O.ox 30, Hefe, Anhu, Chna. 2. Department of

More information

Mathematics 256 a course in differential equations for engineering students

Mathematics 256 a course in differential equations for engineering students Mathematcs 56 a course n dfferental equatons for engneerng students Chapter 5. More effcent methods of numercal soluton Euler s method s qute neffcent. Because the error s essentally proportonal to the

More information

An Efficient Illumination Normalization Method with Fuzzy LDA Feature Extractor for Face Recognition

An Efficient Illumination Normalization Method with Fuzzy LDA Feature Extractor for Face Recognition www.mer.com Vol.2, Issue.1, pp-060-065 ISS: 2249-6645 An Effcent Illumnaton ormalzaton Meod w Fuzzy LDA Feature Extractor for Face Recognton Behzad Bozorgtabar 1, Hamed Azam 2 (Department of Electrcal

More information

Skew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach

Skew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach Angle Estmaton and Correcton of Hand Wrtten, Textual and Large areas of Non-Textual Document Images: A Novel Approach D.R.Ramesh Babu Pyush M Kumat Mahesh D Dhannawat PES Insttute of Technology Research

More information

Solutions to Programming Assignment Five Interpolation and Numerical Differentiation

Solutions to Programming Assignment Five Interpolation and Numerical Differentiation College of Engneerng and Coputer Scence Mechancal Engneerng Departent Mechancal Engneerng 309 Nuercal Analyss of Engneerng Systes Sprng 04 Nuber: 537 Instructor: Larry Caretto Solutons to Prograng Assgnent

More information

Complex Numbers. Now we also saw that if a and b were both positive then ab = a b. For a second let s forget that restriction and do the following.

Complex Numbers. Now we also saw that if a and b were both positive then ab = a b. For a second let s forget that restriction and do the following. Complex Numbers The last topc n ths secton s not really related to most of what we ve done n ths chapter, although t s somewhat related to the radcals secton as we wll see. We also won t need the materal

More information

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,

More information

R s s f. m y s. SPH3UW Unit 7.3 Spherical Concave Mirrors Page 1 of 12. Notes

R s s f. m y s. SPH3UW Unit 7.3 Spherical Concave Mirrors Page 1 of 12. Notes SPH3UW Unt 7.3 Sphercal Concave Mrrors Page 1 of 1 Notes Physcs Tool box Concave Mrror If the reflectng surface takes place on the nner surface of the sphercal shape so that the centre of the mrror bulges

More information

Kent State University CS 4/ Design and Analysis of Algorithms. Dept. of Math & Computer Science LECT-16. Dynamic Programming

Kent State University CS 4/ Design and Analysis of Algorithms. Dept. of Math & Computer Science LECT-16. Dynamic Programming CS 4/560 Desgn and Analyss of Algorthms Kent State Unversty Dept. of Math & Computer Scence LECT-6 Dynamc Programmng 2 Dynamc Programmng Dynamc Programmng, lke the dvde-and-conquer method, solves problems

More information

Evolutionary Support Vector Regression based on Multi-Scale Radial Basis Function Kernel

Evolutionary Support Vector Regression based on Multi-Scale Radial Basis Function Kernel Eolutonary Support Vector Regresson based on Mult-Scale Radal Bass Functon Kernel Tanasanee Phenthrakul and Boonserm Kjsrkul Abstract Kernel functons are used n support ector regresson (SVR) to compute

More information