CLUSTERING that discovers the relationship among data

Size: px
Start display at page:

Download "CLUSTERING that discovers the relationship among data"

Transcription

1 Ramp-based Twn Support Vector Clusterng Zhen Wang, Xu Chen, Chun-Na L, and Yuan-Ha Shao arxv: v [cs.lg] 0 Dec 208 Abstract Tradtonal plane-based clusterng methods measure the cost of wthn-cluster and between-cluster by quadratc, lnear or some other unbounded functons, whch may amplfy the mpact of cost. Ths letter ntroduces a ramp cost functon nto the plane-based clusterng to propose a new clusterng method, called ramp-based twn support vector clusterng (RampTWSVC). RampTWSVC s more robust because of ts boundness, and thus t s more easer to fnd the ntrnsc clusters than other planebased clusterng methods. The non-convex programmng problem n RampTWSVC s solved effcently through an alternatng teraton algorthm, and ts local soluton can be obtaned n a fnte number of teratons theoretcally. In addton, the nonlnear manfold-based formaton of RampTWSVC s also proposed by kernel trck. Expermental results on several benchmark datasets show the better performance of our RampTWSVC compared wth other plane-based clusterng methods. Index Terms Nonlnear clusterng, plane-based clusterng, ramp cost, twn support vector machnes, unsupervsed learnng. I. INTRODUCTION CLUSTERING that dscovers the relatonshp among data samples, s one of the most fundamental problems n machne learnng [] [4]. It has been appled to many realworld problems, e.g., marketng, text mnng, and web analyss [5], [6]. In partcular, the partton clusterng methods [], [7] are wdely used n real applcaton for ther smplcty, e.g., the classcal kmeans [8] wth ponts as cluster centers, the k-plane clusterng (kpc) [9] and proxmal-plane clusterng (PPC) [0], [] wth planes as cluster centers. As an extenson of pont center, the plane center has the ablty to dscover comprehensve structures n the sample space. The plane-based clusterng seeks the cluster centers dependng on the current cluster assgnment. When a cluster center s constructed, the smlarty of wthn-cluster s ntensfed (n some methods, the dssmlarty of between-cluster s also ntensfed smultaneously). Therefore, the noses or outlers would sgnfcantly nfluence the cluster centers n plane-based clusterng. For nstance, kpc mnmzes the cost of wthncluster by a quadratc functon, and PPC mnmzes the cost of wthn-cluster and between-cluster by the same one. Subsequently, the twn support vector clusterng (TWSVC) [2] was Submtted n December, 208. Ths work s supported n part by Natonal Natural Scence Foundaton of Chna (Nos. 5030, , 8783, and ), n part by Natural Scence Foundaton of Hanan Provnce (No. 8QN8), and n part by Scentfc Research Foundaton of Hanan Unversty (No. kyqd(sk)804). Zhen Wang s wth the School of Mathematcal Scences, Inner Mongola Unversty, Hohhot, 0002 P.R.Chna (e-mal: wangzhen@mu.edu.cn). Xu Chen s wth the School of Mathematcal Scences, Inner Mongola Unversty, Hohhot, 0002 P.R.Chna (e-mal: pohuozhe@63.com). Chun-Na L s wth the Zhjang College, Zhejang Unversty of Technology, Hangzhou 30024, P.R.Chna (e-mal: na03na@63.com). Yuan-Ha Shao (*Correspondng Author) s wth the School of Economcs and Management, Hanan Unversty, Hakou, , P.R.Chna (e-mal: shaoyuanha2@63.com). (a) Wthn-cluster (b) Between-cluster Fg.. Functons used n kpc, PPC, TWSVC, RTWSVC, FRTWSVC, and RampTWSVC to measure the cost of wthn-cluster and between-cluster. The horzontal axs denotes the devaton of a sample from the cluster center, and the vertcal one denotes the cost to ft the sample. When a cost value s negatve, the cost becomes a reward. proposed, whch hred a pecewse lnear functon to measure the cost of between-cluster but perssted n usng the quadratc functon for the wthn-cluster. Recently, another plane-based clusterng method, called robust twn support vector clusterng (RTWSVC) [3], was proposed by hrng a lnear functon to measure the cost of wthn-cluster and between-cluster. Both TWSVC and RTWSVC reduce the nfluence of noses or outlers to some extent. The ramp functon [4], whch has been appled n semsupervsed and supervsed learnng successfully [5] [7], s a bounded pecewse lnear functon. Therefore, n ths letter, we propose a ramp-based twn support vector clusterng method (RampTWSVC) to further reduce the nfluence of noses or outlers both from wthn-cluster and between-cluster, by ntroducng the ramp functon n the constructon of the cluster center planes. The problem of RampTWSVC s a non-convex programmng problem, and t s recast to a mxed nteger programmng problem. We propose an teratve algorthm to solve the mxed nteger programmng problem, and we prove that the algorthm termnates n a fnte number of teratons at a local soluton. In addton, RampTWSVC s extended to nonlnear case by kernel trck to cope wth the manfold clusterng [8], [9]. Fg. exhbts the cost functons used n several plane-based clusterng methods, where FRTWSVC [3] s a plane-based clusterng method called fast robust twn support vector clusterng. It s obvous from Fg. that only our RampTWSVC uses the bounded cost functons both n wthn-cluster and between-cluster, whch can reduce the nfluence of noses or outlers much more than other methods. Expermental results on the benchmark datasets show the better performance of the proposed RampTWSVC compared wth other plane-based clusterng methods.

2 2 II. REVIEW OF PLANE-BASED CLUSTERING In ths paper, we consdermdata samples{x, x 2,...,x m } n the n-dmensonal real vector space R n. Assumng these m samples belong to k clusters wth ther correspondng labels n {, 2,...,k}, and they are represented by a matrx X = (x,x 2,...,x m ) R n m. We further organze the samples from X wth the current label nto a matrx X and those wth the rest labels nto a matrx ˆX, where =, 2,...,k. For readers convenence, the symbols X, X, and ˆX wll also refer to the correspondng sets, dependng on the specfc context they appear. For example, the symbol X can be comprehended as a matrx belongng to R n m or a set that contans m samples. The th cluster center plane ( =,...,k) s defned as f (x) = w x+b = 0, () where w R n and b R. The followng plane-based clusterng methods share the same kmeans-lke clusterng procedure. Startng from an ntal assgnment of the m samples nto k clusters, all the cluster center planes () are constructed by the current cluster assgnment. Once obtaned all the cluster center planes, the m samples are reassgned by y = argmn w x+b, (2) where denotes the absolute value. The cluster center planes and the sample labels are updated alternately untl some termnate condtons are satsfed. In the followng, we brefly descrbe the dfferent constructons of the cluster center plane by kpc, PPC, TWSVC, and RTWSVC. A. kpc and PPC kpc [9] wshes the cluster center plane close to the current cluster samples. Further on, PPC [0] consders t should also be far away from the dfferent cluster samples. Therefore, the th ( =,...,k) cluster center plane n PPC s constructed by solvng the followng problem mn X w +b e 2 c ˆX w +b e 2 w,b s.t. w 2 =, where denotes the L 2 norm, e s a column vector of ones wth approprate dmenson, and c > 0 s a user-set parameter. The optmzaton problem n kpc s just of the frst term of the objectve of (3). From the objectve of (3), t s obvous that a sample from the current cluster receves a quadratc cost, and a sample from a dfferent cluster receves a quadratc reward. Therefore, noses or outlers from the current cluster or dfferent clusters wll have great mpact on the potental cluster center plane. B. TWSVC and RTWSVC PPC may obtan a cluster center plane whch s far from the current cluster, because the samples from dfferent clusters receve hgh rewards when they are far away from the cluster center plane. In contrast, TWSVC [2] degrades the reward of (3) the samples from dfferent clusters by consderng the problem wth =,...,k as mn w 2 X w +b e 2 +ce (e ˆX w +b e ) +, (4),b where ( ) + replaces the negatve part by zeros. From the second part of (4), t s avalable that a sample wth a devaton n[0,) has mpact on the cluster center plane. Thus, TWSVC s more robust than PPC. However, the ssue of current cluster also exsts because of the quadratc cost n the frst part of (4). Thus, RTWSVC [3] was proposed to decreases the nfluence of current cluster by replacng the L 2 norm n (4) wth L norm. RTWSVC nherts the advantage of TWSVC and decreases the requrement from current cluster. However, the cost of RTWSVC from wthncluster s unbounded from Fg.. In order to eradcate the nfluence of noses or outlers, t s reasonable to hre a bounded functon for the wthn-cluster, whose prncple s smlar to the cost for the between-cluster used n TWSVC. III. RAMPTWSVC Smlar to the above plane-based clusterng methods mentoned n secton 2, our RampTWSVC starts wth an ntal sample labels, then computes each cluster center plane for the current sample labels teratvely, untl some termnate condtons are satsfed. In the followng, we consder to obtan one of the cluster center planes for the gven samples wth ther labels. A. Formaton To obtan the th ( =,...,k) cluster center plane, our RampTWSVC consders the followng problem mn w 2 ( w 2 +b 2 )+c R (x j )+c 2 R 2 (x j ),,b x j X x j ˆX where c > 0 and c 2 > 0 are parameters. R (x) and R 2 (x) are two pecewse lnear functons w.r.t. the devaton f (x) = w x+b (see Fg. ) as 0 f f (x), R (x) = s f f (x) 2 s, f (x) + otherwse, (6) 2+2 f f (x) s, R 2 (x) = + s f f (x) +, f (x) +2+2 s otherwse, (7) where [0,) ands (,0] are two parameters to control the functon form (typcally, we set = 0.3 and s = 0.2 n ths letter). It s obvous that both of the cost functons R (x) (for the current cluster X ) and R 2 (x) (for the dfferent clusters ˆX ) have bounds for large devaton. Thus, the noses or outlers much further from the cluster center plane do not have greater mpact on the cluster center plane when they meet the bound. (5)

3 3 The above property ndcates our RampTWSVC s more robust than RTWSVC. In the followng, we extend the RampTWSVC to nonlnear manfold clusterng, and the solutons to the problems n lnear and nonlnear RampTWSVC are elaborated n next subseton. The plane-based clusterng method can be extend to nonlnear manfold clusterng easly by the kernel trck [20], [2]. By ntroducng a pre-defned kernel functon K(, ), the planebased nonlnear clusterng seeks k cluster center manfolds n the kernel generated space as g (x) = K(x,X) w +b = 0, =,...,k. (8) Then, the nonlnear RampTWSVC consders to ntroduce the ramp functons nto the plane-based nonlnear clusterng. By replacng f (x) wth g (x) n (6) and (7), and substtutng them nto (5), one can easly obtan k optmzaton problems to construct the cluster center manfolds (8). When we obtan the k cluster centers (8), a sample x s assgned to whch cluster dependng on y = argmn K(x,X) w +b. (9) The procedure of the nonlnear case s the same as the lnear one, so the detals are omtted. B. Soluton In ths subsecton, we study the soluton to the problem (5). The correspondng problem n nonlnear RampTWSVC s smlar to the one n lnear case. For convenence, let u = (w,b ), Z be a matrx whose jth column z j s x j wth an addtonal feature (where the correspondng x j belongs to the th cluster), and Ẑ be a matrx whose column smlar as z j (where the correspondng x j does not belongs to the th cluster). Then, the problem (5) s recast to mn u 2 u j 2 +c e ( + Z u ) + +c e ( + +Z u ) + +c 2 e (+ Ẑ u ) + +c 2 e (+ +Ẑ u ) + c e (s 2+ Z u ) + c e (s 2+ +Z u ) + c 2 e (s Ẑ u ) + c 2 e (s+ẑ u ) +. (0) It s easy to see that the above problem s a non-convex programmng problem because of the concave part ( ) +. By ntroducng two auxlary vectors p {,0,} m and p 2 {,0,} m m (where m s the sample number of the current th cluster), the above problem s equvalent to the followng mxed-nteger programmng problem mn u 2 u 2 +c e ( + Z u ) +,p,p 2 +c e ( + +Z u ) + +c 2 e (+ Ẑ u ) + +c 2 e (+ +Ẑ u ) + +c p Z u +c 2 p 2 Ẑ u f zj u > 2 s, s.t. p (j) = f zj u < 2+ +s, z j Z p 2 (j) = 0 otherwse, f zj u > s, f zj u < s, 0 otherwse, z j Ẑ () TABLE I DETAILS OF THE BENCHMARK DATASETS Data m n k (a) Arrhythma (b) Dermatology (c) Ecol (d) Glass (e) Irs (f) Lbras (g) Seeds (h) Wne () Zoo (j) Bupa (k) Echocardogram (l) Heartstatlog (m) Housevotes (n) Ionosphere (o) Sonar (p) Soybean (q) Spect (r) Wpbc *m s the number of samples, n s the one of dmenson, and k s the one of classes. where p (j) and p 2 (j) are the correspondng jth elements of p and p 2, respectvely. Here, we propose an alternatng teraton algorthm to solve the mxed-nteger programmng problem (). Startng wth an ntalzed u (0), t s easy to calculate p (0) and p (0) 2 by the constrants of (). For fxed p (t ) and p (t ) 2 (t =,2,...), the problem () becomes to an unconstraned convex problem and ts soluton can be obtaned by many algorthms easly (e.g., sequental mnmal optmzaton (SMO) [22] and fast Newton-Amjo algorthm [23]). Once obtaned u (t), p (t) and p (t) 2 are updated agan. The loop wll be contnued untl the objectve of () does not decrease any more. Theorem III.. The above alternatng teraton algorthm to solve () termnates n a fnte number of teratons at a local optmal pont, where a local optmal pont of the mxed nteger programmng problem () s defned as the pont (u,p,p 2) f u s the global soluton to the problem () wth fxed (p,p 2 ) and vce versa. Proof. From the procedure of the alternatng teraton algorthm, t s obvous that the global solutons to the problem () wth fxed u or (p,p 2 ) are obtaned n teraton. Snce there s a fnte number of ways to select p and p 2, there are two fnte numbers r,r 2 > 0 such that (p (r) 2 ) = (p (r2),p (r2) 2 ). Thus, we have u (r) = u (r2). That s to say, the objectve values are equal n the r th and r 2 th teratons. Snce p Z u 0 and p 2 Ẑ u 0 are always holds, the objectve value of () keeps non-ncreasng n teraton. Therefore, the objectve s nvarant after the r th teraton, and then the algorthm would termnate at the r th teraton. Let us consder the pont (u (r),p (r) 2 ). From the above proof, we have G(u (r),p (r) 2 ) = G(u (r),p (r+),p (r+) 2 ), where G( ) s the objectve value of (). If there are more than one global soluton to the problem () wth fxed u, we always select the same one for the same u. Thus, we have (p (r) 2 ) = (p (r+),p (r+) 2 ), 2 ) s a local optmal whch ndcates the pont (u (r) pont.,p (r)

4 4 IV. EXPERIMENTAL RESULTS In ths secton, we analyze the performance of our RampTWSVC compared wth kmeans [8], kpc [9], PPC [0], TWSVC [2], RTWSVC [3], and FRTWSVC [3] on several benchmark datasets [24]. All the methods were mplemented by MATLAB207 on a PC wth an Intel Core Duo processor (double 4.2 GHz) wth 6GB RAM. The parameters c n PPC, TWSVC, RTWSVC, FRTWSVC, and c, c 2 n RampTWSVC were selected from {2 = 8, 7,...,7}. For nonlnear case, the Gaussan kernel K(x,x 2 ) = exp{ µ x x 2 2 } [20] was used, and ts parameter µ was selected from {2 = 0, 9,...,5}. The random ntalzaton was used on kmeans, and the nearest neghbor graph (NNG) ntalzaton [2] was used on other methods. In the experments, we used the metrc accuracy (AC) [2] and mutual nformaton (MI) [25] to measure the performance of these methods. Table I shows the detals of the benchmark datasets. Tables II and III exhbt the lnear and nonlnear clusterng methods on the benchmark datasets, respectvely. The hghest metrcs among these methods on each dataset are n bold. Besdes, we also reported the statstcs of these methods n the last rows n Tables II and III, whch s the number of the datasets that each method s the hghest one n terms of AC, MI, or both. From Table II, t can be seen that our lnear RampTWSVC performs better than other lnear methods on fve datasets n terms of both AC and MI, and t s more accurate than other methods on other fve datasets. On the rest eght datasets, our lnear RampTWSVC s also compettve wth the best one. From Table III, t s obvous that our nonlnear RampTWSVC has much hgher AC and MI over other methods on many datasets. V. CONCLUSIONS A plane-based clusterng method (RampTWSVC) has been proposed wth the ramp functon. It contans both the lnear and nonlnear formatons. The cluster center planes n RampTWSVC are obtaned by solvng a seres of non-convex problems, and ther local solutons are guaranteed by a proposed alternatng teraton algorthm n theory. Expermental results on several benchmark datasets have ndcated that our RampTWSVC performs much better than other plane-based clusterng methods on many datasets. For practcal convenence, the correspondng RampTWSVC Matlab code has been uploaded upon Future work ncludes the parameter regulaton and effcent solver desgn for our non-convex problems. REFERENCES [] A. Jan, M. Murty, and P. Flynn, Data clusterng: a revew, ACM computng surveys (CSUR), vol. 3, no. 3, pp , 999. [2] Z. Kang, C. Peng, and Q. Cheng, Robust subspace clusterng va smoothed rank approxmaton, IEEE Sgnal Processng Letters, vol. 22, no., pp , 205. [3] A. Malhotra and I. Schzas, Mlp-based unsupervsed clusterng, IEEE Sgnal Processng Letters, vol. 25, no. 2, pp , 208. [4] Z. Zhou, A bref ntroducton to weakly supervsed learnng, Natonal Scence Revew, vol. 5, no., pp , 207. TABLE II LINEAR CLUSTERING ON BENCHMARK DATASETS kmeans kpc PPC TWSVC RTWSVC FRTWSVC Ours Data AC(%) AC(%) AC(%) AC(%) AC(%) AC(%) AC(%) MI(%) MI(%) MI(%) MI(%) MI(%) MI(%) MI(%) (a) 65.72± ± (b) 69.76± ± (c) 82.9± ± (d) 65.58± ± (e) 84.57± ± (f) 90.84± ± (g) 87.35± ± (h) 7.06± ± () 87.49± ± (j) 50.39± ± (k) 66.4± ± (l) 5.45± ± (m) 78.83± ± (n) 58.89± ± (o) 50.22± ± (p) 93.4± ± (q) 52.97± ± (r) 56.03± ± AC MI Both TABLE III NONLINEAR CLUSTERING ON BENCHMARK DATASETS kmeans kpc PPC TWSVC RTWSVC FRTWSVC Ours Data AC(%) AC(%) AC(%) AC(%) AC(%) AC(%) AC(%) MI(%) MI(%) MI(%) MI(%) MI(%) MI(%) MI(%) (a) 47.32± ± (b) 7.66± ± (c) 79.93± ± (d) 69.27± ± (e) 87.63± ± (f) 90.60± ± (g) 87.02± ± (h) 52.07± ± () 87.4± ± (j) 5.08± ± (k) 7.4± ± (l) 50.83± ± (m) 79.79± ± (n) 62.32± ± (o) 50.6± ± (p) 00.0± ± (q) 60.68± ± (r) 63.40± ± AC MI Both

5 5 [5] M. Berry, Survey of Text Mnng I: Clusterng, Classfcaton, and Retreval. Sprnger, 2004, vol.. [6] R. Iln, Unsupervsed learnng of categorcal data wth competng models, Neural Networks and Learnng Systems, IEEE Transactons on, vol. 23, no., pp , 202. [7] P. Tan, M. Stenbach, and V. Kumar, Introducton to Data Mnng, (st Edton). Boston, MA, USA: Addson-Wesley Longman Publshng Co., Inc., [8] A. Jan and R. Dubes, Algorthms for clusterng data. Prentce-Hall, Inc., 988. [9] P. Bradley and O. Mangasaran, k-plane clusterng, Journal of Global Optmzaton, vol. 6, no., pp , [0] Y. Shao, L. Ba, Z. Wang, X. Hua, and N. Deng, Proxmal plane clusterng va egenvalues, Proceda Computer Scence, vol. 7, pp. 4 47, 203. [] Y. Shao, Y. Guo, Z. Wang, Z. Yang, and N. Deng, k-proxmal plane clusterng, Internatonal Journal of Machne Learnng and Cybernetcs, vol. 8, no. 5, pp , 207. [2] Z. Wang, Y. Shao, L. Ba, and N. Deng, Twn support vector machne for clusterng, IEEE transactons on neural networks and learnng systems, vol. 26, no. 0, pp , 205. [3] Q. Ye, H. Zhao, Z. L, X. Yang, S. Gao, T. Yn, and N. Ye, L- norm dstance mnmzaton-based fast robust twn support vector k- plane clusterng, IEEE transactons on neural networks and learnng systems, vol. 29, no. 9, pp , 208. [4] R. Collobert, F. Snz, J. Weston, and L. Bottou, Large scale transductve svms, Journal of Machne Learnng Research, vol. 7, pp , [5] H. Cevkalp, Best fttng hyperplanes for classfcaton, IEEE transactons on pattern analyss and machne ntellgence, vol. 39, no. 6, pp , 207. [6] Y. Tan, M. Mrzabagher, S. Bamakan, H. Wang, and Q. Qu, Ramp loss one-class support vector machne: A robust and effectve approach to anomaly detecton problems, Neurocomputng, vol. 30, no. 8, pp , 208. [7] D. Lu, Y. Sh, Y. Tan, and X. Huang, Ramp loss least squares support vector machne, Journal of computatonal scence, vol. 4, pp. 6 68, 206. [8] R. Souvenr and R. Pless, Manfold clusterng, n Computer Vson, ICCV Tenth IEEE Internatonal Conference on, vol.. IEEE, 2005, pp [9] W. Cao and R. Haralck, Nonlnear manfold clusterng by dmensonalty, n Pattern Recognton, ICPR th Internatonal Conference on, vol.. IEEE, 2006, pp [20] R. Khemchandan, Jayadeva, and S. Chandra, Optmal kernel selecton n twn support vector machnes, Optmzaton Letters, vol. 3, pp , [2] Z. Wang, Y. Shao, L. Ba, C. L, L. Lu, and N. Deng, Insenstve stochastc gradent twn support vector machnes for large scale problems, Informaton Scences, vol. 462, pp. 4 3, 208. [22] J. Platt, Fast tranng of support vector machnes usng sequental mnmal optmzaton, n Advances n kernel methods-support vector learnng, Cambrdge, MA: MIT Press, 999, pp [23] Z. Wang, Y. Shao, and T. Wu, A ga-based model selecton for smooth twn parametrc-margn support vector machne, Pattern Recognton, vol. 46, no. 8, pp , 203. [24] C. Blake and C. Merz, UCI Repostory for Machne Learnng Databases, mlearn/mlrepostory.html, 998. [25] H. Lu and Y. Fu, Clusterng wth partton level sde nformaton, n In Data Mnng (ICDM), 205 IEEE Internatonal Conference on, IEEE, 205, pp

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

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

Outline. Type of Machine Learning. Examples of Application. Unsupervised Learning

Outline. Type of Machine Learning. Examples of Application. Unsupervised Learning Outlne Artfcal Intellgence and ts applcatons Lecture 8 Unsupervsed Learnng Professor Danel Yeung danyeung@eee.org Dr. Patrck Chan patrckchan@eee.org South Chna Unversty of Technology, Chna Introducton

More information

Unsupervised Learning

Unsupervised Learning Pattern Recognton Lecture 8 Outlne Introducton Unsupervsed Learnng Parametrc VS Non-Parametrc Approach Mxture of Denstes Maxmum-Lkelhood Estmates Clusterng Prof. Danel Yeung School of Computer Scence and

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

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

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points;

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points; Subspace clusterng Clusterng Fundamental to all clusterng technques s the choce of dstance measure between data ponts; D q ( ) ( ) 2 x x = x x, j k = 1 k jk Squared Eucldean dstance Assumpton: All features

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

Machine Learning: Algorithms and Applications

Machine Learning: Algorithms and Applications 14/05/1 Machne Learnng: Algorthms and Applcatons Florano Zn Free Unversty of Bozen-Bolzano Faculty of Computer Scence Academc Year 011-01 Lecture 10: 14 May 01 Unsupervsed Learnng cont Sldes courtesy of

More information

Discriminative Dictionary Learning with Pairwise Constraints

Discriminative Dictionary Learning with Pairwise Constraints Dscrmnatve Dctonary Learnng wth Parwse Constrants Humn Guo Zhuoln Jang LARRY S. DAVIS UNIVERSITY OF MARYLAND Nov. 6 th, Outlne Introducton/motvaton Dctonary Learnng Dscrmnatve Dctonary Learnng wth Parwse

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

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

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

Fast Computation of Shortest Path for Visiting Segments in the Plane

Fast Computation of Shortest Path for Visiting Segments in the Plane Send Orders for Reprnts to reprnts@benthamscence.ae 4 The Open Cybernetcs & Systemcs Journal, 04, 8, 4-9 Open Access Fast Computaton of Shortest Path for Vstng Segments n the Plane Ljuan Wang,, Bo Jang

More information

Cluster Analysis of Electrical Behavior

Cluster Analysis of Electrical Behavior Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School

More information

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 15

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 15 CS434a/541a: Pattern Recognton Prof. Olga Veksler Lecture 15 Today New Topc: Unsupervsed Learnng Supervsed vs. unsupervsed learnng Unsupervsed learnng Net Tme: parametrc unsupervsed learnng Today: nonparametrc

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

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

Unsupervised Learning and Clustering

Unsupervised Learning and Clustering Unsupervsed Learnng and Clusterng Supervsed vs. Unsupervsed Learnng Up to now we consdered supervsed learnng scenaro, where we are gven 1. samples 1,, n 2. class labels for all samples 1,, n Ths s also

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

Local Quaternary Patterns and Feature Local Quaternary Patterns

Local Quaternary Patterns and Feature Local Quaternary Patterns Local Quaternary Patterns and Feature Local Quaternary Patterns Jayu Gu and Chengjun Lu The Department of Computer Scence, New Jersey Insttute of Technology, Newark, NJ 0102, USA Abstract - Ths paper presents

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

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

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

GSLM Operations Research II Fall 13/14

GSLM Operations Research II Fall 13/14 GSLM 58 Operatons Research II Fall /4 6. Separable Programmng Consder a general NLP mn f(x) s.t. g j (x) b j j =. m. Defnton 6.. The NLP s a separable program f ts objectve functon and all constrants are

More information

An Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices

An Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices Internatonal Mathematcal Forum, Vol 7, 2012, no 52, 2549-2554 An Applcaton of the Dulmage-Mendelsohn Decomposton to Sparse Null Space Bases of Full Row Rank Matrces Mostafa Khorramzadeh Department of Mathematcal

More information

Nonlocal Mumford-Shah Model for Image Segmentation

Nonlocal Mumford-Shah Model for Image Segmentation for Image Segmentaton 1 College of Informaton Engneerng, Qngdao Unversty, Qngdao, 266000,Chna E-mal:ccluxaoq@163.com ebo e 23 College of Informaton Engneerng, Qngdao Unversty, Qngdao, 266000,Chna E-mal:

More information

Positive Semi-definite Programming Localization in Wireless Sensor Networks

Positive Semi-definite Programming Localization in Wireless Sensor Networks Postve Sem-defnte Programmng Localzaton n Wreless Sensor etworks Shengdong Xe 1,, Jn Wang, Aqun Hu 1, Yunl Gu, Jang Xu, 1 School of Informaton Scence and Engneerng, Southeast Unversty, 10096, anjng Computer

More information

On Some Entertaining Applications of the Concept of Set in Computer Science Course

On Some Entertaining Applications of the Concept of Set in Computer Science Course On Some Entertanng Applcatons of the Concept of Set n Computer Scence Course Krasmr Yordzhev *, Hrstna Kostadnova ** * Assocate Professor Krasmr Yordzhev, Ph.D., Faculty of Mathematcs and Natural Scences,

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

Programming in Fortran 90 : 2017/2018

Programming in Fortran 90 : 2017/2018 Programmng n Fortran 90 : 2017/2018 Programmng n Fortran 90 : 2017/2018 Exercse 1 : Evaluaton of functon dependng on nput Wrte a program who evaluate the functon f (x,y) for any two user specfed values

More information

Unsupervised Learning and Clustering

Unsupervised Learning and Clustering Unsupervsed Learnng and Clusterng Why consder unlabeled samples?. Collectng and labelng large set of samples s costly Gettng recorded speech s free, labelng s tme consumng 2. Classfer could be desgned

More information

Outline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like:

Outline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like: Self-Organzng Maps (SOM) Turgay İBRİKÇİ, PhD. Outlne Introducton Structures of SOM SOM Archtecture Neghborhoods SOM Algorthm Examples Summary 1 2 Unsupervsed Hebban Learnng US Hebban Learnng, Cntd 3 A

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

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

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour 6.854 Advanced Algorthms Petar Maymounkov Problem Set 11 (November 23, 2005) Wth: Benjamn Rossman, Oren Wemann, and Pouya Kheradpour Problem 1. We reduce vertex cover to MAX-SAT wth weghts, such that the

More information

The Shortest Path of Touring Lines given in the Plane

The Shortest Path of Touring Lines given in the Plane Send Orders for Reprnts to reprnts@benthamscence.ae 262 The Open Cybernetcs & Systemcs Journal, 2015, 9, 262-267 The Shortest Path of Tourng Lnes gven n the Plane Open Access Ljuan Wang 1,2, Dandan He

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

Range images. Range image registration. Examples of sampling patterns. Range images and range surfaces

Range images. Range image registration. Examples of sampling patterns. Range images and range surfaces Range mages For many structured lght scanners, the range data forms a hghly regular pattern known as a range mage. he samplng pattern s determned by the specfc scanner. Range mage regstraton 1 Examples

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

ON SOME ENTERTAINING APPLICATIONS OF THE CONCEPT OF SET IN COMPUTER SCIENCE COURSE

ON SOME ENTERTAINING APPLICATIONS OF THE CONCEPT OF SET IN COMPUTER SCIENCE COURSE Yordzhev K., Kostadnova H. Інформаційні технології в освіті ON SOME ENTERTAINING APPLICATIONS OF THE CONCEPT OF SET IN COMPUTER SCIENCE COURSE Yordzhev K., Kostadnova H. Some aspects of programmng educaton

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

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance Tsnghua Unversty at TAC 2009: Summarzng Mult-documents by Informaton Dstance Chong Long, Mnle Huang, Xaoyan Zhu State Key Laboratory of Intellgent Technology and Systems, Tsnghua Natonal Laboratory for

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

LECTURE : MANIFOLD LEARNING

LECTURE : MANIFOLD LEARNING LECTURE : MANIFOLD LEARNING Rta Osadchy Some sldes are due to L.Saul, V. C. Raykar, N. Verma Topcs PCA MDS IsoMap LLE EgenMaps Done! Dmensonalty Reducton Data representaton Inputs are real-valued vectors

More information

A PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION

A PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION 1 THE PUBLISHING HOUSE PROCEEDINGS OF THE ROMANIAN ACADEMY, Seres A, OF THE ROMANIAN ACADEMY Volume 4, Number 2/2003, pp.000-000 A PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION Tudor BARBU Insttute

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

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

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

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

The Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique

The Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique //00 :0 AM Outlne and Readng The Greedy Method The Greedy Method Technque (secton.) Fractonal Knapsack Problem (secton..) Task Schedulng (secton..) Mnmum Spannng Trees (secton.) Change Money Problem Greedy

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

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

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

An Image Fusion Approach Based on Segmentation Region

An Image Fusion Approach Based on Segmentation Region Rong Wang, L-Qun Gao, Shu Yang, Yu-Hua Cha, and Yan-Chun Lu An Image Fuson Approach Based On Segmentaton Regon An Image Fuson Approach Based on Segmentaton Regon Rong Wang, L-Qun Gao, Shu Yang 3, Yu-Hua

More information

A fast algorithm for color image segmentation

A fast algorithm for color image segmentation Unersty of Wollongong Research Onlne Faculty of Informatcs - Papers (Arche) Faculty of Engneerng and Informaton Scences 006 A fast algorthm for color mage segmentaton L. Dong Unersty of Wollongong, lju@uow.edu.au

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

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

FINDING IMPORTANT NODES IN SOCIAL NETWORKS BASED ON MODIFIED PAGERANK

FINDING IMPORTANT NODES IN SOCIAL NETWORKS BASED ON MODIFIED PAGERANK FINDING IMPORTANT NODES IN SOCIAL NETWORKS BASED ON MODIFIED PAGERANK L-qng Qu, Yong-quan Lang 2, Jng-Chen 3, 2 College of Informaton Scence and Technology, Shandong Unversty of Scence and Technology,

More information

A Deflected Grid-based Algorithm for Clustering Analysis

A Deflected Grid-based Algorithm for Clustering Analysis A Deflected Grd-based Algorthm for Clusterng Analyss NANCY P. LIN, CHUNG-I CHANG, HAO-EN CHUEH, HUNG-JEN CHEN, WEI-HUA HAO Department of Computer Scence and Informaton Engneerng Tamkang Unversty 5 Yng-chuan

More information

TN348: Openlab Module - Colocalization

TN348: Openlab Module - Colocalization TN348: Openlab Module - Colocalzaton Topc The Colocalzaton module provdes the faclty to vsualze and quantfy colocalzaton between pars of mages. The Colocalzaton wndow contans a prevew of the two mages

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

CMPS 10 Introduction to Computer Science Lecture Notes

CMPS 10 Introduction to Computer Science Lecture Notes CPS 0 Introducton to Computer Scence Lecture Notes Chapter : Algorthm Desgn How should we present algorthms? Natural languages lke Englsh, Spansh, or French whch are rch n nterpretaton and meanng are not

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

A New Feature of Uniformity of Image Texture Directions Coinciding with the Human Eyes Perception 1

A New Feature of Uniformity of Image Texture Directions Coinciding with the Human Eyes Perception 1 A New Feature of Unformty of Image Texture Drectons Concdng wth the Human Eyes Percepton Xng-Jan He, De-Shuang Huang, Yue Zhang, Tat-Mng Lo 2, and Mchael R. Lyu 3 Intellgent Computng Lab, Insttute of Intellgent

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

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 CONDENSED FUZZY K-NEAREST NEIGHBOR RULE BASED ON SAMPLE FUZZY ENTROPY

THE CONDENSED FUZZY K-NEAREST NEIGHBOR RULE BASED ON SAMPLE FUZZY ENTROPY Proceedngs of the 20 Internatonal Conference on Machne Learnng and Cybernetcs, Guln, 0-3 July, 20 THE CONDENSED FUZZY K-NEAREST NEIGHBOR RULE BASED ON SAMPLE FUZZY ENTROPY JUN-HAI ZHAI, NA LI, MENG-YAO

More information

For instance, ; the five basic number-sets are increasingly more n A B & B A A = B (1)

For instance, ; the five basic number-sets are increasingly more n A B & B A A = B (1) Secton 1.2 Subsets and the Boolean operatons on sets If every element of the set A s an element of the set B, we say that A s a subset of B, or that A s contaned n B, or that B contans A, and we wrte A

More information

The Research of Ellipse Parameter Fitting Algorithm of Ultrasonic Imaging Logging in the Casing Hole

The Research of Ellipse Parameter Fitting Algorithm of Ultrasonic Imaging Logging in the Casing Hole Appled Mathematcs, 04, 5, 37-3 Publshed Onlne May 04 n ScRes. http://www.scrp.org/journal/am http://dx.do.org/0.436/am.04.584 The Research of Ellpse Parameter Fttng Algorthm of Ultrasonc Imagng Loggng

More information

K-means and Hierarchical Clustering

K-means and Hierarchical Clustering Note to other teachers and users of these sldes. Andrew would be delghted f you found ths source materal useful n gvng your own lectures. Feel free to use these sldes verbatm, or to modfy them to ft your

More information

Semi-supervised Classification Using Local and Global Regularization

Semi-supervised Classification Using Local and Global Regularization Proceedngs of the Twenty-Thrd AAAI Conference on Artfcal Intellgence (2008) Sem-supervsed Classfcaton Usng Local and Global Regularzaton Fe Wang 1, Tao L 2, Gang Wang 3, Changshu Zhang 1 1 Department of

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

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

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

Semi-Supervised Discriminant Analysis Based On Data Structure

Semi-Supervised Discriminant Analysis Based On Data Structure IOSR Journal of Computer Engneerng (IOSR-JCE) e-issn: 2278-0661,p-ISSN: 2278-8727, Volume 17, Issue 3, Ver. VII (May Jun. 2015), PP 39-46 www.osrournals.org Sem-Supervsed Dscrmnant Analyss Based On Data

More information

Hybridization of Expectation-Maximization and K-Means Algorithms for Better Clustering Performance

Hybridization of Expectation-Maximization and K-Means Algorithms for Better Clustering Performance BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 16, No 2 Sofa 2016 Prnt ISSN: 1311-9702; Onlne ISSN: 1314-4081 DOI: 10.1515/cat-2016-0017 Hybrdzaton of Expectaton-Maxmzaton

More information

Learning-Based Top-N Selection Query Evaluation over Relational Databases

Learning-Based Top-N Selection Query Evaluation over Relational Databases Learnng-Based Top-N Selecton Query Evaluaton over Relatonal Databases Lang Zhu *, Wey Meng ** * School of Mathematcs and Computer Scence, Hebe Unversty, Baodng, Hebe 071002, Chna, zhu@mal.hbu.edu.cn **

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

Geometry-aware Metric Learning

Geometry-aware Metric Learning Zhengdong Lu LUZ@CS.UTEXAS.EDU Prateek Jan PJAIN@CS.UTEXAS.EDU Inderjt S. Dhllon INDERJIT@CS.UTEXAS.EDU Dept. of Computer Scence, Unversty of Texas at Austn, Unversty Staton C5, Austn, TX 78712 Abstract

More information

Related-Mode Attacks on CTR Encryption Mode

Related-Mode Attacks on CTR Encryption Mode Internatonal Journal of Network Securty, Vol.4, No.3, PP.282 287, May 2007 282 Related-Mode Attacks on CTR Encrypton Mode Dayn Wang, Dongda Ln, and Wenlng Wu (Correspondng author: Dayn Wang) Key Laboratory

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

Hierarchical clustering for gene expression data analysis

Hierarchical clustering for gene expression data analysis Herarchcal clusterng for gene expresson data analyss Gorgo Valentn e-mal: valentn@ds.unm.t Clusterng of Mcroarray Data. Clusterng of gene expresson profles (rows) => dscovery of co-regulated and functonally

More information

Real-time Joint Tracking of a Hand Manipulating an Object from RGB-D Input

Real-time Joint Tracking of a Hand Manipulating an Object from RGB-D Input Real-tme Jont Tracng of a Hand Manpulatng an Object from RGB-D Input Srnath Srdhar 1 Franzsa Mueller 1 Mchael Zollhöfer 1 Dan Casas 1 Antt Oulasvrta 2 Chrstan Theobalt 1 1 Max Planc Insttute for Informatcs

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

Vectorization of Image Outlines Using Rational Spline and Genetic Algorithm

Vectorization of Image Outlines Using Rational Spline and Genetic Algorithm 01 Internatonal Conference on Image, Vson and Computng (ICIVC 01) IPCSIT vol. 50 (01) (01) IACSIT Press, Sngapore DOI: 10.776/IPCSIT.01.V50.4 Vectorzaton of Image Outlnes Usng Ratonal Splne and Genetc

More information

Clustering using Vector Membership: An Extension of the Fuzzy C-Means Algorithm

Clustering using Vector Membership: An Extension of the Fuzzy C-Means Algorithm Clusterng usng Vector Membershp: An Extenson of the Fuzzy C-Means Algorthm Srnjoy Ganguly 1, Dgbalay Bose, Amt Konar 3 1,,3 Department of Electroncs & Telecommuncaton Engneerng, Jadavpur Unversty, Kolkata,

More information

Laplacian Eigenmap for Image Retrieval

Laplacian Eigenmap for Image Retrieval Laplacan Egenmap for Image Retreval Xaofe He Partha Nyog Department of Computer Scence The Unversty of Chcago, 1100 E 58 th Street, Chcago, IL 60637 ABSTRACT Dmensonalty reducton has been receved much

More information

An Evolvable Clustering Based Algorithm to Learn Distance Function for Supervised Environment

An Evolvable Clustering Based Algorithm to Learn Distance Function for Supervised Environment IJCSI Internatonal Journal of Computer Scence Issues, Vol. 7, Issue 5, September 2010 ISSN (Onlne): 1694-0814 www.ijcsi.org 374 An Evolvable Clusterng Based Algorthm to Learn Dstance Functon for Supervsed

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

Network Intrusion Detection Based on PSO-SVM

Network Intrusion Detection Based on PSO-SVM TELKOMNIKA Indonesan Journal of Electrcal Engneerng Vol.1, No., February 014, pp. 150 ~ 1508 DOI: http://dx.do.org/10.11591/telkomnka.v1.386 150 Network Intruson Detecton Based on PSO-SVM Changsheng Xang*

More information

Module Management Tool in Software Development Organizations

Module Management Tool in Software Development Organizations Journal of Computer Scence (5): 8-, 7 ISSN 59-66 7 Scence Publcatons Management Tool n Software Development Organzatons Ahmad A. Al-Rababah and Mohammad A. Al-Rababah Faculty of IT, Al-Ahlyyah Amman Unversty,

More information

A Two-Stage Algorithm for Data Clustering

A Two-Stage Algorithm for Data Clustering A Two-Stage Algorthm for Data Clusterng Abdolreza Hatamlou 1 and Salwan Abdullah 2 1 Islamc Azad Unversty, Khoy Branch, Iran 2 Data Mnng and Optmsaton Research Group, Center for Artfcal Intellgence Technology,

More information

Intelligent Information Acquisition for Improved Clustering

Intelligent Information Acquisition for Improved Clustering Intellgent Informaton Acquston for Improved Clusterng Duy Vu Unversty of Texas at Austn duyvu@cs.utexas.edu Mkhal Blenko Mcrosoft Research mblenko@mcrosoft.com Prem Melvlle IBM T.J. Watson Research Center

More information

Evaluation of the application of BIM technology based on PCA - Q Clustering Algorithm and Choquet Integral

Evaluation of the application of BIM technology based on PCA - Q Clustering Algorithm and Choquet Integral IETI Transactons on Busness and Management Scences, 2016, Volume 1, Issue 1, 47-55. http://www.et.net/tc An Internatonal Open Access Journal Evaluaton of the applcaton of BIM technology based on PCA -

More information

Review of approximation techniques

Review of approximation techniques CHAPTER 2 Revew of appromaton technques 2. Introducton Optmzaton problems n engneerng desgn are characterzed by the followng assocated features: the objectve functon and constrants are mplct functons evaluated

More information

Available online at ScienceDirect. Procedia Environmental Sciences 26 (2015 )

Available online at   ScienceDirect. Procedia Environmental Sciences 26 (2015 ) Avalable onlne at www.scencedrect.com ScenceDrect Proceda Envronmental Scences 26 (2015 ) 109 114 Spatal Statstcs 2015: Emergng Patterns Calbratng a Geographcally Weghted Regresson Model wth Parameter-Specfc

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