Optimal Fuzzy Clustering in Overlapping Clusters

Size: px
Start display at page:

Download "Optimal Fuzzy Clustering in Overlapping Clusters"

Transcription

1 46 The Internatonal Arab Journal of Informaton Technology, Vol. 5, No. 4, October 008 Optmal Fuzzy Clusterng n Overlappng Clusters Ouafa Ammor, Abdelmoname Lachar, Khada Slaou 3, and Noureddne Ras Department of Mathematcs, Faculty of Scences and Technology of Fes, Morocco ESTM, Moulay Ismal Unversty, Morocco 3 Department of Physcs, Faculty of Scences Dhar Mehraz of Fes, Morocco Abstract: The fuzzy c-means clusterng algorthm has been wdely used to obtan the fuzzy -parttons. Ths algorthm requres that the user gves the number of clusters. To fnd automatcally the rght number of clusters,, for a gven data set, many valdty ndexes algorthms have been proposed n the lterature. Most of these ndexes do not wor well for clusters wth dfferent overlappng degree. They usually have a tendency to fals n selectng the correct optmal clusters number when dealng wth some data sets contanng overlappng clusters. To overcome ths lmtaton, we propose n ths paper, a new and effcent clusters valdty measure for determnaton of the optmal number of clusters whch can deal successfully wth or wthout stuaton of overlappng. Ths measure s based on mamum entropy prncple. Our approach does not requre any parameter adustment, t s then completely automatc. Many smulated and real examples are presented, showng the superorty of our measure to the estng ones. Keywords: Unsupervsed clusterng, cluster valdty ndex, optmal clusters number, overlappng clusters, mamum entropy prncple. Receved November 30, 006; accepted June, 007. Introducton Cluster analyss has been playng an mportant role n solvng many problems n pattern recognton, mage processng, colour mage segmentaton, machne learnng, data mnng, and dfferent felds le medcne, bology, technology, maretng. The am of the cluster valdty s to fnd the parttonng that best fts the underlyng data. A wde varety of clusterng algorthms have been proposed for dfferent applcatons and a good overvew can be found n the lterature [, 4, 6, 7]. Snce there are no predefned classes, t s therefore dffcult to fnd an approprate metrc for measurng f the found clusters confguraton s acceptable or not. The result of a clusterng algorthm can be very dfferent from each other on the same data set, and nput parameters of an algorthm can extremely modfy the behavour and executon of that algorthm. Usually, n well separated clusters, D data sets are used for evaluatng clusterng algorthms as the reader can easly verfy the result. But n case of hgh dmensonal data, the vsualzaton and vsual valdaton s not a trval tas. Therefore some formal methods are needed. The process of evaluatng the results of a clusterng algorthm s called cluster valdty assessment. For ths, there are three dfferent technques: external crtera, nternal crtera and relatve crtera. Both nternal and external crtera are based on statstcal methods and they have hgh computaton demand. A revew of clusterng valdty ndexes that are based on external and nternal crtera can be found n [6]. Also as was mentoned [7], the valdty assessment approaches based on relatve crtera wor well n non overlappng cases. If the data set consdered contans overlappng clusters, then, the maorty of the estng valdty ndex fals to detect the rght number of clusters. In ths paper, we propose a new and effcent measure to determne the optmal number of clusters, based on Mamum Entropy Prncple (MEP), whch not only handles effcently hgh degree overlapped cases, but also are completely automatc, requrng any parameter determnaton. We show also n some examples that t wors well also n non gaussan mxture models. The organzaton of the rest of the paper s as follows. In secton, we brefly revew some valdty measures related to our wor, and also present some of ther shortcomngs. In secton 3, the proposed measure based on MEP s presented and the correspondent algorthm. Secton 4 presents many examples usng artfcal and real data sets to demonstrate the effectveness of the proposed measure. Fnally, secton 5 concludes the paper.. Related Wor The Fuzzy C-Means (FCM) clusterng algorthm has been wdely used to obtan the fuzzy c-partton. Ths algorthm requres that the user predefne the number of clusters ; however, t s not always possble to now the number of clusters n advance. Dfferent fuzzy parttons are obtaned at dfferent values of. Thus, an evaluaton methodology s requred to

2 47 The Internatonal Arab Journal of Informaton Technology, Vol. 5, No. 4, October 008 valdate each of the fuzzy c-parttons and, to obtan an optmal partton (or optmal number of clusters c). Ths quanttatve evaluaton s the subect of cluster valdty. The mathematcal formula used to compute the valdaton s referred to as a cluster valdty ndex. Many clusters valdty ndexes for fuzzy clusterng are proposed n the lterature n order to fnd an optmal number of clusters. [3] two cluster valdty ndexes for fuzzy clusterng: Partton Coeffcent (V PC ) and Partton Entropy (V PE.) These ndexes V PC and V PE are senstve to nose or a weghtng exponent m. Other ndexes such as V FS and V XB whch tae nto account the geometrc propertes of nput data were proposed respectvely [5] and Xe-Ben [5].The V FS ndex s senstve to both hgh and low exponent m. V XB provded a good response over a wde range of choces both for c= to 0 and for <m 7. However, V XB decreases monotoncally as the number of clusters c becomes very large and close to the number of data n. [] extended Xe-Ben ndex V XB to elmnate ts monotonc decreasng tendency. To acheve ths, a punshng functon was ntroduced to the numerator part of Xe and Ben orgnal valdty ndex. [7] have defned a new valdty ndex V S_Dbw. Ths latter explots also the compactness and the separaton propertes of the data set. The compactness s measured by the cluster varance whereas the separaton by the densty between clusters. As was mentoned [8] the ndex V S_Dbw s optmzed for data sets that nclude compact and well-separated clusters that s n non overlappng cases. [0] attempted to determne the optmal number of clusters by measurng the status of the gven partton wth both an under-partton functon and an over-partton functon. The proposed ndex V SV s the sum of the two functons. V SV provdes enhanced performances when compared wth the prevous studes. More recently, a new valdty ndex V OS was proposed [], V OS explots an overlap measure and a separaton measure between clusters. The proposed ndex V OS was defned as the rato of the overlappng degree to the separaton. The overlap measure, whch ndcates the degree of overlap between fuzzy clusters, s obtaned by computng an nter-cluster overlap. The separaton measure, whch ndcates the solaton dstance between fuzzy clusters, s obtaned by computng a dstance between fuzzy clusters. As was mentoned [] the proposed ndex V OS s more relable than other ndexes. Unfortunately, from the tests on the IRIS data that have real overlappng clusters, the authors have seen that V OS does not dscrmnate the two overlappng clusters. 3. The Valdty Index For a gven data set, we obtan, after some clusterng process, a partton on clusters c c c. Now, defne P as a measure of the lns between any pont and the cluster c, for =. As all membershps of any of those clusters c are nown, we can set P =0 for c and, for c, P >0 are normalzed by: P =, for =. () c For all the clusters, we have: (3) P = = c = c P ( ) = The entropy of all the clusters s defned by: where = P P S ln S S s gven by: = c = = c (4) P ln ( P ) + ln( ) (5) S= S ln( ) + = (6) S = P ln c (7) ( P ) () S s the entropy correspondng to the cluster. Ths entropy wll be mamal when all the data ponts of each cluster have the same assocaton wth ther cluster centres. Therefore, the optmal number of clusters s the number whose value of entropy s mamal. In addton to mamzng the above entropy, we use another constrant whch wll be mnmzed. In ths second constrant, for each cluster, the nearest neghbour data ponts to the cluster centre wll be prvleged. The proposed constrant s gven by the followng formula: (8) W = P = c g where s the eucldean dstance, and x represents the pont. We are tryng to reach the hgher possble concentraton around or near each cluster centre.

3 Optmal Fuzzy Clusterng n Overlappng Clusters 48 To satsfy the above two constrans, that s to mamze S whle mnmzng W, s equvalent to mnmze the followng expresson: T = = c T = W S () P ln ( P ) ln( ) (0) under constrants: P = c + P g = c for =,... The lagrange optmzng the formula gven n equaton 0 under the constrants s gven by L = + = c = c P P ln( P ) ln( ) x + g α P = c () where α s the lagrange multplcator assocated to th constrant. We then annul the dervaton of L per P : ln( P ) + g () + + α = 0 we can then gve the expressons of P for =...N, and =... by the followng: [ P Z exp g ] = (3) where Z s a normalzaton coeffcent gven by: Z = exp ( + α ) By replacng the expresson of P gven by (3) n the correspondng constrant expresson, we obtan the expresson of Z gven below: Z = c exp x (4) g then P coeffcents can be computed by: c [ g ] [ g ] exp P = exp (5) now, we defne, our proposed ndex V MEP as the whole entropy: V MEP = S = S ln( ) + = (6) where S s defned by equaton 7 whch use P defned n equaton 5. The optmal number of clusters s then the number whose value of V MEP s mamal. The proposed new algorthm usng the new ndex V MEP. We propose n ths secton our new general algorthm based on the novel ndex V MEP. The optmal number of clusters s the number whose value of V MEP s mamal. The steps of the algorthm are: A. Fx the mamal number of classes K max B. K K max C. Do whle, C..Applcaton of clusterng algorthm (exp: call Fuzzy C-means or -means to defne the classes c c c. and determne the g centres, for =,,) C..Compute the P probablty wth formula 5 C.3.Compute the S entropy wth formula 7 C.4.K K- End D. V MEP = max S, for =, K max. ( The correct number of clusters s then the for whch the mamum s due) 4. Expermental Results To test the performance of proposed valdty V MEP, we use t to determne the optmal number of clusters n some of synthetc data and also n a well nown real data set. However, n earler publcatons, V SV, proposed [0], was compared wth the followng valdty ndexes V PC, V PE, V FS, V XB, V K and V S-Dbw. It provdes enhanced performances; and n the prevous wor of one of the authors [3] t was also mplemented and used to fnd the optmal number of clusters usng Gaussan Mxture Model (GMM), and the EM algorthm for clusterng process, ths scheme was successfully appled to extract the desgn regons n color textle mage. Therefore, n ths nvestgaton, we wll ust compare our proposed ndex V MEP to V SV. In frst, we revew some applcatons of V MEP to GMM. We generate sxteen artfcal data sets. The frst one,, s le the well nown four polonase balls [4], Fgure shows the scatter plot of ths data set; t has 4 compact and well-separated clusters algned n dagonal. Each cluster was generated usng normal dstrbuton; the parameters used for generatng ths data set are gven n Table. Table. Parameters used for generatng. Cluster Number of Covarance Mean Vector Number Ponts Matrx Cluster 000 (-4; -4) ( 0; 0 ) Cluster 000 (0; 0) ( 0; 0 ) Cluster ( 0; 0 )

4 4 The Internatonal Arab Journal of Informaton Technology, Vol. 5, No. 4, October 008 Cluster (8; 8) ( 0; 0 ) The others ffteen Data Sets:,, 5 and 6, are derved from the frst one, by producng a two overlappng clusters wth dfferent degree of overlap. We move the coordnates centre of cluster havng as coordnates centre (0, 0) as shown n Table, to a seres of the followng centers coordnates (, ), (.5;.5), (.6;.6), (.7;.7), (.8;.8), (; ), (.5;.5), (.;.), (3; 3), (3.5; 3.5), (3.5; 3.5), (3.6; 3.6), (3.7; 3.7), (3.; 3.), and fnally whch are the coordnates centre of cluster 3, as shown n Table. We apply V SV and V MEP to these data sets, and we see f V MEP can perform V SV? If yes, how well does t, and up what lmt?. The cluster valdaton results usng V SV and V MEP are shown n Fgure. For the, havng well-separated clusters, both V SV and V MEP can select correctly 4 as optmal number of clusters. For the to 4, whch have two overlappng clusters wth low degree of overlap, also both V SV and V MEP select correctly 4 as the optmal number of clusters. For 5, V SV select 3 whch s a falure result. By ncreasng the degree of overlap n 6, 7, V SV also fals, t select 3 whch s not a correct optmal number of clusters. Instead, V MEP selects correctly 4 clusters for all these data sets (5, 6, and 7). (0;0) (;) (,7;,7) Index VSV Index Index VMEP Fgure. Results of clusters valdaton usng V SV (mnmal value) and V MEP (mamal value), dsplayed from to 7. From the above results, we conclude that V SV can wor correctly only n the presence of a low degree of overlap, and t produces a falure results when dealng wth relatvely a hgh overlappng degree. We then stop to apply V SV to data sets havng a superor overlappng degree such as 8 6; and we contnue to apply only V MEP. The result of applyng V MEP to the 8 to 3, are presented respectvely n Fgure. V MEP can stll wor well; t selects correctly 4 as the optmal number of clusters. In 4 6, the centre coordnates of the moved cluster number are respectvely (3.7; 3.7), (3.; 3.), and. These centers are very close to those of the fxed cluster number 3 whose coordnates centre are. Ths yelds a very hgh overlappng degree. In ths case, we can see n Fgure 3 that the two overlappng clusters represent appromately one cluster. V MEP can not select 4 as optmal number of clusters, t select 3 clusters whch can be consdered as evdent result. 8 (.5;.5) (. ;. ) 0 (3; 3,8,7,6,5,4,3,,,8,7,6,5,4,3,, ndex VME,,8,7,6,5,4,3,, (,8;,8) Num ber of Clusters Index VMPE (3.5; 3.5, 0,8,7,6,5,4,3,, (;) 5 (3,6;3,6) 6 (3,7;3,7) Num ber of Clusters Index VMEP (3.5; 3.5) 3 (3.6; 3.6 ), 0,8,7,6,5,4,3,,, 0,8,7,6,5,4,3,,

5 Optmal Fuzzy Clusterng n Overlappng Clusters 50 Fgure. Results of clusters valdaton usng the proposed V MEP, dsplayed from 8 to 3 Then, we see that V MEP performs clearly V SV, t can stll wor well and select the correct optmal number of clusters for all data sets up 3, as shown n Fgures and 3. From the data sets 4 up 6, as shown n Fgure 3, V MEP can not select 4 as optmal number of clusters; because the clusters number and 3 are extremely overlapped, and they may be regarded as one cluster. We conclude that V SV can wor correctly only n the presence of a low degree of overlap, and t produces a falure results when dealng wth relatvely hgh overlappng degree, whle V MEP stll wors well, and selects correctly 4 as optmal number of clusters wth very hgh overlap. We conclude that V MEP performs clearly V SV for GMM as verfed n our early wor []. The performance of V MEP s also examned usng a well nown real Irs data set []. Results are shown n Fgure 4. Both the two ndex V MEP and V SV select correctly 3 as optmal number of clusters. Here, V SV can wor well because the low degree of overlap. 4 (3.7; 3.7) 5 (3.; 3. ) 6 (4; 4 ) Fgure 3. Results of clusters valdaton usng the proposed V MEP, dsplayed from 4 to 6,8,7,6,5,4,3,,,,8,7,6,5,4,3,,,75,5,5 8,75 8,5 0 respectvely Banana set, Banana set, Banana set3, Banana set4. In all of them, V MEP detects the correct and real number of clusters. Banana set descrbe two banana forms enclosed nto one crcle whch s wrapped by one banana form. The result of applyng V MEP to the Banana set shows that t can select 4 clusters whch s the correct number of clusters for banana set. For banana set, we stay the same two banana forms enclosed now n two symmetrc banana forms wth same centre but wth dfferent radus. In ths case V MEP can select also 4 clusters whch s the correct number of clusters for banana set. The llustraton of the banana set3 show two symmetrc banana forms wth same centre and same radus. We eep nto them the same two banana forms enclosed n banana set and banana set. V MEP wors also well and selects 3 clusters whch s the logc and correct number of clusters. Fnally, we test our new ndex on a combnaton of dfferent forms and overlappng case. The result of ths applcaton showng n graphc BSet4 s very nterestng. V MEP can detect 5 clusters whch s the correct number of clusters. Ths last result completes the performance and the robustness of V MEP. BSet BSet BSet3 BSet Fgure 5. Results of clusters valdaton usng V MEP for some banana forms. Fgure 4. Results of clusters valdaton usng s ndex V SV (mnmal value), and the proposed V MEP (mamal value), appled to the Irs Data Set. Now, what about non GMM, Fgure 5 shows results when V MEP s appled to other forms le banana forms. In the present wor, we generate 4 banana forms named 5. Concluson We ntroduced n ths paper a new formulaton of a cluster valdty ndex for the valdaton of the fuzzy - parttons that are generated by the applcaton of the FCM clusterng algorthm. Ths new ndex can be playng an mportant role n solvng many problems n pattern recognton, ameloraton of the qualty of

6 5 The Internatonal Arab Journal of Informaton Technology, Vol. 5, No. 4, October 008 products n maretng. The proposed ndex V MEP s based on the MEP and the optmal number of clusters s the number * whose value of V MEP s mamal. The performance of our ndex V MEP was examned, n both our generated synthetc data sets and n real data example and a robustness of ths new ndex s completed by the extenson of the method to non-gmm wth overlap. The expermental results show the superorty of our measure V MEP to the estng ones and ts capacty to detect the rght number of clusters wth dfferent shapes and degree of overlap. Fnally, we report also another advantage of our ndex. The defnton of V MEP uses any parameter produced by the adopted clusterng algorthm. Therefore, V MEP s ndependent of any clusterng algorthm. Ths allows us to choose any one, such as Gustafson-Kessel (GK) algorthm whch can deals wth ellpsodal clusters, or EM clusterng algorthm. Ths wll be the subect of our next nvestgaton. References [] Ammor O., Lachar A., Slaou K., and Ras N., New Effcent Approach to Determne the Optmal n Overlappng Cases, n Proceedngs of the IEEE on Advances n Cybernetc Systems, UK, pp. 6-3, 006. [] Anderson E., The IRISes of the Gaspe Pennsula, Bulletn of the Amercan Irs Socety, vol. 5, no. 35, pp. -5, 5. [3] Bezde J., Cluster Valdty wth Fuzzy Sets, Cybernetcs and Systems, vol. 3, no. 3, pp. 58-7, 75. [4] Cembrzyns T., Banc D'essa Sur Les Boules Polonases, Des Tros Crtères de Décson Utlsés Dans La Procédure de Classfcaton, MNDOPT Pour Chosr un Nombre de Classes, RR-0784 Rapport de recherche de l'inria, 86. [5] Fuuyama Y. and Sugeno M., A New Method of Choosng the for the Fuzzy C-Means Method, n Proceedngs of the 5 th Fuzzy Systems Symposum, Japan, pp , 8. [6] Hald M., Batstas Y., and Vazrganns M., Qualty Scheme Assessment n the Clusterng Process, n Proceedngs of the 4 th European Conference on Prncples of Data Mnng and Knowledge Dscovery, London, pp , 000. [7] Hald M. and Vazrganns M., Clusterng Valdty Assessment: Fndng the Optmal Parttonng of a Data Set, n Proceedngs of st IEEE Internatonal Conference on Data Mnng (ICDM'00), USA, pp. 87-4, 00. [8] Hald M., Batstas Y., and Vazrganns M., Clusterng Valdty Checng Methods: Part II, Specal Interest Group on Management of Data (SIGMOD), vol. 3, no. 3, pp. -7, 00. [] Jan A., Murty M., and Flynn P., Data Clusterng a Revew, ACM Computng Surveys, vol. 3, no. 3, pp ,. [0] Km D., Par Y., and Par D., A Novel Valdty Index for Determnaton of the Optmal Number of Clusters, IEICE Transactons on Informaton System, vol. D-E84, no., pp. 8-85, 00. [] Km D., Kwang A, Lee H., and Lee D., On Cluster Valdty Index for Estmaton of the Optmal Number of Fuzzy Clusters, Pattern Recognton, vol. 37, pp , 004. [] Kwon S., Cluster Valdty Index for Fuzzy Clusterng, Pattern Recognton Letters, vol. 34, no., pp , 8. [3] Lachar A., Benslmane R., D Orazo L., and Martuscell E., A System for Textle Desgn Patterns Retreval Part : Desgn Patterns Extracton by Adaptve and Effcent Color Image Segmentaton Method, Journal of the Textle Insttute, no ot, 4r, 005. [4] Lu J. and Yang Y., Multresoluton Color Image Segmentaton, IEEE Transactons on Pattern Analyss and Machne Intellgence, vol. 6, no. 7, pp , 4. [5] Lu X. and Ben G., A Valdty Measure for Fuzzy Clusterng, IEEE Transactons on Pattern Analyss Machne Intellgence, vol. 3, no. 8, pp ,. [6] Sharma S., Appled Multvarate Technques, John Wley and Sons, 6. [7] Trved M. and Bezde J., Low-Level Segmentaton of Aeral Images wth Fuzzy Clusterng, IEEE Transactons on Systems, Man and Cybernetcs, vol. 6, no. 4, pp , 86. Ouafae Ammor s a professor n the Department of Mathematcs at Faculty of Scences and Technology of Fes, Morocco. She has obtened her Master s degree n statstcs from Polytechnque school of Montreal, Canada. Her research focuses on the clusterng n data analyss, pattern recognton, colour mage segmentaton,

7 Optmal Fuzzy Clusterng n Overlappng Clusters 5 medcal mage processng, fuzzy logc and spatal data analyss. Abdelmoname Lachar receved the Master degree n automatc and systems analyss n, and hs PhD degree from the USMBA, Morocco n 004. In computer scences. He s assocate professor n Computer Scences n ESTM at Moulay Ismal Unversty-Morocco. Hs current research nterests nclude mage ndeng and retreval, shape representaton, ndeng and retreval n large shapes databases, colour mage segmentaton, unsupervsed clusterng, cluster valdty ndex, pattern recognton, arabc and latn handwrtten recognton, document clusterng and categorsaton, medcal mage processng, mage compresson and watermarng. Khada Slaou s a professor n the Department of Physcs at Faculty of Scences Dhar Mehraz of Fes, Morocco. She has obtened her Master s n mage processng at Polytechnque Insttute of Toulouse, France, and PHD n data analyss at Unversty Sd Mohammed Ben Abdellah. Her research focuses on sgnal processng, clusterng, and mages processng. Noureddne Ras s full professor at the Unversty of Fez n Morocco. He has a PhD n mathematcal statstcs from the Unversty of Montreal, Canada and a 3 rd Cycle Doctorate n Statstcs and Mathematcal Models from Unversty of Pars-sud, Orsay, France. Hs current research nterests covers bootstrap, spatal data, survey samplng, decson theory, data analyss, data mnng, expermental desgn, lnear models and analyss of varance, wth applcatons n many felds, specally statstcal processng control, arabc handwrtten recognton, and medcal mage processng.

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

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

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

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

Determining Fuzzy Sets for Quantitative Attributes in Data Mining Problems

Determining Fuzzy Sets for Quantitative Attributes in Data Mining Problems Determnng Fuzzy Sets for Quanttatve Attrbutes n Data Mnng Problems ATTILA GYENESEI Turku Centre for Computer Scence (TUCS) Unversty of Turku, Department of Computer Scence Lemmnkäsenkatu 4A, FIN-5 Turku

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

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

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

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

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

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

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data A Fast Content-Based Multmeda Retreval Technque Usng Compressed Data Borko Furht and Pornvt Saksobhavvat NSF Multmeda Laboratory Florda Atlantc Unversty, Boca Raton, Florda 3343 ABSTRACT In ths paper,

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

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

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

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

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

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

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

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

A New Measure of Cluster Validity Using Line Symmetry *

A New Measure of Cluster Validity Using Line Symmetry * JOURAL OF IFORMATIO SCIECE AD EGIEERIG 30, 443-461 (014) A ew Measure of Cluster Valdty Usng Lne Symmetry * CHIE-HSIG CHOU 1, YI-ZEG HSIEH AD MU-CHU SU 1 Department of Electrcal Engneerng Tamkang Unversty

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

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

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

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

Image Alignment CSC 767

Image Alignment CSC 767 Image Algnment CSC 767 Image algnment Image from http://graphcs.cs.cmu.edu/courses/15-463/2010_fall/ Image algnment: Applcatons Panorama sttchng Image algnment: Applcatons Recognton of object nstances

More information

From Comparing Clusterings to Combining Clusterings

From Comparing Clusterings to Combining Clusterings Proceedngs of the Twenty-Thrd AAAI Conference on Artfcal Intellgence (008 From Comparng Clusterngs to Combnng Clusterngs Zhwu Lu and Yuxn Peng and Janguo Xao Insttute of Computer Scence and Technology,

More information

Detection of an Object by using Principal Component Analysis

Detection of an Object by using Principal Component Analysis Detecton of an Object by usng Prncpal Component Analyss 1. G. Nagaven, 2. Dr. T. Sreenvasulu Reddy 1. M.Tech, Department of EEE, SVUCE, Trupath, Inda. 2. Assoc. Professor, Department of ECE, SVUCE, Trupath,

More information

An Entropy-Based Approach to Integrated Information Needs Assessment

An Entropy-Based Approach to Integrated Information Needs Assessment Dstrbuton Statement A: Approved for publc release; dstrbuton s unlmted. An Entropy-Based Approach to ntegrated nformaton Needs Assessment June 8, 2004 Wllam J. Farrell Lockheed Martn Advanced Technology

More information

Analysis of Continuous Beams in General

Analysis of Continuous Beams in General Analyss of Contnuous Beams n General Contnuous beams consdered here are prsmatc, rgdly connected to each beam segment and supported at varous ponts along the beam. onts are selected at ponts of support,

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

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

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

NIVA: A Robust Cluster Validity

NIVA: A Robust Cluster Validity 2th WSEAS Internatonal Conference on COMMUNICATIONS, Heralon, Greece, July 23-25, 2008 NIVA: A Robust Cluster Valdty ERENDIRA RENDÓN, RENE GARCIA, ITZEL ABUNDEZ, CITLALIH GUTIERREZ, EDUARDO GASCA, FEDERICO

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

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

An Accurate Evaluation of Integrals in Convex and Non convex Polygonal Domain by Twelve Node Quadrilateral Finite Element Method

An Accurate Evaluation of Integrals in Convex and Non convex Polygonal Domain by Twelve Node Quadrilateral Finite Element Method Internatonal Journal of Computatonal and Appled Mathematcs. ISSN 89-4966 Volume, Number (07), pp. 33-4 Research Inda Publcatons http://www.rpublcaton.com An Accurate Evaluaton of Integrals n Convex and

More information

Applying EM Algorithm for Segmentation of Textured Images

Applying EM Algorithm for Segmentation of Textured Images Proceedngs of the World Congress on Engneerng 2007 Vol I Applyng EM Algorthm for Segmentaton of Textured Images Dr. K Revathy, Dept. of Computer Scence, Unversty of Kerala, Inda Roshn V. S., ER&DCI Insttute

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

Type-2 Fuzzy Non-uniform Rational B-spline Model with Type-2 Fuzzy Data

Type-2 Fuzzy Non-uniform Rational B-spline Model with Type-2 Fuzzy Data Malaysan Journal of Mathematcal Scences 11(S) Aprl : 35 46 (2017) Specal Issue: The 2nd Internatonal Conference and Workshop on Mathematcal Analyss (ICWOMA 2016) MALAYSIAN JOURNAL OF MATHEMATICAL SCIENCES

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

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

FEATURE EXTRACTION. Dr. K.Vijayarekha. Associate Dean School of Electrical and Electronics Engineering SASTRA University, Thanjavur

FEATURE EXTRACTION. Dr. K.Vijayarekha. Associate Dean School of Electrical and Electronics Engineering SASTRA University, Thanjavur FEATURE EXTRACTION Dr. K.Vjayarekha Assocate Dean School of Electrcal and Electroncs Engneerng SASTRA Unversty, Thanjavur613 41 Jont Intatve of IITs and IISc Funded by MHRD Page 1 of 8 Table of Contents

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

Image Representation & Visualization Basic Imaging Algorithms Shape Representation and Analysis. outline

Image Representation & Visualization Basic Imaging Algorithms Shape Representation and Analysis. outline mage Vsualzaton mage Vsualzaton mage Representaton & Vsualzaton Basc magng Algorthms Shape Representaton and Analyss outlne mage Representaton & Vsualzaton Basc magng Algorthms Shape Representaton and

More information

A Robust Method for Estimating the Fundamental Matrix

A Robust Method for Estimating the Fundamental Matrix Proc. VIIth Dgtal Image Computng: Technques and Applcatons, Sun C., Talbot H., Ourseln S. and Adraansen T. (Eds.), 0- Dec. 003, Sydney A Robust Method for Estmatng the Fundamental Matrx C.L. Feng and Y.S.

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

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

Using Fuzzy Logic to Enhance the Large Size Remote Sensing Images

Using Fuzzy Logic to Enhance the Large Size Remote Sensing Images Internatonal Journal of Informaton and Electroncs Engneerng Vol. 5 No. 6 November 015 Usng Fuzzy Logc to Enhance the Large Sze Remote Sensng Images Trung Nguyen Tu Huy Ngo Hoang and Thoa Vu Van Abstract

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

1. Introduction. Abstract

1. Introduction. Abstract Image Retreval Usng a Herarchy of Clusters Danela Stan & Ishwar K. Seth Intellgent Informaton Engneerng Laboratory, Department of Computer Scence & Engneerng, Oaland Unversty, Rochester, Mchgan 48309-4478

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

A NEW FUZZY C-MEANS BASED SEGMENTATION STRATEGY. APPLICATIONS TO LIP REGION IDENTIFICATION

A NEW FUZZY C-MEANS BASED SEGMENTATION STRATEGY. APPLICATIONS TO LIP REGION IDENTIFICATION A NEW FUZZY C-MEANS BASED SEGMENTATION STRATEGY. APPLICATIONS TO LIP REGION IDENTIFICATION Mhaela Gordan *, Constantne Kotropoulos **, Apostolos Georgaks **, Ioanns Ptas ** * Bass of Electroncs Department,

More information

An Application of Fuzzy c-means Clustering to FLC Design for Electric Ceramics Kiln

An Application of Fuzzy c-means Clustering to FLC Design for Electric Ceramics Kiln An Applcaton of cmeans Clusterng to FLC Desgn for lectrc Ceramcs Kln Watcharacha Wryasuttwong, Somphop Rodamporn lectrcal ngneerng Department, Faculty of ngneerng, Srnaharnwrot Unversty, Nahornnayo 6,

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

Shape Representation Robust to the Sketching Order Using Distance Map and Direction Histogram

Shape Representation Robust to the Sketching Order Using Distance Map and Direction Histogram Shape Representaton Robust to the Sketchng Order Usng Dstance Map and Drecton Hstogram Department of Computer Scence Yonse Unversty Kwon Yun CONTENTS Revew Topc Proposed Method System Overvew Sketch Normalzaton

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

Intra-Parametric Analysis of a Fuzzy MOLP

Intra-Parametric Analysis of a Fuzzy MOLP Intra-Parametrc Analyss of a Fuzzy MOLP a MIAO-LING WANG a Department of Industral Engneerng and Management a Mnghsn Insttute of Technology and Hsnchu Tawan, ROC b HSIAO-FAN WANG b Insttute of Industral

More information

Accounting for the Use of Different Length Scale Factors in x, y and z Directions

Accounting for the Use of Different Length Scale Factors in x, y and z Directions 1 Accountng for the Use of Dfferent Length Scale Factors n x, y and z Drectons Taha Soch (taha.soch@kcl.ac.uk) Imagng Scences & Bomedcal Engneerng, Kng s College London, The Rayne Insttute, St Thomas Hosptal,

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

Fuzzy Filtering Algorithms for Image Processing: Performance Evaluation of Various Approaches

Fuzzy Filtering Algorithms for Image Processing: Performance Evaluation of Various Approaches Proceedngs of the Internatonal Conference on Cognton and Recognton Fuzzy Flterng Algorthms for Image Processng: Performance Evaluaton of Varous Approaches Rajoo Pandey and Umesh Ghanekar Department of

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

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

EYE CENTER LOCALIZATION ON A FACIAL IMAGE BASED ON MULTI-BLOCK LOCAL BINARY PATTERNS

EYE CENTER LOCALIZATION ON A FACIAL IMAGE BASED ON MULTI-BLOCK LOCAL BINARY PATTERNS P.G. Demdov Yaroslavl State Unversty Anatoly Ntn, Vladmr Khryashchev, Olga Stepanova, Igor Kostern EYE CENTER LOCALIZATION ON A FACIAL IMAGE BASED ON MULTI-BLOCK LOCAL BINARY PATTERNS Yaroslavl, 2015 Eye

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

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

A B-Snake Model Using Statistical and Geometric Information - Applications to Medical Images

A B-Snake Model Using Statistical and Geometric Information - Applications to Medical Images A B-Snake Model Usng Statstcal and Geometrc Informaton - Applcatons to Medcal Images Yue Wang, Eam Khwang Teoh and Dnggang Shen 2 School of Electrcal and Electronc Engneerng, Nanyang Technologcal Unversty

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

Lecture 4: Principal components

Lecture 4: Principal components /3/6 Lecture 4: Prncpal components 3..6 Multvarate lnear regresson MLR s optmal for the estmaton data...but poor for handlng collnear data Covarance matrx s not nvertble (large condton number) Robustness

More information

THE FUZZY GROUP METHOD OF DATA HANDLING WITH FUZZY INPUTS. Yuriy Zaychenko

THE FUZZY GROUP METHOD OF DATA HANDLING WITH FUZZY INPUTS. Yuriy Zaychenko 206 5 Knowledge Dalogue - Soluton THE FUZZY GROUP ETHOD OF DATA HANDLING WITH FUZZY INPUTS Yury Zaycheno Abstract: The problem of forecastng models constructng usng expermental data n terms of fuzzness,

More information

Cordial and 3-Equitable Labeling for Some Star Related Graphs

Cordial and 3-Equitable Labeling for Some Star Related Graphs Internatonal Mathematcal Forum, 4, 009, no. 31, 1543-1553 Cordal and 3-Equtable Labelng for Some Star Related Graphs S. K. Vadya Department of Mathematcs, Saurashtra Unversty Rajkot - 360005, Gujarat,

More information

Clustering algorithms and validity measures

Clustering algorithms and validity measures Clusterng algorthms and valdty measures M. Hald, Y. Batstas, M. Vazrganns Department of Informatcs Athens Unversty of Economcs & Busness Emal: {mhal, yanns, mvazrg}@aueb.gr Abstract Clusterng ams at dscoverng

More information

Lecture 13: High-dimensional Images

Lecture 13: High-dimensional Images Lec : Hgh-dmensonal Images Grayscale Images Lecture : Hgh-dmensonal Images Math 90 Prof. Todd Wttman The Ctadel A grayscale mage s an nteger-valued D matrx. An 8-bt mage takes on values between 0 and 55.

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

MULTISPECTRAL REMOTE SENSING IMAGE CLASSIFICATION WITH MULTIPLE FEATURES

MULTISPECTRAL REMOTE SENSING IMAGE CLASSIFICATION WITH MULTIPLE FEATURES MULISPECRAL REMOE SESIG IMAGE CLASSIFICAIO WIH MULIPLE FEAURES QIA YI, PIG GUO, Image Processng and Pattern Recognton Laboratory, Bejng ormal Unversty, Bejng 00875, Chna School of Computer Scence and echnology,

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

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

EVALUATION OF THE PERFORMANCES OF ARTIFICIAL BEE COLONY AND INVASIVE WEED OPTIMIZATION ALGORITHMS ON THE MODIFIED BENCHMARK FUNCTIONS

EVALUATION OF THE PERFORMANCES OF ARTIFICIAL BEE COLONY AND INVASIVE WEED OPTIMIZATION ALGORITHMS ON THE MODIFIED BENCHMARK FUNCTIONS Academc Research Internatonal ISS-L: 3-9553, ISS: 3-9944 Vol., o. 3, May 0 EVALUATIO OF THE PERFORMACES OF ARTIFICIAL BEE COLOY AD IVASIVE WEED OPTIMIZATIO ALGORITHMS O THE MODIFIED BECHMARK FUCTIOS Dlay

More information

Clustering Algorithm of Similarity Segmentation based on Point Sorting

Clustering Algorithm of Similarity Segmentation based on Point Sorting Internatonal onference on Logstcs Engneerng, Management and omputer Scence (LEMS 2015) lusterng Algorthm of Smlarty Segmentaton based on Pont Sortng Hanbng L, Yan Wang*, Lan Huang, Mngda L, Yng Sun, Hanyuan

More information

MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION

MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION Paulo Quntlano 1 & Antono Santa-Rosa 1 Federal Polce Department, Brasla, Brazl. E-mals: quntlano.pqs@dpf.gov.br and

More information

Graph-based Clustering

Graph-based Clustering Graphbased Clusterng Transform the data nto a graph representaton ertces are the data ponts to be clustered Edges are eghted based on smlarty beteen data ponts Graph parttonng Þ Each connected component

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

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

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

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

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

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

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

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

Object-Based Techniques for Image Retrieval

Object-Based Techniques for Image Retrieval 54 Zhang, Gao, & Luo Chapter VII Object-Based Technques for Image Retreval Y. J. Zhang, Tsnghua Unversty, Chna Y. Y. Gao, Tsnghua Unversty, Chna Y. Luo, Tsnghua Unversty, Chna ABSTRACT To overcome the

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

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

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

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

BAYESIAN MULTI-SOURCE DOMAIN ADAPTATION

BAYESIAN MULTI-SOURCE DOMAIN ADAPTATION BAYESIAN MULTI-SOURCE DOMAIN ADAPTATION SHI-LIANG SUN, HONG-LEI SHI Department of Computer Scence and Technology, East Chna Normal Unversty 500 Dongchuan Road, Shangha 200241, P. R. Chna E-MAIL: slsun@cs.ecnu.edu.cn,

More information

Fuzzy C-Means Initialized by Fixed Threshold Clustering for Improving Image Retrieval

Fuzzy C-Means Initialized by Fixed Threshold Clustering for Improving Image Retrieval Fuzzy -Means Intalzed by Fxed Threshold lusterng for Improvng Image Retreval NAWARA HANSIRI, SIRIPORN SUPRATID,HOM KIMPAN 3 Faculty of Informaton Technology Rangst Unversty Muang-Ake, Paholyotn Road, Patumtan,

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

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

High-Boost Mesh Filtering for 3-D Shape Enhancement

High-Boost Mesh Filtering for 3-D Shape Enhancement Hgh-Boost Mesh Flterng for 3-D Shape Enhancement Hrokazu Yagou Λ Alexander Belyaev y Damng We z Λ y z ; ; Shape Modelng Laboratory, Unversty of Azu, Azu-Wakamatsu 965-8580 Japan y Computer Graphcs Group,

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