A Scatter Search Approach for the Minimum Sum-of- Squares Clustering Problem

Size: px
Start display at page:

Download "A Scatter Search Approach for the Minimum Sum-of- Squares Clustering Problem"

Transcription

1 A Scatter Search Approach for the Mnmum Sum-of- Squares Clusterng Problem Joaquín A. Pacheco Department of Appled Economcs, Unversty of Burgos, Plaza Infanta Elena s/n BURGOS 09001, SPAIN Tf: ; fx ; emal: Latest verson: September 19, 003 Abstract A metaheurstc procedure based on the Scatter Search approach s proposed for the non-herarchcal clusterng problem under the crteron of mnmum Sum-of-Squares Clusterng. Ths algorthm ncorporates procedures based on dfferent strateges, such as Local Search, GRASP, Tabu Search or Path Relnkng. The am s to obtan qualty solutons wth short computaton tmes. A seres of computatonal experments has been performed. The proposed algorthm obtans better results than prevously reported methods, especally wth small numbers of clusters. Keywords: Clusterzaton, Metaheurstcs, Scatter Search, Local Search, GRASP, Tabu Search, Path Relnkng 1. Introducton Consder a set X = {x 1, x,..., x N } of N ponts n R q and let m be a predetermned postve nteger. The Mnmum Sum-of-Squares Clusterng (MSSC) problem s to fnd a partton of X nto m dsont subsets (clusters) so that the sum of squared dstances from each pont to the centrod of ts cluster s mnmum. Specfcally, let P m denote the set of all the parttons of X n m sets, where each partton PA P m s defned as PA = {C 1, C,..., C m } and where C denotes each of the clusters that forms PA. Thus, the problem can be expressed as: where the centrod x s defned as The problem can be wrtten as PA m mn x x, P x m = 1 x C l xl C l = 1 xl, wth n = C. n N mn x x k, l= 1 where k l s the cluster to whch pont x l belongs. The desgn of clusters s a well known exploratory Data Analyss ssue called Pattern Recognton. The am s to fnd whether a gven set of cases X has some structure and, n f so, to dsplay t n the form of a partton. Ths problem belongs to the area of Non- Herarchcal cluster desgn, whch has many applcatons n economcs, socal and natural scences. It s known to be NP-Hard [5]. l l

2 Varous exact methods for MSSC can be found n the lterature (see, for example [17] and [7]), some of them, such as the method proposed by du Merle et al. [8], have succeeded n resolvng problems wth up to 150 ponts. For larger-szed problems the use of heurstc algorthms s stll necessary. The most popular are those based on Local Search methods, such as the well-known K-Means [15] and H-Means [14] procedures. In a recent work, Hansen and Mladenovc [13] propose a new Local Search procedure, J-Means, along wth varants H-Means+ or HK-Means. In recent years algorthms usng Metaheurstc strateges have been desgned, such as Smulated Annealng [16], Tabu Search [], Genetc Algorthms [3] or most recently Varable Neghborhood Search or VNS [8], [13] and Memetc Algorthms [19]. An algorthm that s able to obtan good solutons n short tmes s proposed for ths problem. Ths method s based n the a recent Metaheurstc strategy named Scatter Search (SS). Ths method also ncorporates others procedures based n others methods, such as Local Search, Tabu Search, GRASP and Path Relnkng. Ths Scatter Search approach s analyzed and compared wth other recents technques. In all cases, our proposed technque gves adequate solutons, compared wth others recent technques, n reasonable tme, especally wth small values of m.. Soluton Approach The soluton approach that we have developed for the MSSC problem conssts of an adaptaton of scatter search. Scatter Search s an nstance of the so-called evolutonary methods, whch s not based solely on randomzaton as the man mechansm for searchng. SS has been successfully mplemented n a varety of settngs ncludng combnatoral optmzaton and nonlnear optmzaton n contnuous varables. Scatter search embodes prncples and strateges that are stll not emulated by other evolutonary methods, and that prove advantageous for solvng a varety of complex optmzaton problems. More about the orgn and multple applcatons of scatter search can be found n Glover [11], Glover, Laguna and Martí [1] and Laguna [18]. SS uses a set of solutons named Reference Set, (RefSet). RefSet s composed by the best solutons by qualty (subset RefSet1) and dversty (subset RefSet). In each teraton new solutons are generated from those of Reference Set, and then Reference Set s updated wht these new solutons. A statc verson s desgned for ths problem. An outlne of ths mplementaton s shown below. Procedure Statc_Scatter_Search Step 1. Generate an ntal set of solutons P by usng a Dversfcaton- Generaton Method Step. Improve these solutons by an Improvement Method Step 3. Wth these solutons buld an ntal RefSet Step 4. Repeat 4.1. Obtan all subsets of pars from RefSet 4.. Combne these subsets and obtan new solutons 4.3. Improve these solutons by the Improvement Method 4.4. Update RefSet wth these news solutons untl RefSet s stable (.e. no new solutons have been ncluded) {fnal 4.}

3 Step 5. If max_ter teratons (Steps 1-4) elapse wthout mprovement stop else return Step 1 Fgure 1. SS mplementaton Let s denote by n_pob the sze of P (Step 1). Also let s denote by b 1 and b the szes of RefSet1 and RefSet. For buldng the ntal RefSet (Step 3) frst the best solutons (by qualty) are taken from P. Then, the next solutons are added to RefSet by dversty. For that, the followng measure of dversty s used. Let λ be a soluton n P \ RefSet, we defne δ mn (λ) = mn {df(λ,λ ) / λ RefSet}; where df(λ,λ ) = number of assgnments n λ that are dfferent from λ. We then select the canddate soluton λ that maxmzes δ mn (λ). The updatng of the RefSet s based only on qualty. That s, only the new solutons that mprove the qualty of the worst soluton n RefSet are added, (Step 4.4). We now provde descrptons of the dversfcaton, mprovement and combnaton methods..1 Dversfcaton Method Our dversfcaton method s based on GRASP constructons. GRASP, or greedy randomzed adaptve search procedure, s a heurstc that constructs solutons wth controlled randomzaton and a greedy functon. Most GRASP mplementatons also nclude a local search that s used to mprove upon the solutons generated wth the randomzed greedy functon. GRASP was orgnally proposed n the context of a set coverng problem [9]. Detals of the methodology and a survey of applcatons can be found n Feo and Resende [10] and Ptsouls and Resende [0]. The method proposed n ths paper conssts of two stages. In the frst, a subset S wth ponts from X and suffcently far from each other (seed-ponts) s bult. Ths s denoted as S = {x s1, x s,, x sm }. We also use m one-pont clusters, correspondng to the seed ponts,.e., C = {x s }, =1,,m. In the second stage, the remanng ponts are assgned to dfferent clusters accordng to the dstance to ther centrods. The set S s determned as follows: - Determne x *, the farthest pont from the centrod of X, and do S = {x * }. - Whle S < m do: 1 Calculate = mn{ x - x l : x l S}, x S.. Calculate max = max { : x S } 3. Buld L = {x / α max } 4. Choose x * L randomly and do S = S {x * }. The α parameter (0 α 1) controls the level of randomzaton for the greedy selectons. Randomzaton decreases as the value of α ncreases. Ths controlled randomzaton results n a samplng procedure where the best soluton found s typcally 3

4 better than the one found by settng α = 1. A udcous selecton of the value of α provdes a balance between dversfcaton and soluton qualty. The frst tme that the dversfcaton method s employed (Step 1 n Fgure 1), there s no hstory assocated wth the number of tmes each pont has been selected as seed pont. However, ths nformaton s valuable when the method s appled to rebuld the reference. The nformaton s stored n the followng array: freq() = number of tmes that pont x has been selected to belong to S n the executon of Step 0 and prevous executons of Step 8 of Fgure 1. The nformaton accumulated n freq() s used to modfy the values n the applcaton of the frst phase of dversfcaton method. The modfed evaluaton s: = β max freq freq () max where freq max = max { freq() : }. The modfed values are used to calculate max and execute the dversfcaton method. Large values of β encourage the selecton of seed ponts that have not been frequently made. The use of frequency nformaton wthn a dversfcaton method s nspred by Campos, et al. [6]. In phase of our dversfcaton method, the ponts of X \ S are assgned as follows: 1. Set A = X \ S, (A = Set of unassgned ponts). For each pont x A and for each cluster C, = 1,, m, calculate the ncrease n obectve functon that results from assgnng pont x to cluster C. Ths value Γ s calculated as follows: n Γ = x x n + 1 where x s the centrod of C and n = C. 3. Calculate Γ ** = mn { Γ : x A, = 1,, m } 4. Assgn x * to C * and set A = A { x * } 5. If A return to, else stop..3 Improvement Method As an mprovement method the H-Means+ algorthm [13] s used together wth a smple Tabu Search procedure. The H-means+ algorthm s a varant of the well-known H- Means algorthm [14]. The Tabu Search procedure has been proposed by Pacheco and Valenca [19] and s descrbed shortly below. Ths Tabu Search procedure uses the neghborng moves employed n K-Means. These moves consst at each step n the movement of an entty from a cluster to a dfferent one. In order to avod repettve cyclng when a move whch conssts n movng pont x from cluster C l to cluster C s performed, pont x s 4

5 prevented from returnng to the cluster C l for a certan number of teratons. Specfcally, defne Matrx_tabu (l, ) = the number of the teraton n whch pont x leaves cluster C l. The Tabu Search method s descrbed below, where P denotes a ntal soluton wth a value f. The parameter τ ndcates the number of teratons durng whch a pont s not allowed to return to the leavng cluster. The parameter κ ndcates the maxmum number of unmproved teratons. In our case we use the stop crteron κ = 100. After dfferent tests, τ was set as m. Tabu Search 1 Do Matrx_tabu(,) = τ, =1,...,m, = 1,,N Do δ = 0 and P* = P, f* = f and η =0; 3 Repeat (3.1) δ = δ + 1 (3.) Determne v ** = mn {v / =1,...,m; = 1,...,N ; x C verfyng nter > Matrx_tabu (,) + τ or f + v < f* ( aspraton crteron )} (3.3) Reassgn x * to C * ; (3.4) Do Matrx_tabu (l*,*) = δ (l* beng the prevous cluster of x * ); (3.5) If f (the value of the current soluton P) < f* then do: P* = P, f* = f and η = δ ; untl (δ η > κ) or another termnaton crteron Where v s the change n the value of the obectve functon when x s reassgned to C. The followng formula s obtaned from Späth [] to smplfy the calculatons n K- Means. Let C l be the cluster to whch x belongs, then the value of v s calculated as follows: n nl v = x x x l x. n + 1 n 1.4 Combnaton Method New solutons are generated from combnng pars of reference set solutons. The number of solutons generated from each combnaton depends on the relatve qualty of the solutons beng combned. Let λ and λ be the reference solutons beng combned, where <. Assume, as before, that the reference set s ordered n a way that λ 1 s the best soluton and λ b s the worst. Then, the number of solutons generated from each combnaton s: 3 f b 1 and b 1, f b 1 and > b 1 and 1 f > b 1 and > b 1. Each subset wth two elements of RefSet s used to generate new solutons. To do so, we use a strategy called Path Relnkng. The basc dea s as follows: a path to on the two ntal solutons s bult. A number of ntermedate ponts (solutons) from the path' or chan are selected as new solutons. The am s for the ntermedate ponts or solutons to be as equdstant as possble from each other. The mprovement method descrbed n Sect.. s then appled to these ntermedate solutons. Fgure depcts ths dea. l 5

6 λ λ Improvement λ* λ** Fgure.- Generaton of New Solutons by usng Path Relnkng From every par of solutons of the reference set, n fgure λ and λ, a path to on them s bult. Solutons n these ntermedate preselected postons wthn the path are selected and mproved. In ths way new solutons are generated, (n fgure λ* y λ**). In order to buld the path that ons λ and λ, ponts wth dfferent clusters are selected n the dfferent steps and the shft s carred out. At each step the best possble shft s selected. In ths way, the ntermedate solutons n each step have another element n common wth λ. Path Relnkng s a strategy tradtonally assocated wth the ntensfcaton phase of the Tabu Search. The underlyng dea s that n the path between two good solutons there should be solutons of smlar qualty (n some cases, even better solutons). See Glover, Laguna and Martí [1] for more detals. 3. Parameter Fne Tunng One of the most tme consumng tasks n the development of metaheurstc procedures for optmzaton s the tunng of search parameters. Adenso-Daz and Laguna [1] propose an automated parameter tunng system called CALIBRA that employs statstcal analyss technques and a local search procedure to create a systematc way of fne-tunng algorthms. The goal of CALIBRA s to provde an automated system to fne-tune algorthms, where a user needs only to specfy a tranng set of nstances and a measure of performance. CALIBRA works wth a set of problem nstances and a range of values for each search parameter. The procedure utlzes desgn of experments and heurstcs to search for the best parameter values. The qualty of a set of values s tested on the specfed set of problem nstances. CALIBRA s avalable at opalo.etsg.unov.es/~adenso/fle_d.html, where a user s manual can also be found. Our complete set of test problems conssts of 15 nstances, as we pont out n Secton 4. We selected a relatvely small tranng set of 3 problems, because we determned that ths sample was representatve and that the parameter values found wth CALIBRA would also perform well when appled to the entre test set. The parameters to be adusted were α and β n the (0.1, 0.9) range and b 1 and b n the (3, 7) range. The values for MaxIter and PSze were set to and 0, respectvely. CALIBRA obtaned 6

7 the followng values, after runnng for approxmately 5 hours on a Pentum III machne at 600 MHz: α = 0.8, β = 0.5, b 1 = 5 and b = 5. Although a computatonal tme of 5 hours may seems excessve, t actually represents a very reasonable effort when compared to manual parameter fne-tunng. We use these parameter values for all of the experments reported n Secton Computatonal Experments To compare the effcences of our Scatter Search algorthm and those of other recent strateges, a seres of tests s performed. Next the results of ths set of computatonal experments usng these algorthms are shown. The followng algorthms are tested: Memetc-HK: Memetc Algorthm proposed n Pacheco and Valenca [19], HybMem: Hybrd Algorthm proposed n Pacheco and Valenca [19], VNS-HK: the VNS Algorthm, proposed n Hansen and Mladenovc [13], wth HK-Means as the Local Search method VNS-J: the VNS Algorthm, wth J-Means as the Local Search method. SS: Scatter Search Algorthm proposed n Secton. For ths work we have used our own mplementaton of every algorthm. The HybMem algorthm uses an ntal populaton generated by the greedy-random method descrbed n Beltrán and Pacheco (001). The VNS-HK and VNS-J algorthms use the best of these solutons as the ntal soluton. The sets of problems nstances used n testng are: () the 575 () 1060 and () 3038 ponts of the plane taken from TSPLIB [1] data base. All the tests n the current work are performed on a personal computer wth a Pentum III 600 MHz processor. All the algorthms have been mplemented n Pascal usng Borland Delph 5.0. In () the runnng tme s lmted to 300 seconds for all the algorthms. We set m = 5, 10, 15, 0, 30, 40, and 100. In tables 1 the solutons obtaned for each algorthm are presented. Table 1. Results for 575 TSPLIB m Memetc HybMem VNS-HK VNS-J SS , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,4 7

8 In (), we set m = 10, 0, 30,..., 150. These test data were prevously used n Hansen and Mladenovc [13], where the best soluton known for every value of m, (except m = 40) s reported. These were obtaned on a SUN Ultra I System workstaton wth 10 mnutes computaton tme. The runnng tme s lmted to 600 seconds for all the algorthms. In table the solutons obtaned for each algorthm and the percent devaton wth respect to the best-known soluton (dvt) are presented. Table. Results and percent desvaton respect to the prevous best know soluton. m Memetc HybMem VNS-HK VNS-J SS dvt : 0 0 0,185 0, ,181 0,05 1,757 1, ,899 1,48 0,79,54 0, ,89,503,19,43 0, ,47 1,773 0,71 1, ,09 3,398 1,173 4,91 0, ,318 3,77 0,353 0, , , , , ,7 3,516 3,34 0,304,644 0, , , , ,3 4,33,711 0,13 4,18 0, , , , ,7 4,03 4,03 0,77 3,08 1, , , ,6 4,383 4,383 1,513 3,943 1, , , , , ,4 5,5 5,086 1,68,531 0, , , , , ,8 5,18 4,35 0,366 3,643 0,511 In (), we set m = 10, 0,, 50, 100, 150,..., 400. These test data were prevously used n Hansen and Mladenovc [13], where the best soluton known for every value of m, s reported. These were obtaned on a SUN Ultra I System workstaton wth 50 mnutes computaton tme. The runnng tme s lmted to 600 seconds for all the algorthms. In table 9 the solutons obtaned for each algorthm and the percent devaton wth respect to the best-known soluton (dvt) are presented. Table 3. Results and percent desvaton respect to the prevous best know soluton. m Memetc HybMem VNS-HK VNS-J SS

9 m Memetc HybMem VNS-HK VNS-J SS , , ,109 0,94 0,709 0, ,17 0 0,346 0,46-0, , , ,6-0,001-0,083-0,068-0,01-0, , , , ,7,604,109 0,937,564 0, , , , , ,4,665,665 0,419,539 0, , , , , ,1 3,54 3,54 0,397 3,3 1, , , , , ,3 3,769 3,769 1,151 3,446, , , , , ,4 3,49 3,49 0,808,946, , , , , ,3 4,6 4,144 1,384 3,955, , , , , ,94 3,988 3,988 1,766 3,757 3, , , , , ,37 3, 3, 1,567 3,18 1, , , , , ,5,39,39 1,05 1,97 1,44 Tables 1-3 yelds the follow observatons: - The HybMem and Memetc-HK algorthms yeld good solutons when the number of clusters, m, s low. However, the qualty of ther solutons deterorates as the value of m ncreases. - The VNS-J, and specally the VNS-HK algorthms become more compettve than the other algorthms as the value of m ncrease. For hgh values of m (m 50 n (), m 90 n (), m 00 n ()), the VNS-HK algorthm yelds the best soluton out of all the strateges proposed n 14 cases. - In 6 out of 41 cases, the SS algorthm yelds the best soluton. Furthermore, t s especally effcent wth lower m values (m < 50 n (), m < 90 n (), m < 00 n ()). In such cases, t always yelds the best soluton (except m = 150 n ()). Besdes n () SS equals the earler best known soluton for m =10, 0, 30, and 70 (for m=40 no soluton was known). For m = 50, 60, and 80 the devatons from the earler best known soluton are very small. In () SS equals the earler best known soluton for m=10, 0, 30 and mproves that soluton for m = 40 and For hgher values of m, SS stll proves to be very compettve,.e., n sx out of twenty cases t yelds the best soluton (m = 130 and 140). In the other ffteen nstances, t holds second place, after the VNS-HK algorthm. Devatons n relaton to the earler best solutons are stll low. - Globally speakng, the VNS-HK and the SS algorthms are the best strateges n these tests. SS seems to be better, as t yelds the best soluton n most cases (6 compared to ). 9

10 - Another nterestng aspect s the robustness that SS shows n comparson to other strateges. In () no case s the best known soluton yelded by SS further than 1.350%. VNS-HK goes up to.19%. The other strateges show greater dstances: up to 5.5% for Memetc-HK, 5.086% for HybMem, and 4.91% for VNS-J. Fgure 3 shows the evoluton of the soluton found for the dfferent algorthms, accordng to the computatonal tme for N = 1060 and m = Computatonal Tme n Seconds Memetc HybMem VNS-HK VNS-J SS Fgure 3.- Evoluton of the soluton value found for the dfferent strateges accordng to the computatonal tme As shown n fgure 3, VNS-HK, and especally SS, have a very greedy character. After a few seconds, they yeld solutons that are better than those produced by other strateges after usng up the maxmum computatonal tme (600 seconds). 5.- Conclusons The contrbuton of our work s the development of a specalzed and sophstcated scatter search procedure for the soluton of the cluster desgn problem. Ths contrbuton s mportant, because our dversfcaton, mprovement and combnaton methods ntroduce novel features that can be adapted to other stuatons. Our method brngs together several search strateges and mechansms, such as GRASP constructons, local search, tabu search and Path Relnkng wthn the framework of scatter search. Usng nstances from the lterature, we were able to show the mert of our SS desgn. In partcular, our experments show that our method obtans the best overall solutons aganst others recent methods. Our method s specally effectve for small values of m. Wth hgher values of m t s only surpassed by the VNS algorthm proposed by Hansen and Mladenovc (001), wth HK-Means as the Local Search Procedure. However, the followng s worth notng: n ther work, VNS s used wth J-Means and J-Means+ as a local search procedure, whereas n our work the use of J-Means n VNS leads to worse solutons than VNS wth HK-Means and also to worse solutons than the methods we propose. 10

11 Fnally our scatter search approach shows two mportant features: t s a greedy method (t found good solutons n very short calculaton tme) and s able to evolve (mprove these solutons wth more calculaton tme). References [1] Adenso-Díaz B and. Laguna M. Automated Fne Tunng of Algorthms wth Taguch Fractonal Expermental Desgns and Local Search. Unversty of Colorado at Boulder; 001. [] Al-Sultan KH. A Tabu Search Approach to the Clusterng Problem. Pattern Recognton 1995;8: [3] Babu GP and Murty MN. A Near-Optmal Intal Seed Value Selecton n K-means Algorthm usng Genetc Algorthms. Pattern Recognton Letters 1993;14: [4] Beltrán M and Pacheco J. Nuevos métodos para el dseño de cluster no erárqucos. Una aplcacón a los muncpos de Castlla y León. Estadístca Española. Insttuto Naconal de Estadístca 001;43, 148:09-4. (avalable n [5] Brucker P. On the Complexty of Clusterng Problems. Lecture Notes n Economcs and Mathematcal Systems 1978;157: [6] Campos V, Glover F, Laguna M and Martí R. An Expermental Evaluaton of a Scatter Search for the Lnear Orderng Problem. Journal of Global Optmzaton 001; 1: [7] Dehr G. Evaluaton of a Branch and Bound Algorthm for Clusterng. SIAM J.Sc.Statst.Comput. 1985; 6: [8] du Merle, O., Hansen, P., Jaumard, B. and Mladenovc, N. An Interor Pont Algorthm for Mnmum Sum of Squares Clusterng. SIAM Journal on Scentfc Computng 000; 1(4): [9] Feo, T.A. and Resende, M.G.C. A Probablstc heurstc for a computatonally dffcult Set Coverng Problem. Operatons Research Letters 1989; 8: [10] Feo, T.A. and Resende, M.G.C. Greedy Randomzed Adaptve Search Procedures. Journal of Global Optmzaton 1995;:1-7. [11] Glover, F. A Template for Scatter Search and Path Relnkng. n Artfcal Evoluton, Lecture Notes n Computer Scence, 1363, J.-K. Hao, E. Lutton, E. Ronald, M. Schoenauer and D. Snyers (Eds.) Sprnger 1998, pp [1] Glover, F., Laguna, M. and Martí, R. Fundamentals of Scatter Search and Path Relnkng. Control and Cybernetcs 000; 39,3: [13] Hansen, P. and Mladenovc, N. J-Means: A new Local Search Heurstc for Mnmum Sum-of- Squares Clusterng. Pattern Recognton 001; 34(): [14] Howard, R. Classfyng a Populaton nto Homogeneous Groups. In Lawrence, J.R. (eds.), Operatonal Research n the Socal Scences. Tavstock Publ., London [15] Jancey, R.C. Multdmensonal Group Analyss. Australan J. Botany 1966;14: [16] Klen, R.W. and Dubes, R.C. Experments n Proecton and Clusterng by Smulated Annealng. Pattern Recognton 1989; : [17] Koontz, W.L.G., Narendra, P.M. and Fukunuga, K. A Branch and Bound Clusterng Algorthm. IEEE Transactons on Computers 1975; C-4: [18] Laguna, M. Scatter Search, n Handbook of Appled Optmzaton, P. M. Pardalos and M. G. C. Resende (Eds.), Oxford Unversty Press, New York 00, pp [19] Pacheco, J. and Valenca, O. Desgn of Hbrds for the Mnmun Sum-of-Squares Clusterng Problem. Computatonal Statstcs and Data Analyss 003; 43,: [0] Ptsouls, L.S. and Resende, M.G.C. Greedy Randomzed Adaptve Search Procedures n Handbook of Appled Optmzaton, P. M. Pardalos and M. G. C. Resende (Eds.), Oxford Unversty Press 00, pp [1] Renelt, G. TSPLIB: A Travellng Salesman Problem Lbrary. ORSA Journal on Computng 1991; 3: [] Späth, H. Cluster Analyss Algorthms for Data Reducton and Classfcaton of Obects. Ells Horwood, Chchester

Hybrid Heuristics for the Maximum Diversity Problem

Hybrid Heuristics for the Maximum Diversity Problem Hybrd Heurstcs for the Maxmum Dversty Problem MICAEL GALLEGO Departamento de Informátca, Estadístca y Telemátca, Unversdad Rey Juan Carlos, Span. Mcael.Gallego@urjc.es ABRAHAM DUARTE Departamento de Informátca,

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

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

Meta-heuristics for Multidimensional Knapsack Problems

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

More information

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

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

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

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

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

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

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

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

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

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms Course Introducton Course Topcs Exams, abs, Proects A quc loo at a few algorthms 1 Advanced Data Structures and Algorthms Descrpton: We are gong to dscuss algorthm complexty analyss, algorthm desgn technques

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

Wishing you all a Total Quality New Year!

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

More information

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 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

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

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

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

The Codesign Challenge

The Codesign Challenge ECE 4530 Codesgn Challenge Fall 2007 Hardware/Software Codesgn The Codesgn Challenge Objectves In the codesgn challenge, your task s to accelerate a gven software reference mplementaton as fast as possble.

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

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

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

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

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

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

expermental results on NRP nstances. Secton V dscusses some related wor and Secton VI concludes ths paper. II. PRELIMINARIES Ths secton gves the defnt

expermental results on NRP nstances. Secton V dscusses some related wor and Secton VI concludes ths paper. II. PRELIMINARIES Ths secton gves the defnt A Hybrd ACO Algorthm for the Next Release Problem He Jang School of Software Dalan Unversty of Technology Dalan 116621, Chna janghe@dlut.edu.cn Jngyuan Zhang School of Software Dalan Unversty of Technology

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

Maximum Variance Combined with Adaptive Genetic Algorithm for Infrared Image Segmentation

Maximum Variance Combined with Adaptive Genetic Algorithm for Infrared Image Segmentation Internatonal Conference on Logstcs Engneerng, Management and Computer Scence (LEMCS 5) Maxmum Varance Combned wth Adaptve Genetc Algorthm for Infrared Image Segmentaton Huxuan Fu College of Automaton Harbn

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

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

X- Chart Using ANOM Approach

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

More information

Edge Detection in Noisy Images Using the Support Vector Machines

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

More information

Feature Reduction and Selection

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

More information

Compiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz

Compiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz Compler Desgn Sprng 2014 Regster Allocaton Sample Exercses and Solutons Prof. Pedro C. Dnz USC / Informaton Scences Insttute 4676 Admralty Way, Sute 1001 Marna del Rey, Calforna 90292 pedro@s.edu Regster

More information

An Efficient Genetic Algorithm with Fuzzy c-means Clustering for Traveling Salesman Problem

An Efficient Genetic Algorithm with Fuzzy c-means Clustering for Traveling Salesman Problem An Effcent Genetc Algorthm wth Fuzzy c-means Clusterng for Travelng Salesman Problem Jong-Won Yoon and Sung-Bae Cho Dept. of Computer Scence Yonse Unversty Seoul, Korea jwyoon@sclab.yonse.ac.r, sbcho@cs.yonse.ac.r

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

Assignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009.

Assignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009. Farrukh Jabeen Algorthms 51 Assgnment #2 Due Date: June 15, 29. Assgnment # 2 Chapter 3 Dscrete Fourer Transforms Implement the FFT for the DFT. Descrbed n sectons 3.1 and 3.2. Delverables: 1. Concse descrpton

More information

Biostatistics 615/815

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

More information

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

A Unified Framework for Semantics and Feature Based Relevance Feedback in Image Retrieval Systems

A Unified Framework for Semantics and Feature Based Relevance Feedback in Image Retrieval Systems A Unfed Framework for Semantcs and Feature Based Relevance Feedback n Image Retreval Systems Ye Lu *, Chunhu Hu 2, Xngquan Zhu 3*, HongJang Zhang 2, Qang Yang * School of Computng Scence Smon Fraser Unversty

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

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

Overview. Basic Setup [9] Motivation and Tasks. Modularization 2008/2/20 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION

Overview. Basic Setup [9] Motivation and Tasks. Modularization 2008/2/20 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION Overvew 2 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION Introducton Mult- Smulator MASIM Theoretcal Work and Smulaton Results Concluson Jay Wagenpfel, Adran Trachte Motvaton and Tasks Basc Setup

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

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

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

Journal of Chemical and Pharmaceutical Research, 2014, 6(6): Research Article. A selective ensemble classification method on microarray data

Journal of Chemical and Pharmaceutical Research, 2014, 6(6): Research Article. A selective ensemble classification method on microarray data Avalable onlne www.ocpr.com Journal of Chemcal and Pharmaceutcal Research, 2014, 6(6):2860-2866 Research Artcle ISSN : 0975-7384 CODEN(USA) : JCPRC5 A selectve ensemble classfcaton method on mcroarray

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

PARETO BAYESIAN OPTIMIZATION ALGORITHM FOR THE MULTIOBJECTIVE 0/1 KNAPSACK PROBLEM

PARETO BAYESIAN OPTIMIZATION ALGORITHM FOR THE MULTIOBJECTIVE 0/1 KNAPSACK PROBLEM PARETO BAYESIAN OPTIMIZATION ALGORITHM FOR THE MULTIOBJECTIVE 0/ KNAPSACK PROBLEM Josef Schwarz Jří Očenáše Brno Unversty of Technology Faculty of Engneerng and Computer Scence Department of Computer Scence

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

Clustering Algorithm Combining CPSO with K-Means Chunqin Gu 1, a, Qian Tao 2, b

Clustering Algorithm Combining CPSO with K-Means Chunqin Gu 1, a, Qian Tao 2, b Internatonal Conference on Advances n Mechancal Engneerng and Industral Informatcs (AMEII 05) Clusterng Algorthm Combnng CPSO wth K-Means Chunqn Gu, a, Qan Tao, b Department of Informaton Scence, Zhongka

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

GA-Based Learning Algorithms to Identify Fuzzy Rules for Fuzzy Neural Networks

GA-Based Learning Algorithms to Identify Fuzzy Rules for Fuzzy Neural Networks Seventh Internatonal Conference on Intellgent Systems Desgn and Applcatons GA-Based Learnng Algorthms to Identfy Fuzzy Rules for Fuzzy Neural Networks K Almejall, K Dahal, Member IEEE, and A Hossan, Member

More information

A Facet Generation Procedure. for solving 0/1 integer programs

A Facet Generation Procedure. for solving 0/1 integer programs A Facet Generaton Procedure for solvng 0/ nteger programs by Gyana R. Parja IBM Corporaton, Poughkeepse, NY 260 Radu Gaddov Emery Worldwde Arlnes, Vandala, Oho 45377 and Wlbert E. Wlhelm Teas A&M Unversty,

More information

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Proceedngs of the Wnter Smulaton Conference M E Kuhl, N M Steger, F B Armstrong, and J A Jones, eds A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Mark W Brantley Chun-Hung

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

THE PATH PLANNING ALGORITHM AND SIMULATION FOR MOBILE ROBOT

THE PATH PLANNING ALGORITHM AND SIMULATION FOR MOBILE ROBOT Journal of Theoretcal and Appled Informaton Technology 30 th Aprl 013. Vol. 50 No.3 005-013 JATIT & LLS. All rghts reserved. ISSN: 199-8645 www.jatt.org E-ISSN: 1817-3195 THE PATH PLANNING ALGORITHM AND

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

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

Optimizing Document Scoring for Query Retrieval

Optimizing Document Scoring for Query Retrieval Optmzng Document Scorng for Query Retreval Brent Ellwen baellwe@cs.stanford.edu Abstract The goal of ths project was to automate the process of tunng a document query engne. Specfcally, I used machne learnng

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

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

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

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

Comparison of Heuristics for Scheduling Independent Tasks on Heterogeneous Distributed Environments

Comparison of Heuristics for Scheduling Independent Tasks on Heterogeneous Distributed Environments Comparson of Heurstcs for Schedulng Independent Tasks on Heterogeneous Dstrbuted Envronments Hesam Izakan¹, Ath Abraham², Senor Member, IEEE, Václav Snášel³ ¹ Islamc Azad Unversty, Ramsar Branch, Ramsar,

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

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

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

More information

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

BIN XIA et al: AN IMPROVED K-MEANS ALGORITHM BASED ON CLOUD PLATFORM FOR DATA MINING

BIN XIA et al: AN IMPROVED K-MEANS ALGORITHM BASED ON CLOUD PLATFORM FOR DATA MINING An Improved K-means Algorthm based on Cloud Platform for Data Mnng Bn Xa *, Yan Lu 2. School of nformaton and management scence, Henan Agrcultural Unversty, Zhengzhou, Henan 450002, P.R. Chna 2. College

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

On the Efficiency of Swap-Based Clustering

On the Efficiency of Swap-Based Clustering On the Effcency of Swap-Based Clusterng Pas Fränt and Oll Vrmaok Department of Computer Scence, Unversty of Joensuu, Fnland {frant, ovrma}@cs.oensuu.f Abstract. Random swap-based clusterng s very smple

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

Multi-objective Optimization Using Self-adaptive Differential Evolution Algorithm

Multi-objective Optimization Using Self-adaptive Differential Evolution Algorithm Mult-objectve Optmzaton Usng Self-adaptve Dfferental Evoluton Algorthm V. L. Huang, S. Z. Zhao, R. Mallpedd and P. N. Suganthan Abstract - In ths paper, we propose a Multobjectve Self-adaptve Dfferental

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

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

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

Complexity Analysis of Problem-Dimension Using PSO

Complexity Analysis of Problem-Dimension Using PSO Proceedngs of the 7th WSEAS Internatonal Conference on Evolutonary Computng, Cavtat, Croata, June -4, 6 (pp45-5) Complexty Analyss of Problem-Dmenson Usng PSO BUTHAINAH S. AL-KAZEMI AND SAMI J. HABIB,

More information

Quality Improvement Algorithm for Tetrahedral Mesh Based on Optimal Delaunay Triangulation

Quality Improvement Algorithm for Tetrahedral Mesh Based on Optimal Delaunay Triangulation Intellgent Informaton Management, 013, 5, 191-195 Publshed Onlne November 013 (http://www.scrp.org/journal/m) http://dx.do.org/10.36/m.013.5601 Qualty Improvement Algorthm for Tetrahedral Mesh Based on

More information

Load Balancing for Hex-Cell Interconnection Network

Load Balancing for Hex-Cell Interconnection Network Int. J. Communcatons, Network and System Scences,,, - Publshed Onlne Aprl n ScRes. http://www.scrp.org/journal/jcns http://dx.do.org/./jcns.. Load Balancng for Hex-Cell Interconnecton Network Saher Manaseer,

More information

Mathematics 256 a course in differential equations for engineering students

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

More information

Imperialist Competitive Algorithm with Variable Parameters to Determine the Global Minimum of Functions with Several Arguments

Imperialist Competitive Algorithm with Variable Parameters to Determine the Global Minimum of Functions with Several Arguments Fourth Internatonal Conference Modellng and Development of Intellgent Systems October 8 - November, 05 Lucan Blaga Unversty Sbu - Romana Imperalst Compettve Algorthm wth Varable Parameters to Determne

More information

A new segmentation algorithm for medical volume image based on K-means clustering

A new segmentation algorithm for medical volume image based on K-means clustering Avalable onlne www.jocpr.com Journal of Chemcal and harmaceutcal Research, 2013, 5(12):113-117 Research Artcle ISSN : 0975-7384 CODEN(USA) : JCRC5 A new segmentaton algorthm for medcal volume mage based

More information

Helsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr)

Helsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr) Helsnk Unversty Of Technology, Systems Analyss Laboratory Mat-2.08 Independent research projects n appled mathematcs (3 cr) "! #$&% Antt Laukkanen 506 R ajlaukka@cc.hut.f 2 Introducton...3 2 Multattrbute

More information

NGPM -- A NSGA-II Program in Matlab

NGPM -- A NSGA-II Program in Matlab Verson 1.4 LIN Song Aerospace Structural Dynamcs Research Laboratory College of Astronautcs, Northwestern Polytechncal Unversty, Chna Emal: lsssswc@163.com 2011-07-26 Contents Contents... 1. Introducton...

More information

Research of Dynamic Access to Cloud Database Based on Improved Pheromone Algorithm

Research of Dynamic Access to Cloud Database Based on Improved Pheromone Algorithm , pp.197-202 http://dx.do.org/10.14257/dta.2016.9.5.20 Research of Dynamc Access to Cloud Database Based on Improved Pheromone Algorthm Yongqang L 1 and Jn Pan 2 1 (Software Technology Vocatonal College,

More information

Concurrent Apriori Data Mining Algorithms

Concurrent Apriori Data Mining Algorithms Concurrent Apror Data Mnng Algorthms Vassl Halatchev Department of Electrcal Engneerng and Computer Scence York Unversty, Toronto October 8, 2015 Outlne Why t s mportant Introducton to Assocaton Rule Mnng

More information

SURFACE PROFILE EVALUATION BY FRACTAL DIMENSION AND STATISTIC TOOLS USING MATLAB

SURFACE PROFILE EVALUATION BY FRACTAL DIMENSION AND STATISTIC TOOLS USING MATLAB SURFACE PROFILE EVALUATION BY FRACTAL DIMENSION AND STATISTIC TOOLS USING MATLAB V. Hotař, A. Hotař Techncal Unversty of Lberec, Department of Glass Producng Machnes and Robotcs, Department of Materal

More information

K-means Optimization Clustering Algorithm Based on Hybrid PSO/GA Optimization and CS validity index

K-means Optimization Clustering Algorithm Based on Hybrid PSO/GA Optimization and CS validity index Orgnal Artcle Prnt ISSN: 3-6379 Onlne ISSN: 3-595X DOI: 0.7354/jss/07/33 K-means Optmzaton Clusterng Algorthm Based on Hybrd PSO/GA Optmzaton and CS valdty ndex K Jahanbn *, F Rahmanan, H Rezae 3, Y Farhang

More information

Optimization of integrated circuits by means of simulated annealing. Jernej Olenšek, Janez Puhan, Árpád Bűrmen, Sašo Tomažič, Tadej Tuma

Optimization of integrated circuits by means of simulated annealing. Jernej Olenšek, Janez Puhan, Árpád Bűrmen, Sašo Tomažič, Tadej Tuma Optmzaton of ntegrated crcuts by means of smulated annealng Jernej Olenšek, Janez Puhan, Árpád Bűrmen, Sašo Tomažč, Tadej Tuma Unversty of Ljubljana, Faculty of Electrcal Engneerng, Tržaška 25, Ljubljana,

More information

Simulation Based Analysis of FAST TCP using OMNET++

Simulation Based Analysis of FAST TCP using OMNET++ Smulaton Based Analyss of FAST TCP usng OMNET++ Umar ul Hassan 04030038@lums.edu.pk Md Term Report CS678 Topcs n Internet Research Sprng, 2006 Introducton Internet traffc s doublng roughly every 3 months

More information

Parallel matrix-vector multiplication

Parallel matrix-vector multiplication Appendx A Parallel matrx-vector multplcaton The reduced transton matrx of the three-dmensonal cage model for gel electrophoress, descrbed n secton 3.2, becomes excessvely large for polymer lengths more

More information

CHAPTER 2 DECOMPOSITION OF GRAPHS

CHAPTER 2 DECOMPOSITION OF GRAPHS CHAPTER DECOMPOSITION OF GRAPHS. INTRODUCTION A graph H s called a Supersubdvson of a graph G f H s obtaned from G by replacng every edge uv of G by a bpartte graph,m (m may vary for each edge by dentfyng

More information

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

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

More information

Stochastic optimization algorithm with probability vector in mathematical function minimization and travelling salesman problem

Stochastic optimization algorithm with probability vector in mathematical function minimization and travelling salesman problem Stochastc optmzaton algorthm wth probablty vector n mathematcal functon mnmzaton and travellng salesman problem JAN POHL, VÁCLAV JIRSÍK, PETR HONZÍK Department of Control and Instrumentaton Brno Unversty

More information

Ant Colony Optimization Applied to Minimum Weight Dominating Set Problem

Ant Colony Optimization Applied to Minimum Weight Dominating Set Problem Ant Colony Optmzaton Appled to Mnmum Weght Domnatng Set Problem Raa JOVANOVIC Mlan TUBA Dana SIMIAN Texas AM Unversty Faculty of Computer Scence Department of Computer Scence at Qatar Megatrend Unversty

More information

Problem Set 3 Solutions

Problem Set 3 Solutions Introducton to Algorthms October 4, 2002 Massachusetts Insttute of Technology 6046J/18410J Professors Erk Demane and Shaf Goldwasser Handout 14 Problem Set 3 Solutons (Exercses were not to be turned n,

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