Extending the Concept of Fuzzy Rule Interpolation with the Interpolation of Fuzziness
|
|
- Sharlene Charlene Briggs
- 5 years ago
- Views:
Transcription
1 WCCI 01 IEEE World Congress on Computatonal Intellgene June, 10-15, 01 - Brsbane, ustrala FUZZ IEEE Extendng the Conept of Fuzzy Rule Interpolaton wth the Interpolaton of Fuzzness Szlveszter Kovás Department of Informaton Tehnology Unversty of Mskol Mskol, Hungary szkovas@t.un-mskol.hu bstrat Fuzzy Rule Interpolaton (FRI) methods are not always sutable for desrbng hanges n the onluson fuzzness. For example, t s dffult to desrbe ases n whh the onluson for a rsp observaton must be fuzzy, or n whh an nrease n the fuzzness of an observaton yelds less fuzzness n the onluson. Ths problem s manly nherted from a lak of nformaton n the model, and orgnates from the defeny of the fuzzy rule representaton. In ths paper, a novel rule representaton onept alled double fuzzy dot s suggested, whh helps the elaboraton of FRI methods n order to be able to better handle the fuzzness of the onluson. Two smple examples of suh FRI methods are also brefly ntrodued n ths paper. Keywords- fuzzy rule nterpolaton, nterpolaton of the fuzzness, fuzzy funton, double fuzzy dot rule representaton. I. INTRODUCTION ommon problem wth exstng fuzzy reasonng methods based on fuzzy nterpolaton s that they have dffultes n expressng stuatons n whh the onluson to a rsp observaton s fuzzy, or n whh an nrease n the fuzzness of an observaton yelds less fuzzness n the onluson. The man reason for ths weakness n expressng the level of fuzzness of the onluson as a funton of the level of fuzzness of the observaton s manly nherted from the shortage of nformaton n suh models, whh orgnates from the defeny of the fuzzy dot rule representaton. s a soluton to ths problem, the paper proposes a novel rule representaton onept alled double fuzzy dot, whh helps the elaboraton of fuzzy rule nterpolaton (FRI) methods able to better handle the fuzzness of the onluson. Two smple examples of FRI methods based on the suggested double fuzzy dot rule representaton are also ntrodued brefly n ths paper. II. FUZZY FUNCTIONS, INTERPOLTION OF FUZZY DT There are two ommon ways to defne fuzzy funtons. The frst s the lassal desrptve way, where the fuzzy funton s vewed as a speal fuzzy relaton. The seond s the delaratve way, whh onsders the fuzzy funton as a mappng between two fuzzy spaes. In the desrptve vew the fuzzy rules represent relatons between rule anteedents and onsequent fuzzy sets. In ths ase, the entre rule-base s also onsdered to be a fuzzy relaton of the anteedent and onsequent unverses, bult upon the dsjunton of the fuzzy rule relatons. The lassal ompostonal rule of nferene [3] also follows the desrptve vew, as the onluson fuzzy set s ganed as a omposton of the fuzzy observaton and the fuzzy rule-base relaton. In the rest of the paper we wll onsder the fuzzy funton form the delaratve vew, as a mappng between two fuzzy spaes. pplyng the notaton ntrodued n [1], [] by Perfleva et al. the fuzzy funton as a mappng an be defned n the followng way: Let L X and L Y are fuzzy spaes on X and Y, respetvely. X Y orrespondene f : L L s a fuzzy funton f for X every, B L, = B mples f ( ) = f( B). = B ff x X x = B x., ( ) ( ) X Y fuzzy funton f : L L s extensonal f for every X, B L, ( B) ( f( ) f( B) ), where s a smlarty L X, L. Y relaton defned on In ths ase, the fuzzy nterpolaton of fuzzy data an be onsdered as an extenson of a fuzzy funton: Let fuzzy data be gven by a set of pars {(, B ) = 1,..., n}, where L X, B L Y, = 1,..., n, and 1,..., n are dfferent wth respet to =. Moreover, let f : B be a fuzzy funton on the doman = { 1,..., n }, and P { : D L Y }, where D L X, s a hosen subset of fuzzy funtons. The problem s to fnd P (P s a set of nterpolatng funtons) suh that the nterpolaton ondton ( ) = B, = 1,..., n s fulflled. In ths ase, s alled an nterpolatng fuzzy funton for fuzzy data {(, B ) = 1,..., n}. In other words, the nterpolatng fuzzy funton s an extenson of a fuzzy funton f on the doman D. III. FUZZY RULE INTERPOLTION METHODS Turnng the notaton of fuzzy funton and fuzzy nterpolaton to the form of fuzzy rules and fuzzy rule nterpolaton, n the delaratve vew of the fuzzy funton, the U.S. Government work not proteted by U.S. opyrght 1106
2 n fuzzy rules of the fuzzy rule-base represent the rules as fuzzy data {(, B ) = 1,..., n}, alled fuzzy dots. In ths ase, the way of fuzzy reasonng based on fuzzy nterpolaton an be understood as a searh for a sutable nterpolatng fuzzy funton ( ) = B, = 1,..., n. Based on the nterpolatng fuzzy funton, the onluson y L Y for observaton x L X s smply ganed as y = (x). There are numerous methods of fuzzy nterpolaton whh have appeared n the lterature n the last 0 years. One of the frst FRI tehnques was publshed by Kózy and Hrota [7]. It s usually referred as the KH method. It s applable to onvex and normal fuzzy (CNF) sets. It determnes the onluson by ts α-uts n suh a way that the rato of dstanes between the onluson and the onsequents should be dental wth the ones between the observaton and the anteedents for all mportant α-uts. It s shown n e.g. [8], [9] that the onluson of the KH method s not always dretly nterpretable as a fuzzy set. Ths drawbak motvated many alternatve solutons. modfaton was proposed by Vass, Kalmár and Kózy [10] (VKK method), where the onluson s omputed based on the dstane of the entre ponts and the wdths of the α-uts, nstead of lower and upper dstanes. The VKK method dereases the applablty lmt of the KH method, but does not elmnate t ompletely. The tehnque annot be appled f any of the anteedent sets s sngleton (the wdth of the anteedent s support must be nonzero). In spte of the dsadvantages, KH s popular beause ts smplty nfers advantageous omplexty propertes. It was generalzed n several ways. mong them the stablzed KH nterpolator s emerged, as t s proved to hold the unversal approxmaton property [11], [1]. Ths method takes nto aount all flakng rules of an observaton n the alulaton of the onluson n extent to the nverse of the dstane of anteedents and observaton. The unversal approxmaton property holds f the dstane funton s rased to the power of the nput s dmenson. nother modfaton of KH s the modfed alphaut based nterpolaton (MCI) method [13], whh allevates ompletely the abnormalty problem. MCI s man dea s the followng: t transforms fuzzy sets of the nput and output unverses to suh a spae where abnormalty s exluded, then omputes the onluson there, whh s fnally transformed bak to the orgnal spae. MCI uses vetor representaton of fuzzy sets and orgnally applable to CNF sets [14]. MCI s one of the pratally applable FRI methods [15], sne t preserves advantageous omputatonal and approxmate nature of KH, whle t exludes ts abnormalty. n applaton orented aspet of the FRI, the low omputatonal and resoure demand s also emergng n the onept of FIVE (Fuzzy Interpolaton based on Vague Envronment) [16]. nother fuzzy nterpolaton tehnque was proposed by Kózy et al. [17]. It s alled onservaton of relatve fuzzness (CRF) method, whh noton means that the left (rght) fuzzness of the approxmated onluson n proporton to the flankng fuzzness of the neghborng onsequent should be the same as the (left) rght fuzzness of the observaton n proporton to the flankng fuzzness of the neghborng anteedent. The tehnque s applable to CNF sets. n mproved fuzzy nterpolaton tehnque for multdmensonal nput spaes (IMUL) was proposed n [18], and desrbed n detals n [15]. IMUL apples a ombnaton of CRF and MCI methods, and mxes advantages of both. The ore of the onluson s determned by MCI method, whle ts flanks by CRF. The man advantages of ths method are ts applablty for multdmensonal problems and ts relatve smplty. In [] a novel FRI dea able to handle hghly overlappng rule anteedents by nonlnear mappng s appears. There are a lot of works done n the dreton fuzzy nterpolaton based on fuzzy relaton vew of fuzzy rules by Dubos, Prade et al. [3]. The suggested nterpolaton by onvex ompleton method [4] s based on the extenson of nterval nterpolaton to fuzzy nterpolaton. Coneptually dfferent approahes were proposed by Bouhon-Meuner et al. [5], [6]. The suggested analogybased nterpolaton frst nterpolates the referene pont poston of the onluson, then onstruts the shape of onluson fuzzy sets based on the smlarty (dstngushablty) relaton of the rule anteedents and observaton wth respet to the smlarty of the orrespondng rule onsequents and the onluson. smlar two step method dea s appearng n the work of Barany et al. [4]. They extend the onept to be able to handle arbtrary shaped fuzzy sets. The suggested two step General Methodology (GM) frst generate an nterpolated ntermedate rule n the referene pont poston of the observaton, then n the seond step a sngle rule reasonng method (revson funton) s appled to determne the fnal fuzzy onluson based on the smlarty of the fuzzy observaton and an nterpolated observaton. Pratal applaton of the GM appears n the work of Johanyák et al. [7]. The suggested Polar Cut nterpolaton solves the task of the sngle rule reasonng step by smlarty of arbtrary shaped fuzzy sets defned based on polar uts. n extenson of the GM s also appearng n the work of Shen et al. [8]. The suggested sale and move transformaton an extend the orgnal method to extrapolaton. The ode of many of the above mentoned FRI methods are freely avalable as a part of the MTLB FRI Toolbox [19], together wth a sparse fuzzy model dentfaton toolbox [0] s avalable at [1]. IV. INTERPOLTION OF THE FUZZINESS It an be easly stated, that many of the above mentoned methods have dffultes when the hange of the fuzzness of the onluson n the funton of the fuzzness of the observaton have to be delberately defned. E.g. for a rsp observaton the onluson have to be fuzzy, or f nreasng the fuzzness of the observaton have to lead to dereasng the fuzzness of the onluson. The problem s straghtforward n the ase f the FRI method s defned for a sngle dmensonal anteedent unverse and two flankng anteedent rules to the observaton, lke the orgnal KH method, the orgnal GM, or the analogy-based nterpolaton. The freedom of ndependently handlng the fuzzness s mssng from the defnton of the methods. On the other hand t s also a lak of nformaton. There s no way for expressng the hange of the fuzzness n the poston of the observaton havng two neghbourng fuzzy rules only. 1107
3 In ase of FRI methods able to handle multple rules from the rule base (usually defned as an extenson of the orgnal method) by takng more rules nto the onsderaton the hange of the fuzzness an be expressed e.g. by overlappng rules havng dfferent fuzzness on the anteedent and onsequent sde. Therefore the man reason of the dffultes n expressng the level of fuzzness of the onluson n the funton of the level of fuzzness of the observaton s manly nherted from the shortage of nformaton orgnates from the defeny of the fuzzy rule representaton. To dsmss the problem of delberately defnng the hange of the fuzzness of the onluson n the funton of the fuzzness of the observaton n most of the FRI methods there are two ommon preoneptons hold about the nterpolaton of the fuzzness: In ase of rsp observaton, the onluson must be rsp. If the level of fuzzness n the observaton s nreasng, the level of the fuzzness n the onluson must be also nrease. The nterpolatng fuzzy funton usually hosen to be sutable for smoothly desrbng a sparse fuzzy rule-base, a rule-base, whh not fully overs the anteedent unverse, and usually ontans only a few fuzzy rules,.e. fuzzy dots, {(, B ) = 1,..., n}, where n s usually relatvely small. The nterpolatng fuzzy funton s usually takng less are about how the fuzzness s hangng n the lose neghborhood of a gven rule. It s also a lak of nformaton,.e. n a sparse rulebase t s also unommon to have overlappng rules ( j =0, j 1,..., n). Hene n the neghborhood of a gven rule the only exstng nformaton about the hange of the fuzzness s the fuzzness relaton of the anteedent and onsequent fuzzy sets of the rule, whh s a onstant n a rule represented as a fuzzy dot. To somehow handle the problem of havng observatons wth dfferent fuzzness than the fuzzness of the rule anteedent, most of the fuzzy nterpolaton methods follow some gudelnes alled axomat approah of the fuzzy nterpolaton. omprehensve overvew of the axomat approah of the fuzzy nterpolaton an be found n []. The most ommon gudelne for handlng the fuzzness relaton of the observaton and onluson n the axomat approah of the fuzzy nterpolaton funton s the Monotonty ondton. In [5] t s the ondton I and defned for the sngle dmensonal anteedent ase, n [6] t s alled Property 6. and extended for multdmensonal anteedents n the followng way: Monotonty [6]: Let *, + are two observatons suh that * s more spef n all dmenson than +, that s for + j =1,...,n, * + j j, then ( * j) ( j ),.e. the same relaton should be held for the two onlusons. In other words the monotonty ondtons mean that we have no nformaton about the rate or lmt the hange of the fuzzness of the onluson wth respet to the hange of the fuzzness of the observaton, only that they have to hange n the same dreton (nreasng or dereasng). The monotonty property s orgnally ntrodued n [5], where t had no partular reason save that the author proves that the method proposed n that paper satsfy ths property. To have some gudelne about the extreme ases of the fuzzness, some fuzzy rule nterpolaton methods also holds the property, that n ase of a rsp observaton, t has to have rsp onluson. But ths ondton s usually not appearng among the axomat approah of the fuzzy nterpolaton, exept n [6] as Property 9. : The fuzzness of the onluson: mappng of a FRI s allowed to generate sngleton onluson n any of the next two ases: a) Whenever the observaton * s sngleton, then the onluson ( * ) should be sngleton as well. b) Whenever all B, where denotes ndes of rules that ontrbutes to the alulaton of onluson ( * ) and the observaton * are sngleton, then ( * ) should also be sngleton. Both the axomat approahes of the fuzzy nterpolaton,.e. the monotonty property and fuzzness of the onluson property are gvng some gudelne for handlng the fuzzness of the onluson n ase of lak of nformaton about the hange of the fuzzness. But they are reatng unneessary restrtons for the reasonng f the nterpolaton of the fuzzness an be delberately handled. V. EXPRESSING THE FUZZINESS CHNGE To be able to express the hange of the fuzzness n a fuzzy rule, the fuzzy dot representaton of the fuzzy rules has to be extended by the ablty of handlng the fuzzness hange n the loaton of the fuzzy dot. In ths paper for ths purpose a smple way s suggested. The onept of the sngle fuzzy dot rule representaton has to be extended to two overlappng fuzzy rules, to the double fuzzy dot rule representaton. Ths ase a sngle fuzzy rule s bult upon two fuzzy rules, where n both the overlappng rules the anteedent and onsequent fuzzy sets have the same referene pont, but they an have dfferent fuzzness. For smplfyng the explanaton and vsualzaton of the man onept, n the further part of the paper we restrt the shape of the fuzzy sets to normal soseles trangular shaped membershp funtons (nludng the sngleton). Ths restrton s a loss of generalty, but serves for the better demonstraton purposes, as the goal of ths paper s to ntrodue the man double fuzzy dot onept wthout solvng the omplex and general problem of arbtrary shaped fuzzy sets. In ase of normal soseles trangular shaped fuzzy sets the referene pont s a sngle pont, the ore: µ ore( ) = 1 Ths ase the th rule of the double fuzzy dot rule representaton has the followng form: p, q p, q B, 1108
4 p p whh s bult upon two overlappng fuzzy rules: B q q p p q q and B, where ( ) = B and ( ) = B, where n both the overlappng rules the anteedent and the onsequent p q fuzzy sets have the same referene pont: ore ( ) = ore( ), p q ore ( B ) = ore( B ), = 1,..., n, but they an have dfferent fuzzness (see e.g. on Fg. 1. and Fg..). For demonstratve purposes n ths paper to be the frst example the extenson of the fuzzy sngle rule reasonng (the revson funton n GM [4]) was hosen. Ths ase the sngle rule s a double fuzzy dot rule p, q p, q p p q q B, where ( ) =B, ( ) =B, and the =. onluson s ganed by the nterpolatng funton: ( *) B* The nterpolatng funton ould be defned based on the lose X, =, lose Y ( B, B) = 1: loseness relaton ( ) 1 The loser the observaton * to one of the element of the double fuzzy dot anteedent j, the loser s the onluson B * to the orrespondng onsequent B j : X j j ( ) lose ( B B ), Y * lose *,. Wthout more detaled alulatons, the onlusons of some relevant double fuzzy dot rule onfguratons are ntrodued on Fg Fg.. Fgure 1. Double fuzzy dot rule representaton when the dreton of the fuzzness hange s remanng the same. Fgure 3. Sngle Rule Reasonng Double fuzzy dot rule representaton when p, B p are fuzzy and q, B q are sngletons. Fgure. Double fuzzy dot rule representaton when the dreton of the fuzzness hange s reversng. Please note, that n some of the exstng FRI methods the double fuzzy dot rule representaton n the rule base means ontradtve rules. (Rules havng same (.e. hghly overlappng) anteedents and dfferent onsequents. VI. EXMPLE: SINGLE RULE RESONING Most of the exstng FRI methods followng the delaratve way of the fuzzy funton nterpolaton ould be extended wth the apablty of nterpolaton of the fuzzness by ntrodung the double fuzzy dot rule representaton. Fgure 4. Sngle Rule Reasonng Double fuzzy dot rule representaton when p, B p, B q are fuzzy and q s sngleton. 1109
5 Fgure 5. Sngle Rule Reasonng Double fuzzy dot rule representaton when p, B p and q, B q are fuzzy. Fgure 8. Sngle Rule Reasonng Double fuzzy dot rule representaton when the dreton of the fuzzness hange s reversng and q, B p are sngletons. Fgure 6. Sngle Rule Reasonng Double fuzzy dot rule representaton when p, B p, q are fuzzy and B q s sngleton. VII. EXMPLE: DOUBLE-LINER FUZZY INTERPOLTION For havng a more omplex example, n the followngs, the smplfed verson of the sngle dmensonal double-lnear fuzzy nterpolaton method s extended by the ablty of nterpolaton of the fuzzness. The smplfaton of the example s a restrton of the orgnal method from handlng trapezodal to trangular shaped fuzzy sets only. Ths method has a very smple way for vsualzaton of the fuzzy nterpolaton and hene sutable for the demonstraton purposes of ths paper. The double-lnear fuzzy nterpolaton method was orgnally ntrodued by Detynek et al. n [9]. The method was defned for sngle dmensonal anteedent unverse only, where all fuzzy sets n the rule-base are normal and onvex trapezodal shaped fuzzy sets defned on the unverse of R real numbers. The rule-base onssts of two rules 1 B and 1 B, where, are fuzzy sets of 1 L X, X R, and B,B 1 are fuzzy sets of L Y, Y R. The rule-base s sparse n the sense that the support sets of 1 and are dsjont and 1, where s a preedene relaton. The preedene relaton s also exsts among B,B. 1 Modfyng the equatons of [9] aordng to trangular shaped fuzzy sets, B, havng a trangular observaton * we get the followng lnear equaton for the B * onluson (see Fg. 9 for notaton): ( ) = α α (1) * ( ) B = α α () B* B1 + 1 Fgure 7. Sngle Rule Reasonng Double fuzzy dot rule representaton when the dreton of the fuzzness hange s reversng. 1110
6 * α = (3) 1 = (4) * * 1 B* B1 + B 1 1 where B * s the ore of the onluson. unertantes of the observaton *L, to the newly nterpolated rule anteedent el must be the same as the rate of the unertantes of the newly nterpolated rule onsequent B el and the onluson B *L we are lookng for (see e.g. on Fg. 10): B L L = * * el B Please note, that n onept the double-lnear fuzzy nterpolaton method s also very smlar to the two step nterpolaton methodology ntrodued n [4] (see also on Fg. 11). el (7) Fgure 9. Notaton for the 1, and * trangular shaped rule anteeden fuzzy sets and observaton (the notaton s the same for the onsequent sets and onluson B 1, B and B *) For the left el, B el (and rght el, B er ) unertantes of the nterpolated rule e B e we get smlar lnear equatons (see Fg. 10 for notaton): ( ) L = α α (5) el 1L + 1 ( ) B L = α α (6) BeL B1L + 1 Fgure 11. Double-lnear nterpolaton of two fuzzy rules 1 B 1, B, the nterpolated rule e B e and the onluson * B * To extend the double-lnear fuzzy nterpolaton method to be able to handle the nterpolaton of the fuzzness by the double fuzzy dot rule representaton s very smple. The man onept s remanng the same. There are no hanges n the use of Eq. (1) (6), exept the nterpolated fuzzy rule must be alulated separately for the p B p and the q B q rulebases. s a result we get the double fuzzy dot representaton of the nterpolated fuzzy rule pq e B pq e (see e.g. on Fg. 13). For alulatng the left B *L (and rght B *R ) unertantes of the onluson we an follow the onept of the orgnal method (Eq. (7)), save ths ase we have to keep the relatve unertantes of the observaton to the two newly reated rules,.e. the rate of the unertantes of the observaton *L, to the two newly nterpolated rule anteedents p el and q el must be the same as the rate of the unertantes of the two newly nterpolated rule onsequents B p el and B q el and the onluson B *L we are lookng for (see e.g. on Fg. 1, Fg.13 and Fg.14): Fgure 10. Calulatng the left el, B el unertantes of the nterpolated rule e B e and the onluson * B * The left B *L (and rght B *R ) unertantes of the onluson an be alulated by keepng the relatve unertantes of the observaton to the newly reated rule,.e. the rate of the B p el * L q * L * L = B p q el+ p el el el q el q el B p el (8) 1111
7 Fgure 14. The nterpolated rule p,q e B p,q e and the onluson * B * n ase of double fuzzy dot rule representaton. Fgure 1. Calulatng the left p,q el, B p,q el unertantes of the nterpolated rule p,q e B p,q e and the onluson * B * n ase of double fuzzy dot rule representaton. VIII. CONCLUSION The double fuzzy dot rule representaton proposed n ths paper ntrodues a novel approah for extendng FRI methods n ways that are able to nterpolate arbtrary hanges of fuzzness n the onluson. The nterpolated fuzzness adds a new dmenson for amelorated fuzzy funton defntons based on FRI methods. CKNOWLEDGMENT Ths researh was supported by the Hungaran Natonal Sentf Researh Fund grant no: OTK K Ths researh was arred out as part of the TMOP-4..1.B- 10//KONV projet wth support by the European Unon, o-fnaned by the European Soal Fund. Fgure 13. Double-lnear nterpolaton of two fuzzy rules p,q 1 B p,q 1, p,q B p,q and the nterpolated rule p,q e B p,q e n ase of double fuzzy dot rule representaton. REFERENCES [1] I. Perfleva, Fuzzy funton as an approxmate soluton to a system of fuzzy relaton equatons, Fuzzy Sets and Systems 147, 004, pp [] I. Perfleva, et al., Interpolaton of fuzzy data: nalytal approah and overvew, Fuzzy Sets and Systems, 010, do: /j.fss [3] E. H. Mamdan and S. sslan, n experment n lngust synthess wth a fuzzy log ontroller, Int. J. of Man Mahne Studes, (7), 1975, pp [4] P. Barany, L. T. Kózy, and T. D. Gedeon, Generalzed Conept for Fuzzy Rule Interpolaton, IEEE Trans. on Fuzzy Systems, vol. 1, No. 6, 004, pp [5] S. Jene, Interpolatng and extrapolatng fuzzy quanttes revsted an axomat approah, Soft Comput., vol. 5., 001, pp [6] D. Tkk, Z. C. Johanyák, Sz. Kovás, K. W. Wong, Fuzzy Rule Interpolaton and Extrapolaton Tehnques: Crtera and Evaluaton Gudelnes, Journal of dvaned Computatonal Intellgene and Intellgent Informats, ISSN , Vol.15, No.3, 011, pp [7] L. T. Kózy and K. Hrota, Rule nterpolaton by α-level sets n fuzzy approxmate reasonng, In J. BUSEFL, utomne, UR-CNRS. Vol. 46. Toulouse, Frane, 1991, pp [8] L. T. Kózy and Sz. Kovás, On the preservaton of the onvexty and peewse lnearty n lnear fuzzy rule nterpolaton, Tokyo Inst. Tehnol., Yokohama, Japan, Teh. Rep. TR 93-94/40, LIFE Char Fuzzy Theory, [9] L. T. Kózy, and Sz. Kovás, Shape of the Fuzzy Conluson Generated by Lnear Interpolaton n Trapezodal Fuzzy Rule Bases, n Proeedngs of the nd European Congress on Intellgent Tehnques and Soft Computng, ahen, 1994, pp
8 [10] G. Vass, L. Kalmár, and L. T. Kózy, Extenson of the fuzzy rule nterpolaton method, n Pro. Int. Conf. Fuzzy Sets Theory pplatons (FST 9), Lptovsky M., Czehoslovaka, 199, pp [11] D. Tkk, I. Joó, L. T. Kózy, P. Várlak, B. Moser, and T. D. Gedeon, Stablty of nterpolatve fuzzy KH-ontrollers, Fuzzy Sets and Systems, 15(1), January, 00, pp [1] D. Tkk: Notes on the approxmaton rate of fuzzy KH nterpolator. Fuzzy Sets and Systems, 138(), Sept., 003, pp [13] D. Tkk and P. Barany, Comprehensve analyss of a new fuzzy rule nterpolaton method, In IEEE Trans. Fuzzy Syst., vol. 8, No. 3, June, 000, pp [14] Y. Yam, and L. T. Kózy, Representng membershp funtons as ponts n hgh dmensonal spaes for fuzzy nterpolaton and extrapolaton, Dept. Meh. utomat. Eng., Chnese Unv. Hong Kong, Teh. Rep.CUHK-ME-97-03, [15] K. W. Wong, D. Tkk, T. D. Gedeon, and L. T. Kózy, Fuzzy Rule Interpolaton for Multdmensonal Input Spaes Wth pplatons, IEEE Transatons on Fuzzy Systems, ISSN , Vol. 13, No. 6, Deember, 005, pp [16] Sz. Kovás, Extendng the Fuzzy Rule Interpolaton "FIVE" by Fuzzy Observaton, dvanes n Soft Computng, Computatonal Intellgene, Theory and pplatons, Bernd Reush (Ed.), Sprnger Germany, ISBN , 006, pp [17] L.T. Kózy, K. Hrota, and T. D. Gedeon, Fuzzy rule nterpolaton by the onservaton of relatve fuzzness, Tehnal Report TR 97/. Hrota Lab, Dept. of Comp. Int. and Sys. S., Tokyo Inst. of Tehn., Yokohama, [18] K. W. Wong, T. D. Gedeon, and D. Tkk: n mproved multdmensonal α-ut based fuzzy nterpolaton tehnque, In Pro. Int. Conf rtfal Intellgene n Sene and Tehnology (IST 000), Hobart, ustrala, 000, pp [19] Zs. Cs. Johanyák, D. Tkk, Sz. Kovás, K. W. Wong Fuzzy Rule Interpolaton Matlab Toolbox FRI Toolbox, Pro. of the IEEE World Congress on Computatonal Intellgene (WCCI'06), 15th Int. Conf. on Fuzzy Systems (FUZZ-IEEE'06), July 16-1, Vanouver, BC, Canada, Omnpress. ISBN , 006, pp [0] Zs. Cs. Johanyák, Sparse fuzzy model dentfaton Matlab toolbox - RuleMaker toolbox, Proeedngs of IEEE 6 th Internatonal Conferene on Computatonal Cybernets ICCC, Stara Lesná, Slovaka, 008, pp [1] The FRI Toolbox s avalable at: [] K. Uehara, S. Sato, and K. Hrota, Inferene for Nonlnear Mappng wth Sparse Fuzzy Rules Based on Mult-Level Interpolaton, Journal of dvaned Computatonal Intellgene and Intellgent Informats, ISSN , Vol.15, No.3, 011, pp [3] D. Dubos, R. Martn-Clouare, H. Prade, Pratal omputng n fuzzy log, Fuzzy Computng, M. M. Gupta, T. Yamakawa (Eds.), North- Holland, [4] L. Ughetto, D. Dubos, H. Prade, Fuzzy nterpolaton by onvex ompleton of sparse rule bases, Proeedngs of FUZZ-IEEE 000, Internatonal Conferene, San ntono, 000, pp [5] B. Bouhon-Meuner, J. Delehamp, C. Marsala, N. Melloul, M. Rfq, L. Zerrouk, nalogy and fuzzy nterpolaton n ase of sparse rules, Proeedngs of the EUROFUSE-SIC Jont Conferene, 1999, pp [6] B. Bouhon-Meuner, C. Marsala, M. Rfq, Interpolatve reasonng based on gradualty, Proeedngs of FUZZ-IEEE 000, Internatonal Conferene, San ntono, 000, pp [7] Z. C. Johanyák, Sz. Kovás, Fuzzy rule nterpolaton based on polar uts, Computatonal Intellgene, Theory and pplatons, B. Reush (Ed.), Sprnger, 006 pp [8] Z. Huang and Q. Shen, Fuzzy nterpolatve reasonng va sale and move transformatons, IEEE Trans. Fuzzy Syst., 14(), 006, pp [9] M. Detynek, C. Marsala, and M. Rfq, Double-Lnear Fuzzy Interpolaton Method, Proeedngs of the 011 IEEE Internatonal Conferene On Fuzzy Systems, (FUZZ-IEEE 011), Tape, Tawan June 7-30, 011, pp
Distance based similarity measures of fuzzy sets
Johanyák, Z. C., Kovács S.: Dstance based smlarty measures of fuzzy sets, SAMI 2005, 3 rd Slovakan-Hungaran Jont Symposum on Appled Machne Intellgence, Herl'any, Slovaka, January 2-22 2005, ISBN 963 75
More informationConnectivity in Fuzzy Soft graph and its Complement
IOSR Journal of Mathemats (IOSR-JM) e-issn: 2278-5728, p-issn: 2319-765X. Volume 1 Issue 5 Ver. IV (Sep. - Ot.2016), PP 95-99 www.osrjournals.org Connetvty n Fuzzy Soft graph and ts Complement Shashkala
More informationType-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 informationSum 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 informationHermite Splines in Lie Groups as Products of Geodesics
Hermte Splnes n Le Groups as Products of Geodescs Ethan Eade Updated May 28, 2017 1 Introducton 1.1 Goal Ths document defnes a curve n the Le group G parametrzed by tme and by structural parameters n the
More informationA 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 informationThe Simulation of Electromagnetic Suspension System Based on the Finite Element Analysis
308 JOURNAL OF COMPUTERS, VOL. 8, NO., FEBRUARY 03 The Smulaton of Suspenson System Based on the Fnte Element Analyss Zhengfeng Mng Shool of Eletron & Mahanal Engneerng, Xdan Unversty, X an, Chna Emal:
More informationSession 4.2. Switching planning. Switching/Routing planning
ITU Semnar Warsaw Poland 6-0 Otober 2003 Sesson 4.2 Swthng/Routng plannng Network Plannng Strategy for evolvng Network Arhtetures Sesson 4.2- Swthng plannng Loaton problem : Optmal plaement of exhanges
More informationBottom-Up Fuzzy Partitioning in Fuzzy Decision Trees
Bottom-Up Fuzzy arttonng n Fuzzy eson Trees Maej Fajfer ept. of Mathemats and Computer Sene Unversty of Mssour St. Lous St. Lous, Mssour 63121 maejf@me.pl Cezary Z. Janow ept. of Mathemats and Computer
More informationProgressive scan conversion based on edge-dependent interpolation using fuzzy logic
Progressve san onverson based on edge-dependent nterpolaton usng fuzzy log P. Brox brox@mse.nm.es I. Baturone lum@mse.nm.es Insttuto de Mroeletróna de Sevlla, Centro Naonal de Mroeletróna Avda. Rena Meredes
More informationTsinghua 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 informationA 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 informationUsing fuzzy rule interpolation based automata for controlling navigation and collision avoidance behaviour of a robot
Using fuzzy rule interpolation based automata for controlling navigation and collision avoidance behaviour of a robot Dávid Vincze, Szilveszter Kovács Department of Information Technology, University of
More informationSparse Fuzzy Model Identification Matlab Toolox RuleMaker Toolbox
Johanyák, Z. C.: Sparse Fuzzy Model Identfcaton Matlab Toolbox - RuleMaker Toolbox, IEEE 6th Internatonal Conference on Computatonal Cybernetcs, November 27-29, 2008, Stara Lesná, Slovaka, pp. 69-74. Sparse
More informationResearch on Neural Network Model Based on Subtraction Clustering and Its Applications
Avalable onlne at www.senedret.om Physs Proeda 5 (01 ) 164 1647 01 Internatonal Conferene on Sold State Deves and Materals Sene Researh on Neural Networ Model Based on Subtraton Clusterng and Its Applatons
More informationPerformance Evaluation of Information Retrieval Systems
Why System Evaluaton? Performance Evaluaton of Informaton Retreval Systems Many sldes n ths secton are adapted from Prof. Joydeep Ghosh (UT ECE) who n turn adapted them from Prof. Dk Lee (Unv. of Scence
More informationTAR based shape features in unconstrained handwritten digit recognition
TAR based shape features n unonstraned handwrtten dgt reognton P. AHAMED AND YOUSEF AL-OHALI Department of Computer Sene Kng Saud Unversty P.O.B. 578, Ryadh 543 SAUDI ARABIA shamapervez@gmal.om, yousef@s.edu.sa
More informationInterval uncertain optimization of structures using Chebyshev meta-models
0 th World Congress on Strutural and Multdsplnary Optmzaton May 9-24, 203, Orlando, Florda, USA Interval unertan optmzaton of strutures usng Chebyshev meta-models Jngla Wu, Zhen Luo, Nong Zhang (Tmes New
More informationS1 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 informationA new paradigm of fuzzy control point in space curve
MATEMATIKA, 2016, Volume 32, Number 2, 153 159 c Penerbt UTM Press All rghts reserved A new paradgm of fuzzy control pont n space curve 1 Abd Fatah Wahab, 2 Mohd Sallehuddn Husan and 3 Mohammad Izat Emr
More informationImprovement 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 informationFuzzy Modeling for Multi-Label Text Classification Supported by Classification Algorithms
Journal of Computer Senes Orgnal Researh Paper Fuzzy Modelng for Mult-Label Text Classfaton Supported by Classfaton Algorthms 1 Beatrz Wlges, 2 Gustavo Mateus, 2 Slva Nassar, 2 Renato Cslagh and 3 Rogéro
More informationSolving 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 informationProblem 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 informationAnalysis 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 informationSteganalysis of DCT-Embedding Based Adaptive Steganography and YASS
Steganalyss of DCT-Embeddng Based Adaptve Steganography and YASS Qngzhong Lu Department of Computer Sene Sam Houston State Unversty Huntsvlle, TX 77341, U.S.A. lu@shsu.edu ABSTRACT Reently well-desgned
More informationMatrix-Matrix Multiplication Using Systolic Array Architecture in Bluespec
Matrx-Matrx Multplaton Usng Systol Array Arhteture n Bluespe Team SegFault Chatanya Peddawad (EEB096), Aman Goel (EEB087), heera B (EEB090) Ot. 25, 205 Theoretal Bakground. Matrx-Matrx Multplaton on Hardware
More informationA MPAA-Based Iterative Clustering Algorithm Augmented by Nearest Neighbors Search for Time-Series Data Streams
A MPAA-Based Iteratve Clusterng Algorthm Augmented by Nearest Neghbors Searh for Tme-Seres Data Streams Jessa Ln 1, Mha Vlahos 1, Eamonn Keogh 1, Dmtros Gunopulos 1, Janwe Lu 2, Shouan Yu 2, and Jan Le
More informationComputing Volumes of Solids Enclosed by Recursive Subdivision Surfaces
EUROGRAPHI 97 / D. Fellner and L. zrmay-kalos (Guest Edtors) olume 6, (997), Number 3 omputng olumes of olds Enlosed by Reursve ubdvson urfaes Jörg Peters and Ahmad Nasr Abstrat The volume of a sold enlosed
More informationColor Texture Classification using Modified Local Binary Patterns based on Intensity and Color Information
Color Texture Classfaton usng Modfed Loal Bnary Patterns based on Intensty and Color Informaton Shvashankar S. Department of Computer Sene Karnatak Unversty, Dharwad-580003 Karnataka,Inda shvashankars@kud.a.n
More informationParallelism 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 informationTopic 5: semantic analysis. 5.5 Types of Semantic Actions
Top 5: semant analyss 5.5 Types of Semant tons Semant analyss Other Semant tons Other Types of Semant tons revously, all semant atons were for alulatng attrbute values. In a real ompler, other types of
More informationFuzzy set approximation using polar co-ordinates and linguistic term shifting
Johanyák, Z. C., Kovács, S.: Fuzzy set approximation using polar co-ordinates and linguistic term shifting, SAMI 2006, 4 rd Slovakian-Hungarian Joint Symposium on Applied Machine Intelligence, Herl'any,
More informationFor instance, ; the five basic number-sets are increasingly more n A B & B A A = B (1)
Secton 1.2 Subsets and the Boolean operatons on sets If every element of the set A s an element of the set B, we say that A s a subset of B, or that A s contaned n B, or that B contans A, and we wrte A
More informationProper 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 informationPerformance Evaluation of TreeQ and LVQ Classifiers for Music Information Retrieval
Performane Evaluaton of TreeQ and LVQ Classfers for Mus Informaton Retreval Matna Charam, Ram Halloush, Sofa Tsekerdou Athens Informaton Tehnology (AIT) 0.8 km Markopoulo Ave. GR - 19002 Peana, Athens,
More informationLink Graph Analysis for Adult Images Classification
Lnk Graph Analyss for Adult Images Classfaton Evgeny Khartonov Insttute of Physs and Tehnology, Yandex LLC 90, 6 Lev Tolstoy st., khartonov@yandex-team.ru Anton Slesarev Insttute of Physs and Tehnology,
More informationb * -Open Sets in Bispaces
Internatonal Journal of Mathematcs and Statstcs Inventon (IJMSI) E-ISSN: 2321 4767 P-ISSN: 2321-4759 wwwjmsorg Volume 4 Issue 6 August 2016 PP- 39-43 b * -Open Sets n Bspaces Amar Kumar Banerjee 1 and
More informationModule 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 informationHelsinki 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 informationA METHOD FOR RANKING OF FUZZY NUMBERS USING NEW WEIGHTED DISTANCE
Mathematcal and omputatonal pplcatons, Vol 6, No, pp 359-369, ssocaton for Scentfc Research METHOD FOR RNKING OF FUZZY NUMERS USING NEW WEIGHTED DISTNE T llahvranloo, S bbasbandy, R Sanefard Department
More information(1) The control processes are too complex to analyze by conventional quantitative techniques.
Chapter 0 Fuzzy Control and Fuzzy Expert Systems The fuzzy logc controller (FLC) s ntroduced n ths chapter. After ntroducng the archtecture of the FLC, we study ts components step by step and suggest a
More information,.,,
ISSN 49-99 6. 9... /.......... 989.... 85-9.... - /.... //.. 5.. 8.. 5-55. 4... /.... //... 978.... 65-7. 5... :.... - :.: 5..6. /. 99. 46. Vtor su leturer Vtor alh Prof. PhD teh. s. lexandr Ddy ssos.
More information12/2/2009. 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 informationA NOTE ON FUZZY CLOSURE OF A FUZZY SET
(JPMNT) Journal of Process Management New Technologes, Internatonal A NOTE ON FUZZY CLOSURE OF A FUZZY SET Bhmraj Basumatary Department of Mathematcal Scences, Bodoland Unversty, Kokrajhar, Assam, Inda,
More informationNUMERICAL 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 informationReducing Frame Rate for Object Tracking
Reducng Frame Rate for Object Trackng Pavel Korshunov 1 and We Tsang Oo 2 1 Natonal Unversty of Sngapore, Sngapore 11977, pavelkor@comp.nus.edu.sg 2 Natonal Unversty of Sngapore, Sngapore 11977, oowt@comp.nus.edu.sg
More informationAn 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 informationSome Tutorial about the Project. Computer Graphics
Some Tutoral about the Project Lecture 6 Rastersaton, Antalasng, Texture Mappng, I have already covered all the topcs needed to fnsh the 1 st practcal Today, I wll brefly explan how to start workng on
More informationCompiler 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 information2x x l. Module 3: Element Properties Lecture 4: Lagrange and Serendipity Elements
Module 3: Element Propertes Lecture : Lagrange and Serendpty Elements 5 In last lecture note, the nterpolaton functons are derved on the bass of assumed polynomal from Pascal s trangle for the fled varable.
More informationLecture #15 Lecture Notes
Lecture #15 Lecture Notes The ocean water column s very much a 3-D spatal entt and we need to represent that structure n an economcal way to deal wth t n calculatons. We wll dscuss one way to do so, emprcal
More informationLecture 5: Multilayer Perceptrons
Lecture 5: Multlayer Perceptrons Roger Grosse 1 Introducton So far, we ve only talked about lnear models: lnear regresson and lnear bnary classfers. We noted that there are functons that can t be represented
More informationComputer Animation and Visualisation. Lecture 4. Rigging / Skinning
Computer Anmaton and Vsualsaton Lecture 4. Rggng / Sknnng Taku Komura Overvew Sknnng / Rggng Background knowledge Lnear Blendng How to decde weghts? Example-based Method Anatomcal models Sknnng Assume
More informationMachine Learning 9. week
Machne Learnng 9. week Mappng Concept Radal Bass Functons (RBF) RBF Networks 1 Mappng It s probably the best scenaro for the classfcaton of two dataset s to separate them lnearly. As you see n the below
More informationCluster 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 informationPath Following Control of a Spherical Robot Rolling on an Inclined Plane
Sensors & ransduers, Vol., Speal Issue, May 3, pp. 4-47 Sensors & ransduers 3 by IFSA http://www.sensorsportal.om Path Followng Control of a Spheral Robot Rollng on an Inlned Plane ao Yu, Hanxu Sun, Qngxuan
More informationFULLY AUTOMATIC IMAGE-BASED REGISTRATION OF UNORGANIZED TLS DATA
FULLY AUTOMATIC IMAGE-BASED REGISTRATION OF UNORGANIZED TLS DATA Martn Wenmann, Bors Jutz Insttute of Photogrammetry and Remote Sensng, Karlsruhe Insttute of Tehnology (KIT) Kaserstr. 12, 76128 Karlsruhe,
More informationEXPRESSION OF DUAL EULER PARAMETERS USING THE DUAL RODRIGUES PARAMETERS AND THEIR APPLICATION TO THE SCREW TRANSFORMATION
Mathematal and Computatonal Applatons, Vol. 6, No., pp. 68-689,. Assoaton for Sentf Researh EXPRESSION OF DUAL EULER PARAMETERS USING THE DUAL RODRIGUES PARAMETERS AND THEIR APPLICATION TO THE SCREW TRANSFORMATION
More informationLOCAL BINARY PATTERNS AND ITS VARIANTS FOR FACE RECOGNITION
IEEE-Internatonal Conferene on Reent Trends n Informaton Tehnology, ICRTIT 211 MIT, Anna Unversty, Chenna. June 3-5, 211 LOCAL BINARY PATTERNS AND ITS VARIANTS FOR FACE RECOGNITION K.Meena #1, Dr.A.Suruland
More informationX- 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 informationFeature 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 informationPattern Classification: An Improvement Using Combination of VQ and PCA Based Techniques
Ameran Journal of Appled Senes (0): 445-455, 005 ISSN 546-939 005 Sene Publatons Pattern Classfaton: An Improvement Usng Combnaton of VQ and PCA Based Tehnques Alok Sharma, Kuldp K. Palwal and Godfrey
More informationParallel Numerics. 1 Preconditioning & Iterative Solvers (From 2016)
Technsche Unverstät München WSe 6/7 Insttut für Informatk Prof. Dr. Thomas Huckle Dpl.-Math. Benjamn Uekermann Parallel Numercs Exercse : Prevous Exam Questons Precondtonng & Iteratve Solvers (From 6)
More informationMachine 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 informationLoad 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 informationAn 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 informationOptimal shape and location of piezoelectric materials for topology optimization of flextensional actuators
Optmal shape and loaton of pezoeletr materals for topology optmzaton of flextensonal atuators ng L 1 Xueme Xn 2 Noboru Kkuh 1 Kazuhro Satou 1 1 Department of Mehanal Engneerng, Unversty of Mhgan, Ann Arbor,
More informationInverse-Polar Ray Projection for Recovering Projective Transformations
nverse-polar Ray Projecton for Recoverng Projectve Transformatons Yun Zhang The Center for Advanced Computer Studes Unversty of Lousana at Lafayette yxz646@lousana.edu Henry Chu The Center for Advanced
More informationEnhanced AMBTC for Image Compression using Block Classification and Interpolation
Internatonal Journal of Computer Applcatons (0975 8887) Volume 5 No.0, August 0 Enhanced AMBTC for Image Compresson usng Block Classfcaton and Interpolaton S. Vmala Dept. of Comp. Scence Mother Teresa
More informationReading. 14. Subdivision curves. Recommended:
eadng ecommended: Stollntz, Deose, and Salesn. Wavelets for Computer Graphcs: heory and Applcatons, 996, secton 6.-6., A.5. 4. Subdvson curves Note: there s an error n Stollntz, et al., secton A.5. Equaton
More informationMultiscale Heterogeneous Modeling with Surfacelets
759 Multsale Heterogeneous Modelng wth Surfaelets Yan Wang 1 and Davd W. Rosen 2 1 Georga Insttute of Tehnology, yan.wang@me.gateh.edu 2 Georga Insttute of Tehnology, davd.rosen@me.gateh.edu ABSTRACT Computatonal
More information3D Metric Reconstruction with Auto Calibration Method CS 283 Final Project Tarik Adnan Moon
3D Metrc Reconstructon wth Auto Calbraton Method CS 283 Fnal Project Tark Adnan Moon tmoon@collge.harvard.edu Abstract In ths paper, dfferent methods for auto camera calbraton have been studed for metrc
More informationAdaptive Class Preserving Representation for Image Classification
Adaptve Class Preservng Representaton for Image Classfaton Jan-Xun M,, Qankun Fu,, Wesheng L, Chongqng Key Laboratory of Computatonal Intellgene, Chongqng Unversty of Posts and eleommunatons, Chongqng,
More informationOutline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1
4/14/011 Outlne Dscrmnatve classfers for mage recognton Wednesday, Aprl 13 Krsten Grauman UT-Austn Last tme: wndow-based generc obect detecton basc ppelne face detecton wth boostng as case study Today:
More informationAnalysis of ray stability and caustic formation in a layered moving fluid medium
Analyss of ray stablty and aust formaton n a layered movng flud medum Davd R. Bergman * Morrstown NJ Abstrat Caust formaton ours wthn a ray skeleton as optal or aoust felds propagate n a medum wth varable
More informationPost-Processing of Air Entrainment on NASIR Flow Solver Results for Skimming Flow over Stepped Chutes
Proeedngs of the 9th WSEAS Internatonal Conferene on Automat Control, Modelng & Smulaton, Istanbul, Turke, Ma 7-9, 7 Post-Proessng of Ar Entranment on NASIR Flow Solver Results for Skmmng Flow over Stepped
More informationInternational Journal of Pharma and Bio Sciences HYBRID CLUSTERING ALGORITHM USING POSSIBILISTIC ROUGH C-MEANS ABSTRACT
Int J Pharm Bo S 205 Ot; 6(4): (B) 799-80 Researh Artle Botehnology Internatonal Journal of Pharma and Bo Senes ISSN 0975-6299 HYBRID CLUSTERING ALGORITHM USING POSSIBILISTIC ROUGH C-MEANS *ANURADHA J,
More informationA Toolbox for Easily Calibrating Omnidirectional Cameras
A oolbox for Easly Calbratng Omndretonal Cameras Davde Saramuzza, Agostno Martnell, Roland Segwart Autonomous Systems ab Swss Federal Insttute of ehnology Zurh EH) CH-89, Zurh, Swtzerland {davdesaramuzza,
More informationMultilabel Classification with Meta-level Features
Multlabel Classfaton wth Meta-level Features Sddharth Gopal Carnege Mellon Unversty Pttsburgh PA 523 sgopal@andrew.mu.edu Ymng Yang Carnege Mellon Unversty Pttsburgh PA 523 ymng@s.mu.edu ABSTRACT Effetve
More informationTopology Design using LS-TaSC Version 2 and LS-DYNA
Topology Desgn usng LS-TaSC Verson 2 and LS-DYNA Wllem Roux Lvermore Software Technology Corporaton, Lvermore, CA, USA Abstract Ths paper gves an overvew of LS-TaSC verson 2, a topology optmzaton tool
More informationClustering incomplete data using kernel-based fuzzy c-means algorithm
Clusterng noplete data usng ernel-based fuzzy -eans algorth Dao-Qang Zhang *, Song-Can Chen Departent of Coputer Sene and Engneerng, Nanjng Unversty of Aeronauts and Astronauts, Nanjng, 210016, People
More informationThe use of the concept of vague environment in approximate fuzzy reasoning
Kovác, Sz., Kóczy,.T.: The ue of the concept of vague envronment n approxmate fuzzy reaonng, Fuzzy Set Theory and pplcaton, Tatra Mountan Mathematcal Publcaton, Mathematcal Inttute Slovak cademy of Scence,
More informationsuch that is accepted of states in , where Finite Automata Lecture 2-1: Regular Languages be an FA. A string is the transition function,
* Lecture - Regular Languages S Lecture - Fnte Automata where A fnte automaton s a -tuple s a fnte set called the states s a fnte set called the alphabet s the transton functon s the ntal state s the set
More informationChapter 6 Programmng the fnte element method Inow turn to the man subject of ths book: The mplementaton of the fnte element algorthm n computer programs. In order to make my dscusson as straghtforward
More informationGSLM 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 informationPROPERTIES OF BIPOLAR FUZZY GRAPHS
Internatonal ournal of Mechancal Engneerng and Technology (IMET) Volume 9, Issue 1, December 018, pp. 57 56, rtcle ID: IMET_09_1_056 valable onlne at http://www.a aeme.com/jmet/ssues.asp?typeimet&vtype
More informationA Flexible Solution for Modeling and Tracking Generic Dynamic 3D Environments*
A Flexble Soluton for Modelng and Trang Gener Dynam 3D Envronments* Radu Danesu, Member, IEEE, and Sergu Nedevsh, Member, IEEE Abstrat The traff envronment s a dynam and omplex 3D sene, whh needs aurate
More informationA 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 informationSimulation: Solving Dynamic Models ABE 5646 Week 11 Chapter 2, Spring 2010
Smulaton: Solvng Dynamc Models ABE 5646 Week Chapter 2, Sprng 200 Week Descrpton Readng Materal Mar 5- Mar 9 Evaluatng [Crop] Models Comparng a model wth data - Graphcal, errors - Measures of agreement
More informationMulti-Collaborative Filtering Algorithm for Accurate Push of Command Information System
Mult-Collaboratve Flterng Algorthm for Aurate Push of Command Informaton System Cu Xao-long,, Du Bo, Su Guo-Png, Yu Yan Urumq Command College of CPAPF, Urumq 83000, Chna XnJang Tehnal Insttute of Physs
More informationAssignment # 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 informationCluster ( Vehicle Example. Cluster analysis ( Terminology. Vehicle Clusters. Why cluster?
Why luster? referene funton R R Although R and R both somewhat orrelated wth the referene funton, they are unorrelated wth eah other Cluster (www.m-w.om) A number of smlar ndvduals that our together as
More informationImproved Accurate Extrinsic Calibration Algorithm of Camera and Two-dimensional Laser Scanner
JOURNAL OF MULTIMEDIA, VOL. 8, NO. 6, DECEMBER 013 777 Improved Aurate Extrns Calbraton Algorthm of Camera and Two-dmensonal Laser Sanner Janle Kong, Le Yan*, Jnhao Lu, Qngqng Huang, and Xaokang Dng College
More informationFuzzy Rule Interpolation Matlab Toolbox FRI Toolbox
6 IEEE International onference on Fuzzy Systems Sheraton Vancouver Wall entre Hotel, Vancouver, B, anada July 6-, 6 Fuzzy Rule Interpolation Matlab Toolbox FRI Toolbox Zsolt saba Johanyák, Domonkos Tikk,
More informationSupport 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 informationLecture 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 informationLoop Transformations, Dependences, and Parallelization
Loop Transformatons, Dependences, and Parallelzaton Announcements Mdterm s Frday from 3-4:15 n ths room Today Semester long project Data dependence recap Parallelsm and storage tradeoff Scalar expanson
More informationWhat are the camera parameters? Where are the light sources? What is the mapping from radiance to pixel color? Want to solve for 3D geometry
Today: Calbraton What are the camera parameters? Where are the lght sources? What s the mappng from radance to pel color? Why Calbrate? Want to solve for D geometry Alternatve approach Solve for D shape
More informationScalable Parametric Runtime Monitoring
Salable Parametr Runtme Montorng Dongyun Jn Patrk O Nel Meredth Grgore Roşu Department of Computer Sene Unversty of Illnos at Urbana Champagn Urbana, IL, U.S.A. {djn3, pmeredt, grosu}@s.llnos.edu Abstrat
More information