DATA DIMENSIONALITY REDUCTION METHODS FOR ORDINAL DATA

Size: px
Start display at page:

Download "DATA DIMENSIONALITY REDUCTION METHODS FOR ORDINAL DATA"

Transcription

1 DATA DIMENIONALITY REDUCTION METHOD FOR ORDINAL DATA Matn Pokop Hana Řezanková Abstact Fom questonnae suvey we fequently get data, the values ae expessed n odnal (e.g. Lket) scale. The questonnae contans usually a lot of questons, so we get multdmensonal data matx. To smplfy calculatons wth the data t s useful to educe dmensonalty of the dataset. Fo odnal data we use dffeent o mpoved methods compaed to quanttatve data. Ths atcle ncludes the ovevew and compason of dmensonalty educton methods (e.g. pncpal component analyss, facto analyss, multdmensonal scalng, cluste analyss...). Fom these methods we get goups of smla vaables (latent classes), n some cases we can ceate ntepetaton of these new vaables. Key wods: odnal data, dmensonalty educton, latent class models, multdmensonal scalng, cluste analyss JEL Code: C3, C6, C8 Intoducton Am of ths theoetcal study s to descbe data dmensonalty educton methods especally fo odnal data. Ths knd of data we fequently get fom questonnae suveys. Thus we solve the methods to educe the numbe of vaables chaactezng ndvdual objects. Fom bg amount of vaables (questons) we make new latent vaables, whch ae ceated by goups of ognal vaables. Conventonal data dmensonalty educton methods usually assume quanttatve vaables, so we have to use modfed o dffeent statstcal methods. Thee exst seveal methods sometmes wth dffeent esults, so we compae the esults fom vaous methods. Applcaton of the methods n ths text wll be usually descbed n the softwae R. ome methods especally fo categocal data ae descbed n detal (latent class models). 523

2 1 Ovevew of methods Basc methods of the data dmensonalty educton ae pncpal component analyss PCA, facto analyss FA and multdmensonal scalng MD. Classcal FA methods assume lnea elatons among ognal vaables, new latent vaables ae contnuous and nomally dstbuted. Conventonal facto analyss s usually based on coelaton matx analyss, e.g. usng ank coellaton coeffcent. Fo moe detals see Hebák (2007). Common methods of latent vaables dentfcaton ae latent class models. Thee exst many methods and dffeent methods ae avalable n statstcal softwae packages, e.g. latent class cluste models LCC, dscete facto analyss models DFacto, latent tat analyss LTA, latent pofle analyss LPA, latent class egesson models LCR etc. Fo some of these methods n detal see obíšek and Řezanková (2011). 2 Pncpal component analyss ome methods ae based on multdmensonal space pojecton nto the space wth lowe dmenson. Basc method s pncpal component analyss. The am s to fnd eal dmenson of the data. To fnd eal dmensonalty ognal dataset X s tansfomed to the new coodnate system by an othogonal lnea tansfomaton. Let F (esp. G ) be the vecto of the ows coodnates (esp. columns) on the axs on ank s. These two vectos ae elated by the tanston fomula, e. g. n the case of PCA (equatons 1 and 2) thee ae F G 1 k x m G k, 1 k k x p G, k k k (1) (2) whee F denotes the coodnate of the ndvdual on the axs s, of the vaable k on the axs s, assocated to the vaable k, the egenvalue assocated wth the axs s, p the weght assocated to the ndvdual. 524 G denotes the coodnate m the weght k utable count of components goes fom the vaance, whch s explaned by the sum of the vaance of ognal vaables o fom sceeplot of egenvalues o fom the count of egenvalues, whch ae hghe than 1, f we use coelaton matx nstead of covaance matx. utablty and lmtaton of ths method conssts n the esult, when we get fom hgh count of vaables small count of components wth hgh popoton of explaned vaablty.

3 Hgh dependence of obseved vaables s also sutable, stong coelaton among ognal vaables and components too. Instead of conventonal pncpal component analyss fo quanttatve vaables t s possble to use categocal pncpal component analyss CATPCA, whch tansfoms categocal vaables nto quanttatve vaables and does not assume lnea elatons among vaables. Accodng to ebasten et al. (2008) although a PCA appled on categocal data would yeld esults compaable to those obtaned fom a Multple Coespondence Analyss (facto scoes and egenvalues ae lnealy elated), thee ae moe appopate technques to deal wth mxed data types, namely Multple Facto Analyss fo mxed data avalable n the FactoMneR R package. Multple Facto Analyss fom the same package s also an opton. 3 Multdmensonal scalng Othe method based on multdmensonal space pojecton nto the space wth lowe dmenson s multdmensonal scalng MD. ettng of axes (dmensons) s smla to PCA components settng. Multdmensonal scalng s moe geneal than facto analyss, because t s based on any elaton matx among vaables o ndvduals. Method MD s smla to cluste analyss, because t uses dstance matx of vaables o ndvduals pas. Ths dstance can be based on smlaty measue. mlaty of two vaables can be estmated by some of mutual symmetc smlaty measues. Basc smlaty measue of two quanttatve vaables s Peason coelaton coeffcent. To measue smlaty of odnal vaables t s possble to use e.g. peaman o Kendall ank coelaton coefcent o symmetc ommes coeffcent. Fo detals see e.g. Hendl (2006). Accodng to Holland (2008) nonmetc multdmensonal scalng (MD, also NMD and NM) s an odnaton technque, that dffes n seveal ways fom nealy all othe odnaton methods. In most odnaton methods, many axes ae calculated, but only a few ae vewed, owng to gaphcal lmtatons. In MD, a small numbe of axes ae explctly chosen po to the analyss and the data ae ftted to those dmensons; thee ae no hdden axes of vaaton. econd, most othe odnaton methods ae analytcal and theefoe esult n a sngle unque soluton to a set of data. In contast, MD s a numecal technque, that teatvely seeks a soluton and stops computaton when an acceptable soluton has been found, o t stops afte some pe-specfed numbe of attempts. As a esult, an MD odnaton s not a unque soluton and a subsequent MD analyss on the same set of data and followng the 525

4 same methodology wll lkely esult n a somewhat dffeent odnaton. Thd, MD s not an egenvalue-egenvecto technque lke pncpal components analyss o coespondence analyss, that odnates the data such that axs 1 explans the geatest amount of vaance, axs 2 explans the next geatest amount of vaance, and so on. As a esult, an MD odnaton can be otated, nveted, o centeed to any desed confguaton. Unlke othe odnaton methods, MD makes few assumptons about the natue of the data. Fo example, pncpal components analyss assumes lnea elatonshps and ecpocal aveagng assumes modal elatonshps. MD makes nethe of these assumptons, so s well suted fo a wde vaety of data. MD also allows the use of any dstance measue of the samples, unlke othe methods, whch specfy patcula measues, such as covaance o coelaton n PCA o the mpled ch-squaed measue n detended coespondence analyss. The method stats wth a matx of data consstng of n ows of samples and p columns of vaables, Fom ths symmetcal matx of all pawse dstances among samples s calculated wth an appopate dstance measue, such as Eucldean dstance, Manhattan dstance (cty block dstance), and Bay dstance. The MD odnaton wll be pefomed on ths dstance matx. Next, a desed numbe of m dmensons s chosen fo the odnaton. Dstances among samples n statng confguaton ae calculated, typcally wth a Eucldean metc. These dstances ae egessed aganst the ognal dstance matx and the pedcted odnaton dstances fo each pa of samples s calculated. A vaety of egesson methods can be used, ncludng lnea, polynomal, and non-paametc appoaches. In any case, the egesson s ftted by least-squaes. The goodness of ft of the egesson s measued based on the sum of squaed dffeences between odnaton-based dstances and the dstances pedcted by the egesson. Ths goodness of ft s called stess and can be calculated n seveal ways, e.g. fom equaton 3 wth one of the most common beng Kuskal s tess tess d h, h d h, dˆ 2 h h 2, (3) whee d h s the odnated dstance between samples h and, and dˆ s the dstance pedcted fom the egesson. Ths confguaton s then mpoved by movng the postons of samples n odnaton space by a small amount n the decton of steepest descent, the decton n whch stess changes most apdly. The odnaton dstance matx s ecalculated, the egesson pefomed agan and stess ecalculated, and ths ente pocedue of nudgng 526

5 samples and ecalculatng stess s epeated untl some small specfed toleance value s acheved o untl the pocedue conveges by falng to acheve any lowe values of stess, whch ndcates that a mnmum (pehaps local) has been found. A scee dagam (stess vesus numbe of dmensons) can then be plotted, on whch one can dentfy the pont beyond whch addtonal dmensons do not substantally lowe the stess value. A second cteon fo the appopate numbe of dmensons s the ntepetablty of the odnaton, that s, whethe the esults make sense. tess nceases both wth the numbe of samples and wth the numbe of vaables. R has two man MD functons avalable, somd, whch s pat of the MA lbay, and metamd, whch s pat of the vegan lbay. The metamd outne allows geate automaton of the odnaton pocess, so s usually the pefeed method. The metamd functon uses somd n ts calculatons as well as seveal helpe functons. The metamd outne also has the useful default behavo of followng the odnaton wth a otaton va pncpal components analyss such that MD axs 1 eflects the pncpal souce of vaaton and so on, as s chaactestc of egenvalue methods. 3 Ovevew of latent class and latent class egesson models Accodng to Lnze (2011) latent class analyss s a statstcal technque fo the analyss of multvaate categocal data. When obseved data take the fom of a sees of categocal esponses (as fo example, n publc opnon suveys, ndvdual-level votng data, studes of nte-ate elablty, o consume behavo and decson-makng), t s often ou nteest to nvestgate souces of confoundng between the obseved vaables, dentfy and chaacteze clustes of smla cases, and appoxmate the dstbuton of obsevatons acoss many vaables of nteest. Latent class models ae a useful tool fo accomplshng these goals. The latent class model seeks to statfy the coss-classffcaton table of obseved (manfest) vaables by an unobseved (latent) categocal vaable, that elmnates all confoundng between the manfest vaables. Responses to all of the manfest vaables ae assumed to be statstcally ndependent. The model, n effect, pobablstcally goups each obsevaton nto a latent class, whch n tun poduces expectatons about how that obsevaton wll espond on each manfest vaable. Although the model does not automatcally detemne the numbe of latent classes n a gven data set, t does offe a vaety of pasmony and goodness of ft statstcs, that the we may use n ode to make a theoetcally and empcally assessment. 527

6 Because the unobseved latent vaable s nomnal (membeshp of a class), the latent class model s actually a type of fnte mxtue model. The component dstbutons n the mxtue ae coss-classffcaton tables of equal dmenson to the obseved table of manfest vaables, and, followng the assumpton of condtonal ndependence, the fequency n each cell of each component table s smply the poduct of the espectve class-condtonal magnal fequences (the paametes estmated by the latent class model ae the popoton of obsevatons n each latent class, and the pobabltes of obsevng each esponse to each manfest vaable, condtonal on latent class). A weghted sum of these component tables foms an appoxmaton (o, densty estmate) of the dstbuton of cases acoss the cells of the obseved table. Obsevatons wth smla sets of esponses on the manfest vaables wll tend to cluste wthn the same latent classes. An extenson of ths basc model pemts the ncluson of covaates to pedct latent class membeshp. Wheeas n the basc model, evey obsevaton has the same pobablty of belongng to each latent class po to obsevng the esponses to the manfest vaables, n the moe geneal latent class egesson model, these po pobabltes vay by ndvdual as a functon of some set of ndependent vaables. polca s a softwae package fo the estmaton of latent class and latent class egesson models fo polytomous outcome vaables (vaables wth moe than two dstnct categoes), mplemented n the R The basc latent class model s a fnte mxtue model, n whch the component dstbutons ae assumed to be mult-way coss-classffcaton tables wth all vaables mutually ndependent. The latent class egesson model futhe enables us to estmate the effects of covaates on pedctng latent class membeshp. polca uses expectaton-maxmzaton and Newton-Raphson algothms to fnd maxmum lkelhood estmates of the model paametes. 4 Latent class models Accodng to Lnze (2011) the basc latent class model s a fnte mxtue model n whch the component dstbutons ae assumed to be mult-way coss-classfcaton tables wth all vaables mutually ndependent. uppose we obseve J polytomous categocal vaables (the manfest vaables), each of whch contans K j possble outcomes, fo ndvduals = 1,...,N. The manfest vaables may have dffeent numbes of outcomes, hence the ndexng by j. Denote as Y jk the obseved values of the J manfest vaables such that Y jk = 1 f espondent gves the k-th esponse to the j-th vaable, and Y jk = 0 othewse, whee j = 1,..., J and k = 1,...,K j. The latent class model appoxmates the obseved jont dstbuton of the manfest 528

7 vaables as the weghted sum of a fnte numbe R of consttuent coss-classfcaton tables. Let π denote the class-condtonal pobablty, that an obsevaton n class = 1,...,R poduces the k-th outcome on the j-th vaable. Wthn each class, fo each manfest vaable, K j theefoe k 1 1. Futhe denote as p the R mxng popotons that povde the weghts n the weghted sum of the component tables, wth p 1. The values of p ae also efeed to as the po pobabltes of latent class membeshp, as they epesent the uncondtonal pobablty that an ndvdual wll belong to each class befoe takng nto account the esponses Y jk povded on the manfest vaables. The pobablty that an ndvdual n class poduces a patcula set of J outcomes on the manfest vaables, assumng condtonal ndependence of the outcomes Y gven class membeshps, s the poduct f Y J K j ; jk Y. (4) j 1 k 1 The pobablty densty functon acoss all classes s the weghted sum P Y R J K j Y, p jk. (5) p 1 j 1 k 1 The paametes estmated by the latent class model ae p and π. Gven estmates pˆ and ˆ of p and π, espectvely, the posteo pobablty that each ndvdual belongs to each class, condtonal on the obseved values of the manfest vaables, can be calculated usng Bayes' fomula: whee 1,...R Pˆ Y pˆ f Y ; ˆ R pˆ f Y ; ˆ q1 q q wth any fnte mxtue model, the EM algothm s applcable because each ndvdual's class 529, (6). Recall that the ˆ ae estmates of outcome pobabltes condtonal on class. It s mpotant to eman awae that the numbe of ndependent paametes estmated by the latent class model nceases apdly wth R, J, and K j. Gven these values, the numbe of paametes s R K j R 1 j 1. If ths numbe exceeds ethe the total numbe of obsevatons, o one fewe than the total numbe of cells n the coss-classfcaton table of the manfest vaables, then the latent class model wll be undentfed. polca estmates the latent class model by maxmzng the log-lkelhood functon N R K j 1 1 j 1 k 1 J ln L ln p (7) wth espect to p and π, usng the expectaton-maxmzaton (EM) algothm. Ths loglkelhood functon s dentcal n fom to the standad fnte mxtue model log-lkelhood. As Y jk

8 membeshp s unknown and may be teated as mssng data. The EM algothm poceeds teatvely. Begn wth abtay ntal values of pˆ and ˆ, and label them the expectaton step, calculate the mssng class membeshp pobabltes usng Equaton 6, substtutng n pˆ and ˆ pˆ and ˆ. In. In the maxmzaton step, update the paamete estmates by ˆ Y wth maxmzng the log-lkelhood functon gven these posteo P N new 1 pˆ Pˆ Y (8) N 1 as the new po pobabltes and N Y Pˆ Y j new 1 ˆ (9) j N Pˆ Y 1 new as the new class-condtonal outcome pobabltes. In Equaton 9, ˆ s the vecto of length K j of class- condtonal outcome pobabltes fo the j-th manfest vaable; and Y j s the N K j matx of obseved outcomes Y jk on that vaable. The algothm epeats these steps, assgnng the new to the, untl the oveall log-lkelhood eaches a maxmum and ceases to ncement beyond some abtaly small value. j polca takes advantage of the teatve natue of the EM algothm to make t possble to estmate the latent class model even when some of the obsevatons on the manfest vaables ae mssng. Although polca does offe the opton to lstwse delete obsevatons wth mssng values befoe estmatng the model, t s not necessay to do so. Instead, when detemnng the poduct n Equaton 4 and the sum n the numeato of Equaton 9, polca smply excludes fom the calculaton any manfest vaables wth mssng obsevatons. The pos ae updated n Equaton 6 usng as many o as few manfest vaables as ae obseved fo each ndvdual. Dependng on the ntal values chosen fo complexty of the latent class model beng estmated, the EM algothm may only fnd a local maxmum of the log-lkelhood functon, athe than the desed global maxmum. Fo ths eason, t s always advsable to e-estmate a patcula model a couple of tmes when usng polca, n an attempt to fnd the global maxmze to be taken as the maxmum lkelhood soluton. One of the benefts of latent class analyss, n contast to othe statstcal technques fo clusteed data, s the vaety of tools avalable fo assessng model ft and detemnng an appopate numbe of latent classes R fo a gven data set. In some applcatons, the numbe 530 pˆ and ˆ, and the

9 of latent classes wll be selected fo pmaly theoetcal easons. In othe cases, howeve, the analyss may be of a moe exploatoy natue, wth the objectve beng to locate the best fttng o most pasmonous model. We may then begn by fttng a complete ndependence model wth R = 1, and then teatvely nceasng the numbe of latent classes by one untl a sutable ft has been acheved. Addng an addtonal class to a latent class model wll ncease the ft of the model, but at the sk of fttng to nose, and at the expense of estmatng a futhe K 1 1 model paametes. Pasmony ctea seek to stke a balance between j j ove- and unde-fttng the model to the data by penalzng the log-lkelhood by a functon of the numbe of paametes beng estmated. The two most wdely used pasmony measues ae the Bayesan nfomaton cteon, o BIC and Akake nfomaton cteon, o AIC. Pefeed models ae those that mnmze values of the BIC and/o AIC. Let Λ epesent the maxmum log-lkelhood of the model and Φ epesent the total numbe of estmated paametes. Then, AIC = -2Λ+2Φ and BIC = -2Λ+ Φ lnn. polca calculates these paametes automatcally when estmatng the latent class model. The BIC wll usually be moe appopate fo basc latent class models because of the elatve smplcty. Calculatng Peason's Χ 2 goodness of ft and lkelhood ato ch-squae (G 2 ) statstcs fo the obseved vesus pedcted cell counts s anothe method to help detemne how well a patcula model fts the data, fo moe detals see Lnze (2011). Lke the AIC and BIC, these statstcs ae outputted automatcally afte callng polca. 5 Optmal scalng by Gf methods The challenge wth categocal vaables s to fnd a sutable way to epesent dstances between vaable categoes and ndvduals n the factoal space. To ovecome ths poblem, we can look fo a non-lnea tansfomaton of each vaable, whethe t s nomnal, odnal, polynomal, o numecal wth optmal scalng. Ths s well explaned n Leeuw (2009), and an mplementaton s avalable n the coespondng R package homals. As extenson to ths tansfomaton, havng nonmetc vaables, we can use dmensonalty educton methods, e.g. nonlnea pncpal component analyss (NLPCA). The tem nonlnea petans to nonlnea tansfomatons of the obseved vaables. In Gf temnology, NLPCA can be defned as homogenety analyss wth estctons on the quantfcaton matx. 6 Fuzzy Clusteng 531

10 Accodng to Oksanen (2010) we have so fa woked wth classffcaton methods, whch mplctly assume, that thee ae dstnct classes. The eal stuaton s usually dffeent. If thee ae classes, they ae vague and have ntemedate and untypcal cases. Wth one wod, they ae fuzzy. Fuzzy classffcaton means, that each obsevaton has a cetan pobablty of belongng to a cetan class. In the csp case, t has pobablty 1 of belongng to a cetan class, and pobablty 0 of belongng to any othe class. In a fuzzy case, t has pobablty < 1 fo the best class, and pobabltes > 0 fo seveal othe classes. Fuzzy classffcaton s smla to K-means clusteng n fndng the optmal classffcaton fo a gven numbe of classes, but the poduced classffcaton s fuzzy: the esult s a pobablty pofle of class membeshp. The fuzzy clusteng s povded by functon fanny (equaton 10) n package cluste. Requested membeshp pobabltes we get fom the mnmalzaton of the functon n n 2 2 u u d k h jh j 1 j 1 J F n h1 2 2uh j 1 (10) whee the values of u ae membeshp pobabltes and values of d ae Eucldean dstances among the objects. It s dffcult to show the fuzzy esults gaphcally, but t s possble to use stas functon (wth many optonal paametes) to show the pobablty pofle, and t daws a convex hull of the csp classfcaton. The sze of the secto shows the pobablty of the class membeshp and n clea cases one of the segments s domnant. Concluson Ths study was amed to the ovevew of data dmensonalty educton method especally fo categocal (odnal) data. ome advantages and dffcultes of the methods wee pesented, n futue eseach these methods wll be aplled on eal dataset and compason of the esults wll be pefomed. Refeences 1. Hebák, Pet. Víceozměné statstcké metody 3. Paha: Infomatoum, Hendl, Jan. Přehled statstckých metod: analýza a metaanalýza dat. Paha: Potál, Holland, teven M. Non-metc multdmensonal scalng (MD). Athens: R foge, Le, ebasten, and Josse, Jule, and Husson, Fancos. FactoMneR: An R package fo multvaate analyss. Jounal of statstcal softwae 25 Mach 2008: sec

11 5. Leeuw, Jan, and Ma, Patck. Gf Methods fo Optmal calng n R: The Package homals. Jounal of statstcal softwae 31 Aug. 2009: sec Lnze, Dew, and Lews, Jeffey. polca: An R Package fo Polytomous Vaable Latent Class Analyss. Jounal of statstcal softwae 42 June 2011: sec Oksanen, Ja. Cluste Analyss: Tutoal wth R. Mendeley, obíšek, Lukáš, and Řezanková, Hana. ovnání metod po edukc dmenzonalty aplkovaných na odnální poměnné. Acta Oeconomca Pagensa : Contact Matn Pokop Vysoká škola ekonomcká v Paze nám. W. Chuchlla Paha 3 mas@post.cz Hana Řezanková Vysoká škola ekonomcká v Paze nám. W. Chuchlla Paha 3 hana.ezankova@vse.cz 533

COMPARISON OF DIMENSIONALITY REDUCTION METHODS APPLIED TO ORDINAL DATA

COMPARISON OF DIMENSIONALITY REDUCTION METHODS APPLIED TO ORDINAL DATA The 7 th Intenatonal Days of tatstcs and Economcs, Pague, eptembe 19-21, 2013 COMPARION OF DIMENIONALITY REDUCTION METHOD APPLIED TO ORDINAL DATA Matn Pokop Hana Řezanková Abstact The pape deals wth the

More information

A SAS Macro for Finding Optimal k-means Clustering in One Dimension with Size Constraints

A SAS Macro for Finding Optimal k-means Clustering in One Dimension with Size Constraints Pape SD-02 A SAS Maco fo Fndng Optmal k-means Clusteng n One Dmenson wth Sze Constants Fengao Hu, Geoga Regents Unvesty; Robet E. Johnson, Vandeblt Unvesty ABSTRACT Wang and Song (2011) poposed a k-means

More information

FUNDAMENTALS OF RADIOMETRY

FUNDAMENTALS OF RADIOMETRY 1 FUNDAMENTALS OF RADIOMETRY Lectue 5 Radomety 2 Quanttatve measuement of the popetes of EM adaton nteacton wth matte o emsson fom t Radomety deals wth total EM adaton We extend the concept of adans n

More information

Volumetric Textures. Esteban W. Gonzalez Clua 1 Marcelo Dreux 2

Volumetric Textures. Esteban W. Gonzalez Clua 1 Marcelo Dreux 2 Volumetc Textues Esteban W. Gonzalez Clua 1 Macelo Deux 2 1 PUC-Ro - Depatment of Compute Scence Rua Maquês de São Vcente, 255, 22443-900, Ro de Janeo, RJ, Bazl esteban @nf.puc-o.b 2 PUC-RIO Mechancal

More information

Robust Proper Clustering Structure Fuzzy Modeling for Function Approximation

Robust Proper Clustering Structure Fuzzy Modeling for Function Approximation Robust Pope lusteng Stuctue Fuzzy Modelng fo Functon Appoxmaton hh-hng Hsao Depatment of Electcal Engneeng Kao Yuan Unvesty Kaohsung ounty,tawan RO hsao@ooneentustedutw Shun-Feng Su Depatment of Electcal

More information

Effectiveness evaluation of electronic warfare command and control system based on grey AHP method

Effectiveness evaluation of electronic warfare command and control system based on grey AHP method Avalable onlne www.jocp.com Jounal of hemcal and Phamaceutcal Reseach, 014, 6(7):535-54 Reseach Atcle ISSN : 0975-7384 ODEN(USA) : JPR5 Effectveness evaluaton of electonc wafae command and contol system

More information

Planar Curve Representation and Matching

Planar Curve Representation and Matching Plana Cuve Repesentaton and Matchng Mahe Al-Khayat and Fahad Kamanga Compute Scence and Engneeng Depatment Unvesty of Texas at Alngton Alngton, Texas 7609, USA [al-khay kamanga]@cse.uta.edu Abstact In

More information

An Optimization Procedure In A Production Line Of Sokat Soap Industry, Ikotun, Lagos State. Adamawa State, Nigeria.

An Optimization Procedure In A Production Line Of Sokat Soap Industry, Ikotun, Lagos State. Adamawa State, Nigeria. An Optmzaton Pocedue In A Poducton Lne Of Sokat Soap Industy, Ikotun, Lagos State. 1 O.S. Balogun, 2 A.A. Olatun and 3 A.A. Momoh 1 Depatment of Statstcs and Opeatons Reseach, Modbbo Adama Unvesty of Technology,

More information

Outlier Detection in 3D Coordinate Transformation with Fuzzy Logic

Outlier Detection in 3D Coordinate Transformation with Fuzzy Logic Acta Montanstca Slovaca Ročník 7 (22), číslo, -8 Outle Detecton n 3D Coodnate ansfomaton wth Fuzzy Logc Yasemn Ssman, Aslan Dlave 2 and Sebahattn Bektas Coodnate measuements nevtably contan outles that

More information

Risk assessment research for power transformer based on fuzzy synthetic evaluation

Risk assessment research for power transformer based on fuzzy synthetic evaluation Avalable onlne www.jocp.com Jounal of Chemcal and Phamaceutcal Reseach, 23, 5(2):62-68 Reseach Atcle ISSN : 975-7384 CODEN(USA) : JCPRC5 Rsk assessment eseach fo powe tansfome based on fuzzy synthetc evaluaton

More information

arxiv: v1 [math.co] 21 Jan 2016

arxiv: v1 [math.co] 21 Jan 2016 PROOF OF BIJECTION FOR COMBINATORIAL NUMBER SYSTEM axv:60.05794v [math.co] Jan 06 ABU BAKAR SIDDIQUE, SAADIA FARID, AND MUHAMMAD TAHIR Abstact. Combnatoal numbe system epesents a non-negatve natual numbes

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

FUZZY ARTMAP AND NEURAL NETWORK APPROACH TO ONLINE PROCESSING OF INPUTS WITH MISSING VALUES

FUZZY ARTMAP AND NEURAL NETWORK APPROACH TO ONLINE PROCESSING OF INPUTS WITH MISSING VALUES FUZZY ARTMAP AD EURAL ETWORK APPROACH TO OLIE PROCESSIG OF IPUTS WITH MISSIG VALUES F.V. elwamondo* and T. Mawala* * School of Electcal and Infomaton Engneeng, Unvesty of the Wtwatesand, Johannesbug, Pvate

More information

One-Dimensional Linear Local Prototypes for Effective Selection of Neuro-Fuzzy Sugeno Model Initial Structure

One-Dimensional Linear Local Prototypes for Effective Selection of Neuro-Fuzzy Sugeno Model Initial Structure One-Dmensonal Lnea Local Pototypes fo Effectve Selecton of euo-fuzzy Sugeno Model Intal Stuctue Jace Kabzńs Insttute of Automatc Contol, Techncal Unvesty of Lodz, Stefanowsego 8/22, 90 924 Lodz, Poland,

More information

A Bayesian Approach toward Active Learning for Collaborative Filtering

A Bayesian Approach toward Active Learning for Collaborative Filtering A Bayesan Appoach towad Actve Leanng fo Collaboatve Flteng Rong Jn Depatment of Compute Scence and Engneeng Mchgan State Unvesty ong@cse.cmu.edu Abstact Collaboatve flteng s a useful technque fo eplotng

More information

NODAL AND LOOP ANALYSIS TECHNIQUES. Develop systematic techniques to determine all the voltages and currents in a circuit NODE ANALYSIS

NODAL AND LOOP ANALYSIS TECHNIQUES. Develop systematic techniques to determine all the voltages and currents in a circuit NODE ANALYSIS NODAL AND LOOP ANALYSS TECHNQUES LEANNG GOALS NODAL ANALYSS LOOP ANALYSS Develop systematic techniques to determine all the voltages and currents in a circuit NODE ANALYSS One of the systematic ways to

More information

Adjusting the Contact Surface of Forming Tools in Order to Compensate for Elastic Deformations during the Process

Adjusting the Contact Surface of Forming Tools in Order to Compensate for Elastic Deformations during the Process Adjustng the Contact Suface of Fomng Tools n Ode to Compensate fo Elastc Defomatons dung the Pocess Knut Goßmann, Hajo Weme, Andè Hadtmann, Las Pente, Sebastan Kechenbaue Insttute fo Machne Tools and Contol

More information

An Object Based Auto Annotation Image Retrieval System

An Object Based Auto Annotation Image Retrieval System Poceedngs of the 5th WSEAS Intenatonal Confeence on Telecommuncatons and Infomatcs, Istanbul, Tukey, May 27-29, 2006 (pp509-54) An Obect Based Auto Annotaton Image Reteval System Pe-Cheng Cheng, Been-Chan

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

Soochow University: Description and Analysis of the Chinese Word Sense Induction System for CLP2010

Soochow University: Description and Analysis of the Chinese Word Sense Induction System for CLP2010 Soochow Unvesty: Descpton and Analyss of the Chnese Wod Sense Inducton System fo CLP2010 Hua Xu Bng Lu Longhua Qan Guodong Zhou Natual Language Pocessng Lab School of Compute Scence and Technology Soochow

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

Hadoop based Feature Selection and Decision Making Models on Big Data

Hadoop based Feature Selection and Decision Making Models on Big Data Indan Jounal of Scence and Technology, Vol 9(0), DOI: 0.7485/jst/206/v90/88905, Mach 206 ISSN (Pnt) : 0974-6846 ISSN (Onlne) : 0974-5645 Hadoop based Featue Selecton and Decson Makng Models on Bg Data

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

Refraction Ray Normal

Refraction Ray Normal Ray Nomal wave cests θ Bounday θ θ θ less dense, n1 moe dense, n2 (>n1) Moe Smply: θ θ Note: (Lght) Rays bend towads the nomal when gong fom a egon of low ndex of efacton to a egon of hgh ndex of efacton

More information

Supervised and Unsupervised Text Classification via Generic Summarization

Supervised and Unsupervised Text Classification via Generic Summarization Intenatonal Jounal of Compute Infomaton Systems and Industal Management Applcatons. ISSN 50-7988 Volume 5 (03) pp. 509-55 MIR Labs, www.mlabs.net/jcsm/ndex.html Supevsed and Unsupevsed Text Classfcaton

More information

Real Time Face Recognition Using Polynomial Regression and Sub-region Color Component Distribution

Real Time Face Recognition Using Polynomial Regression and Sub-region Color Component Distribution Real Tme Face Recognton Usng Polynomal Regesson and Sub-egon Colo Component Dstbuton Manshanka Mondal, Md. Almas Hossan, Md. Matu Rahman, Kamul Hasan Talukde* Compute Scence and Engneeng Dscplne Khulna

More information

AN EXPLICIT GROWTH MODEL OF THE STEREO REGION GROWING ALGORITHM FOR PARALLEL PROCESSING

AN EXPLICIT GROWTH MODEL OF THE STEREO REGION GROWING ALGORITHM FOR PARALLEL PROCESSING Intenatonal Achves of Photogammety, Remote Sensng and Spatal Infomaton Scences, Vol. XXXVIII, Pat 5 AN EXPLICIT GROWTH MODEL OF THE STEREO REGION GROWING ALGORITHM FOR PARALLEL PROCESSING Dongoe Shn*,

More information

Journal of Engineering Science and Technology Review 7 (1) (2014) Research of Anti-Noise Image Salient Region Extraction Method

Journal of Engineering Science and Technology Review 7 (1) (2014) Research of Anti-Noise Image Salient Region Extraction Method Jest Jounal of Engneeng Scence and echnology Revew 7 () (04) 43 47 JOURNAL OF Engneeng Scence and echnology Revew www.jest.og Reseach of Ant-Nose Image Salent Regon Extacton Method Bng XU,*, Shy XIE and

More information

Performance Evaluation of Neighborhood Signature Techniques for Peer-to-Peer Search

Performance Evaluation of Neighborhood Signature Techniques for Peer-to-Peer Search Pefomance Evaluaton of Neghbohood gnatue Technques fo Pee-to-Pee each e L Wang-Chen Lee * Anand vasubamanam Depatment of Compute cence and Engneeng Pennsylvana tate Unvesty Unvesty Pak, PA 16802, UA E-al:

More information

Flood Runoff Simulation Using Grid-based Radar Rainfall and Vflo TM Model

Flood Runoff Simulation Using Grid-based Radar Rainfall and Vflo TM Model Flood Runoff Smulaton Usng Gd-based Rada Ranfall and Vflo TM Model HUI SEONG NOH 1, BYUNG SIK KIM 2, NA RAE KANG 3, HUNG SOO KIM 4 1 Dept. of Cvl Engneeng, Inha Unvesty, Incheon, Koea, heesung80@hanmal.net

More information

Dynamic Optimization of Structures Subjected to Earthquake

Dynamic Optimization of Structures Subjected to Earthquake IACSIT Intenatonal Jounal of Engneeng and Technology, Vol. 8, No. 4, August 06 Dynamc Optmzaton of Stuctues Subjected to Eathquake Aleza Lavae and Aleza Lohasb Abstact To educe the oveall tme of stuctual

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

Weighted Piecewise LDA for Solving the Small Sample Size Problem in Face Verification

Weighted Piecewise LDA for Solving the Small Sample Size Problem in Face Verification TNN05-P485 Weghted Pecewse LDA fo Solvng the Small Sample Sze Poblem n Face Vefcaton Maos Kypeountas, Anastasos Tefas, Membe, IEEE, and Ioanns Ptas, Seno Membe, IEEE Abstact A novel algothm that can be

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

Recognizing Facial Expressions Based on Gabor Filter Selection

Recognizing Facial Expressions Based on Gabor Filter Selection Recognzng Facal Expessons Based on Gabo Flte Selecton Zang Zhang, Xaomn Mu, Le Gao School of Infomaton Engneeng Zhengzhou Unvest Zhengzhou, Chna Abstact Recognton of human emotonal state s an mpotant component

More information

Beyond a Simple Physically Based Blinn-Phong Model in Real-Time

Beyond a Simple Physically Based Blinn-Phong Model in Real-Time Pactcal Physcally Based Shadng n Flm and Game Poducton: Beyond a Smple Physcally Based Blnn-Phong Model n Real-Tme. Intoducton Yoshhau Gotanda t-ace, Inc. http://eseach.t-ace.com/ We pesented ou mplementaton

More information

Performance Evaluation of WDM LAN s with Propagation Delay Constraint Analysis

Performance Evaluation of WDM LAN s with Propagation Delay Constraint Analysis Pefomance Evaluaton of WD AN s wth Popagaton Dela Constant Analss IOANNI E. POUNOURAKI Depatment of Electcal & Compute Engneeng Natonal echncal Unvest of Athens 57 73 Zogaphou, Athens GREECE Abstact: -In

More information

Feature Extraction for Collaborative Filtering: A Genetic Programming Approach

Feature Extraction for Collaborative Filtering: A Genetic Programming Approach IJCSI Intenatonal Jounal of Compute Scence Issues Vol. 9 Issue 5 No Septembe 0 www.ijcsi.og 348 Featue xtacton fo Collaboatve Flteng: A Genetc Pogammng Appoach Deepa Anand Depatment of Compute Scence Chst

More information

Rigid Body Segmentation and Shape Description from Dense Optical Flow Under Weak Perspective

Rigid Body Segmentation and Shape Description from Dense Optical Flow Under Weak Perspective , VOL. 19, NO. 2, FEBRUARY 1997 139 Rgd Body Segmentaton and Shape Descpton fom Dense Optcal Flow Unde Weak Pespectve Joseph Webe and Jtenda Malk Abstact We pesent an algothm fo dentfyng and tackng ndependently

More information

Constructing Service Semantic Link Network Based on the Probabilistic Graphical Model

Constructing Service Semantic Link Network Based on the Probabilistic Graphical Model Intenatonal Jounal of omputatonal Intellgence ystems, Vol. 5, No. 6 (Novembe, 2012), 1040-1051 onstuctng evce emantc Lnk Netwok Based on the Pobablstc Gaphcal odel ANPING ZHAO * ollege of ompute and Infomaton

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

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

We are IntechOpen, the world s leading publisher of Open Access books Built by scientists, for scientists. International authors and editors

We are IntechOpen, the world s leading publisher of Open Access books Built by scientists, for scientists. International authors and editors We ae IntechOpen, the wold s leadng publshe of Open Access books Bult by scentsts, fo scentsts 3,900 116,000 10 Open access books avalable Intenatonal authos and edtos Downloads Ou authos ae among the

More information

Enhance the Alignment Accuracy of Active Shape Models Using Elastic Graph Matching

Enhance the Alignment Accuracy of Active Shape Models Using Elastic Graph Matching Enhance the Algnment Accuacy of Actve Shape Models Usng Elastc Gaph Matchng Sanqang Zhao 1,, Wen Gao, Shguang Shan, Baoca Yn 1 1) Multmeda and Intellgent Softwae Technology Lab, Beng Unvesty of Technology,

More information

y and the total sum of

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

More information

Fuzzy clustering ensemble based on mutual information

Fuzzy clustering ensemble based on mutual information Poceedngs of the 6th WSEAS Intenatonal Confeence on Smulaton, Modellng and Optmzaton, Lsbon, Potugal, Septembe 22-24, 26 476 Fuzz clusteng ensemble based on mutual nfomaton YAN GAO SHIWEN GU LIMING XIA

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

A LIGHT SCATTERING METHOD TO MEASURE REAL-TIME PARTICULATE EMISSIONS

A LIGHT SCATTERING METHOD TO MEASURE REAL-TIME PARTICULATE EMISSIONS A Lght Scatteng Method to Measue Real-tme Patculate Emssons Rose Gong & Keyn Wllams 6 th Austalasan Tanspot Reseach Foum Wellngton New Zealand -3 Octobe 003 A LIGHT SCATTERING METHOD TO MEASURE REAL-TIME

More information

Improved Methods on PCA Based Human Face Recognition for Distorted Images

Improved Methods on PCA Based Human Face Recognition for Distorted Images Poceedngs of the Intenatonal MultConfeence of Engnees and Compute Scentsts 016 Vol I,, Mach 16-18, 016, Hong Kong Impoved Methods on PCA Based Human Face Recognton fo Dstoted Images Buce Poon, M. Ashaful

More information

Unsupervised Learning

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

More information

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

Dual Tree Complex Wavelet Transform for Face Recognition Using PCA Algorithm

Dual Tree Complex Wavelet Transform for Face Recognition Using PCA Algorithm B.Dhamenda,INDIA / Intenatonal Jounal of Reseach and Computatonal echnology, Vol.5 Issue.4 ISSN: 0975-5662, Dec, 2013 www.jct.og Dual ee Complex Wavelet ansfom fo Face Recognton Usng PCA Algothm Dhamenda

More information

Interactive NURBS Tutorial in Virtual Reality

Interactive NURBS Tutorial in Virtual Reality Inteacte NURBS Ttoal n Vtal Realty Eng 477 Poect Fa Octobe 10 2006 Edmond Nadle enadle@mch.ed 734 913-6970 Eng 477 Poect Fa Octobe 10 2006 Eng 477 Poect Fa Octobe 10 2006 NURBS Ces and Sfaces = Non-Unfom

More information

NAG Fortran Library Chapter Introduction. G10 Smoothing in Statistics

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

More information

3D Head Tracking Based on Recognition and Interpolation Using a Time-Of- Flight Depth Sensor

3D Head Tracking Based on Recognition and Interpolation Using a Time-Of- Flight Depth Sensor 3D Head Tacng Based on Recognton and Intepolaton Usng a Tme-Of- Flght Depth Senso Salh Bua Götü 1 and Calo Tomas 1,2 1 Canesta Inc., 2 Due Unvesty bgotu@canesta.co tomas@cs.due.edu Abstact Ths pape descbes

More information

THE MODELLING OF INVOLUTE TEETH GENERATION, WITH THE RELATIVE GENERATING TRAJECTORIES METHOD

THE MODELLING OF INVOLUTE TEETH GENERATION, WITH THE RELATIVE GENERATING TRAJECTORIES METHOD THE MODELLING OF INVOLUTE TEETH GENERATION, WITH THE RELATIVE GENERATING TRAJECTORIES METHOD Vgl Gabel TEODOR 1, Ncuşo BAROIU 1, Flon SUSAC 1 and Ncolae OANCEA 1 ABSTRACT: The modellng of a cul of sufaces

More information

Multi-Objective Optimization Using LS-OPT

Multi-Objective Optimization Using LS-OPT 6. LS-DYNA Anwendefoum, fankenthal 007 Optmeung Mult-Objectve Optmzaton Usng LS-OPT Tusha Goel, Nelen Stande Lvemoe Softwae Technology Copoaton, Lvemoe, CA, USA Summay: Most engneeng optmzaton poblems

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

LEAST SQUARES. RANSAC. HOUGH TRANSFORM.

LEAST SQUARES. RANSAC. HOUGH TRANSFORM. LEAS SQUARES. RANSAC. HOUGH RANSFORM. he sldes are from several sources through James Has (Brown); Srnvasa Narasmhan (CMU); Slvo Savarese (U. of Mchgan); Bll Freeman and Antono orralba (MI), ncludng ther

More information

GENERAL MODEL FOR AIRBORNE AND SPACEBORNE LINEAR ARRAY SENSORS

GENERAL MODEL FOR AIRBORNE AND SPACEBORNE LINEAR ARRAY SENSORS GENERAL MODEL FOR AIRBORNE AND SPAEBORNE LINEAR ARRAY SENSORS Danela Pol Insttute of Geodesy and Photogammety, Swss Fedeal Insttute of Technology, Zuch, Swtzeland danela@geod.baug.ethz.ch KEY WORDS: Oentaton,

More information

NURBS Curve Shape Modification and Fairness Evaluation for Computer Aided Aesthetic Design

NURBS Curve Shape Modification and Fairness Evaluation for Computer Aided Aesthetic Design oceedngs of the 7th WSEAS Int. Conf. on Sgnal ocessng, Computatonal Geomet & Atfcal Vson, Athens, Geece, August 4-6, 7 38 NURBS Cuve Shape Modfcaton and Faness Evaluaton fo Compute Aded Aesthetc Desgn

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

Comparing performance of SRAMT-LE vs other layered encoding schemes regarding TCP friendliness

Comparing performance of SRAMT-LE vs other layered encoding schemes regarding TCP friendliness Compang pefomance of SRAMT-LE vs othe layeed encodng schemes egadng TCP fendlness Chstos Bouas, Apostolos Gkamas Reseach Academc Compute Technology Insttute, 61 Rga Feaou St, GR-26221 Patas, Geece Compute

More information

A METHODOLOGY FOR RATING AND RANKING HAZARDS IN MARITIME FORMAL SAFETY ASSESSMENT USING FUZZY

A METHODOLOGY FOR RATING AND RANKING HAZARDS IN MARITIME FORMAL SAFETY ASSESSMENT USING FUZZY Doumas N. Geogos, Nktakos V. Nñtas, ambou A. aa A ETODOOGY FOR RATING AND RANKING AZARDS IN ARITIE FORA SAFETY ASSESSENT USING FUZZY OGIC A ETODOOGY FOR RATING AND RANKING AZARDS IN ARITIE FORA SAFETY

More information

Fast Inductance Extraction of Large VLSI Circuits

Fast Inductance Extraction of Large VLSI Circuits Fast Inductance Extacton of Lage VLSI Ccuts Hemant Mahawa mahawah@cs.tamu.edu Vvek San san@cs.tamu.edu Texas A&M Unvesty College Staton, TX 83 Wepng Sh wsh@cs.tamu.edu Abstact Accuate estmaton of sgnal

More information

A New Regularized Orthogonal Local Fisher Discriminant Analysis for Image Feature Extraction

A New Regularized Orthogonal Local Fisher Discriminant Analysis for Image Feature Extraction ZHONGFENG WANG et al: A NEW REGULARIZED ORHOGONAL LOCAL FISHER DISCRIMINAN A Ne Regulazed Othogonal Local Fshe Dscmnant Analyss fo Image Featue Extacton Zhongfeng WANG, Zhan WANG 2 dept. name of oganzaton,

More information

Vol. 4, No. 6 June 2013 ISSN Journal of Emerging Trends in Computing and Information Sciences CIS Journal. All rights reserved.

Vol. 4, No. 6 June 2013 ISSN Journal of Emerging Trends in Computing and Information Sciences CIS Journal. All rights reserved. Vol 4, No 6 June 03 ISSN 079-8407 Jounal Emegng Tends n Computng and Infomaton Scences 009-03 CIS Jounal All ghts eseved http://wwwcsjounalog Optmal Powe Dstbuton System Placement to Install the RTU wth

More information

Recognition of Shapes for Object Retrieval in Image Databases by Discrete Curve Evolution and Two Consecutive Primitive Edges

Recognition of Shapes for Object Retrieval in Image Databases by Discrete Curve Evolution and Two Consecutive Primitive Edges Poceedngs of the Intenatonal MultConfeence of Engnees and Compute Scentsts 009 Vol I IMECS 009 Mach 8-0 009 Hong Kong Recognton of Shapes fo Object Reteval n Image Databases by Dscete Cuve Evoluton and

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

Collision Avoidance for Multiple Robots till Next Waypoints through Collision Free Polygons

Collision Avoidance for Multiple Robots till Next Waypoints through Collision Free Polygons Collson Avodance fo Multple Robots tll Next Wayponts though Collson Fee Polygons Satsh Peddu* K. Madhava Kshna* * Robotcs Reseach Cente, Intenatonal Insttute of Infomaton Technology, Hydeabad, Inda peddu@eseach.t.ac.n

More information

QFSRD:OrthogenesisEvolutionBasedGeneticAlgorithmforQosFitnessScopeawareRouteDiscoveryinAd Hoc Networks

QFSRD:OrthogenesisEvolutionBasedGeneticAlgorithmforQosFitnessScopeawareRouteDiscoveryinAd Hoc Networks Global Jounal of Compute Scence and Technology: E Netwo, Web & Secuty Volume 5 Issue 3 Veson.0 Yea 205 Type: Double Blnd Pee Revewed Intenatonal Reseach Jounal Publshe: Global Jounals Inc. (USA) Onlne

More information

Efficient VLSI Implementation of Modified Dual Tree Discrete Wavelet Transform Based on Lifting Scheme

Efficient VLSI Implementation of Modified Dual Tree Discrete Wavelet Transform Based on Lifting Scheme Austalan Jounal of Basc and Appled Scences, 9(7) August 05, Pages: 58-66 ISSN:99-878 Austalan Jounal of Basc and Appled Scences Jounal home page: www.ajbasweb.com Effcent VLSI Implementaton of Modfed Dual

More information

SOFT COMPUTING OPTIMIZATION TECHNIQUES FOR SOLAR PHOTOVOLTAIC ARRAYS

SOFT COMPUTING OPTIMIZATION TECHNIQUES FOR SOLAR PHOTOVOLTAIC ARRAYS ARPN Jounal of Engneeng and Appled Scences 2006-20 Asan Reseach Publshng Netwok (ARPN). All ghts eseved. SOFT COMPUTING OPTIMIZATION TECHNIQUES FOR SOLAR PHOTOVOLTAIC ARRAYS Ramapabha R. and Mathu B. L.

More information

POMDP: Introduction to Partially Observable Markov Decision Processes Hossein Kamalzadeh, Michael Hahsler

POMDP: Introduction to Partially Observable Markov Decision Processes Hossein Kamalzadeh, Michael Hahsler POMDP: Intoduction to Patially Obsevable Makov Decision Pocesses Hossein Kamalzadeh, Michael Hahsle 2019-01-02 The R package pomdp povides an inteface to pomdp-solve, a solve (witten in C) fo Patially

More information

Some Advanced SPC Tools 1. Cumulative Sum Control (Cusum) Chart For the data shown in Table 9-1, the x chart can be generated.

Some Advanced SPC Tools 1. Cumulative Sum Control (Cusum) Chart For the data shown in Table 9-1, the x chart can be generated. Some Advanced SP Tools 1. umulatve Sum ontrol (usum) hart For the data shown n Table 9-1, the x chart can be generated. However, the shft taken place at sample #21 s not apparent. 92 For ths set samples,

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

IP Network Design by Modified Branch Exchange Method

IP Network Design by Modified Branch Exchange Method Received: June 7, 207 98 IP Netwok Design by Modified Banch Method Kaiat Jaoenat Natchamol Sichumoenattana 2* Faculty of Engineeing at Kamphaeng Saen, Kasetsat Univesity, Thailand 2 Faculty of Management

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

KANBAN DEVS MODELLING, SIMULATION AND VERIFICATION

KANBAN DEVS MODELLING, SIMULATION AND VERIFICATION KANBAN DEVS MODELLING, SIMULATION AND VERIFICATION Alex Cave and Saed Nahavand School of Engneeng and Technology, Deakn Unvesty, Geelong, Vctoa 3216, Austala nahavand@deakn.edu.au ABSTRACT Kanban Contol

More information

An Entropy-Based Approach to Integrated Information Needs Assessment

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

More information

Help for Time-Resolved Analysis TRI2 version 2.4 P Barber,

Help for Time-Resolved Analysis TRI2 version 2.4 P Barber, Help for Tme-Resolved Analyss TRI2 verson 2.4 P Barber, 22.01.10 Introducton Tme-resolved Analyss (TRA) becomes avalable under the processng menu once you have loaded and selected an mage that contans

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

A Neural Network Model for Storing and Retrieving 2D Images of Rotated 3D Object Using Principal Components

A Neural Network Model for Storing and Retrieving 2D Images of Rotated 3D Object Using Principal Components A Neual Netwok Model fo Stong and Reteving 2D Images of Rotated 3D Object Using Pncipal Components Tsukasa AMANO, Shuichi KUROGI, Ayako EGUCHI, Takeshi NISHIDA, Yasuhio FUCHIKAWA Depatment of Contol Engineeng,

More information

Gravitational Shift for Beginners

Gravitational Shift for Beginners Gavitational Shift fo Beginnes This pape, which I wote in 26, fomulates the equations fo gavitational shifts fom the elativistic famewok of special elativity. Fist I deive the fomulas fo the gavitational

More information

Active Contours/Snakes

Active Contours/Snakes Actve Contours/Snakes Erkut Erdem Acknowledgement: The sldes are adapted from the sldes prepared by K. Grauman of Unversty of Texas at Austn Fttng: Edges vs. boundares Edges useful sgnal to ndcate occludng

More information

Parameter estimation for incomplete bivariate longitudinal data in clinical trials

Parameter estimation for incomplete bivariate longitudinal data in clinical trials Parameter estmaton for ncomplete bvarate longtudnal data n clncal trals Naum M. Khutoryansky Novo Nordsk Pharmaceutcals, Inc., Prnceton, NJ ABSTRACT Bvarate models are useful when analyzng longtudnal data

More information

Interactive Rendering of Deforming NURBS Surfaces

Interactive Rendering of Deforming NURBS Surfaces EUROGRAHICS 97 / D. Fellne and L. Szmay-Kalos Volume 6, (997), Numbe 3 (Guest Edtos) Inteactve Rendeng of Defomng NURBS Sufaces Fedeck W. B. L Rynson W. H. Lau Mak Geen Compute Gaphcs and Meda Laboatoy

More information

ARTICULATED MOTION ANALYSIS FROM MOTION CAPTURE DATA

ARTICULATED MOTION ANALYSIS FROM MOTION CAPTURE DATA ARICULAED MOION ANALYSIS FROM MOION CAPURE DAA J. Fayad, A. Del Bue and P. M. Q. Agua 3. ABSRAC We pesent computatonal methods to extact and model dffeent onts of a genec subect, n an automatc way. he

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

Point-Biserial Correlation Analysis of Fuzzy Attributes

Point-Biserial Correlation Analysis of Fuzzy Attributes Appl Math Inf Sci 6 No S pp 439S-444S (0 Applied Mathematics & Infomation Sciences An Intenational Jounal @ 0 NSP Natual Sciences Publishing o Point-iseial oelation Analysis of Fuzzy Attibutes Hao-En hueh

More information

User Authentication Based On Behavioral Mouse Dynamics Biometrics

User Authentication Based On Behavioral Mouse Dynamics Biometrics User Authentcaton Based On Behavoral Mouse Dynamcs Bometrcs Chee-Hyung Yoon Danel Donghyun Km Department of Computer Scence Department of Computer Scence Stanford Unversty Stanford Unversty Stanford, CA

More information

Fuzzy Membership Function Based on Structural Information of Data Fang-Zhi ZHU a, Hui-Ru WANG b, Zhi-Jian ZHOU c,*

Fuzzy Membership Function Based on Structural Information of Data Fang-Zhi ZHU a, Hui-Ru WANG b, Zhi-Jian ZHOU c,* 06 Intenatona Confeence on Sevce Scence, Technoogy and Engneeng (SSTE 06 ISBN: 978--60595-35-9 Fuzzy Membeshp Functon Based on Stuctua Infomaton of Data Fang-Zh ZHU a, Hu-Ru WANG b, Zh-Jan ZHOU c,* Coege

More information

Multi-View Data Visualization Based Big Data Analysis Using Clustering Similarity Measure

Multi-View Data Visualization Based Big Data Analysis Using Clustering Similarity Measure Mult-Vew Data Vsualzaton Based Bg Data Analyss Usng Clusteng Smlaty Measue Snvasa Rao Madala #1, V N Rajavaman #2, T.Venkata Satya Vvek #3 1,3 Reseach scholos, Depatment of Compute Scence & Engneeng, D

More information

Investigating the Performance of Naïve- Bayes Classifiers and K- Nearest Neighbor Classifiers

Investigating the Performance of Naïve- Bayes Classifiers and K- Nearest Neighbor Classifiers Journal of Convergence Informaton Technology Volume 5, Number 2, Aprl 2010 Investgatng the Performance of Naïve- Bayes Classfers and K- Nearest Neghbor Classfers Mohammed J. Islam *, Q. M. Jonathan Wu,

More information

A Robust Method for Estimating the Fundamental Matrix

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

More information

Simultaneous Position, Velocity, Attitude, Angular Rates, and Surface Parameter Estimation Using Astrometric and Photometric Observations

Simultaneous Position, Velocity, Attitude, Angular Rates, and Surface Parameter Estimation Using Astrometric and Photometric Observations Smultaneous Poston, Velocty, Atttude, Angula Rates, and Suface Paamete Estmaton Usng Astometc and Photometc Obsevatons Chales J. Wettee and C. Channng Chow Pacfc Defense Solutons, LLC Khe, Hawa, USA jac.wettee@pacfcds.com

More information

Classification / Regression Support Vector Machines

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

More information

EECS 730 Introduction to Bioinformatics Sequence Alignment. Luke Huan Electrical Engineering and Computer Science

EECS 730 Introduction to Bioinformatics Sequence Alignment. Luke Huan Electrical Engineering and Computer Science EECS 730 Introducton to Bonformatcs Sequence Algnment Luke Huan Electrcal Engneerng and Computer Scence http://people.eecs.ku.edu/~huan/ HMM Π s a set of states Transton Probabltes a kl Pr( l 1 k Probablty

More information

Advanced Algorithms for Location-Based Smart Mobile Augmented Reality Applications

Advanced Algorithms for Location-Based Smart Mobile Augmented Reality Applications Avalable onlne at www.scencedect.com Poceda Compute Scence 00 (2016) 000 000 www.elseve.com/locate/poceda The 13th Intenatonal Confeence on Moble Systems and Pevasve Computng (MobSPC 2016) Advanced Algothms

More information

Generalization of Lambert s Reflectance Model

Generalization of Lambert s Reflectance Model Genealzaton of Lambet s Reflectance Model Mchael Oen and Shee K. Naya Depatment of Compute Scence, Columba Unvesty Ne Yok, NY 007 Abstact Lambet s model fo body eflecton s dely used n compute gaphcs. It

More information