Performance Evaluation of Mutation / Non- Mutation Based Classification With Missing Data

Size: px
Start display at page:

Download "Performance Evaluation of Mutation / Non- Mutation Based Classification With Missing Data"

Transcription

1 Performace Evaluatio of Mutatio / No- Mutatio Based Classificatio With Missig Data N.C. Viod Research Scholar, Maomaiam Sudaraar Uiversity, Tiruelveli, Tamil Nadu, Idia Dr. M. Puithavalli Research Supervisor, Maomaiam Sudaraar Uiversity, Tiruelveli, Tamil Nadu, Idia Abstract A commo problem ecoutered by may data miig techiques is the missig data. A missig data is defied as a attribute or feature i a dataset which has o associated data value. Correct treatmet of these data is crucial, as they have a egative impact o the iterpretatio ad result of data miig processes. Missig value hadlig techiques ca be grouped ito four categories, amely, complete case aalysis, Imputatio methods, maximum likelihood methods ad machie learig methods. Out of these imputatio methods are the widely used solutio for hadlig missig values. However, there are situatios whe imputatio methods might ot work correctly. This study studies ad aalyzes the performace of two algorithms, oe imputatio based ad aother without imputatio based classificatio o missig data. Keywords-Missig Values, Imputatio, No-imputatio, Classificatio with missig data. 1. INTRODUCTION The process of extractig hidde kowledge from databases is called data miig. The data miig methods have great potetial i predictig future treds ad behaviours, allowig busiesses to make proactive, kowledge-drive decisios. It is a multidiscipliary field, drawig work from areas icludig database techology, artificial itelligece, machie learig, eural etworks, statistics, patter recogitio, kowledgebased systems, kowledge acquisitio, iformatio retrieval, high-performace computig ad data visualizatio [26]. The era of iformatio explosio is curretly ecouterig a sudde upsurge i both size ad umber of databases. This icrease far exceeds the ability of humas to aalyze ad extract kowledge from these databases. A commo problem ecoutered by may database ad data miig applicatios is missig data. A missig data is defied as a attribute / feature i a dataset which has o associated data value stored to it. Correct treatmet of missig data is essetial as they have a egative impact, if utreated, o the iterpretatio result. The situatio whe the missig rate is more tha 15%, is a clear idicatio that the applicatio i questio should implemet some sophisticated tools to hadle them correctly. Several studies have focused o developig algorithms that deal with missig data ([22], [24]). But most of these have some drawbacks whe applied to classificatio tasks. Classificatio is the process of dividig a dataset ito mutually exclusive groups such that the members of each group are as close as possible to oe aother, ad differet groups are as far as possible from oe aother, where distace is measured with respect to specific variable(s) that are used for predictio. Mishadlig missig values durig classificatio may produce erroeous classificatio results ([1], [5], [11], [25]). Geerally, the methods that deal with missig values ca be grouped ito four differet categories, which are based o the techique used to solve the problem. The first is Complete case aalysis. These algorithms igore the records with missig values ad the aalysis is based o the complete data ad ca oly work well with large sized database [24]). The demerit of this approach is that the umber of data discarded has a direct impact o the efficiecy of classificatio. The secod approach is Imputatio method [32]. Imputatio is a class of procedures that aims to fill i the missig values with estimated oes. The process uses kow relatioship that exists amog the complete values i the dataset to estimate the missig data. After fillig the missig data with the estimated value, the classifier lears the modified data set. The third approach is Maximum likelihood method [27]. I this method, the iput data distributio is modeled by usig either Expectatio-Maximizatio (EM) algorithm or by usig the variats of EM to hadle the missig values. This is followed by the classificatio task performed by meas of the Bayes rule [17]. The fourth approach is the machie learig methods, which deals with missig values without ay imputatio ad uses techiques such as decisio trees [30] ad fuzzy eural etwork [15]. Most of the methods i this category require complete iput data matrix for processig ad requires cosiderable efforts. Out of this imputatio methods have bee more widely used. Methods like mea ad mode imputatio [2], hot deck imputatio [24], predictio models [32], artificial eural etwork imputatio [13], recurret eural ISSN : Vol. 5 No. 02 Feb

2 etwork imputatio [18], auto-associative imputatio [20], k-earest eighbour imputatio [19] ad Self- Orgaizig Map imputatio [23] have reported to be successful i hadlig missig values. Recetly, Garcia-Laecia et al. [16] proposed a solutio to hadle missig data usig the popular k-earest eighbour (KNN) imputatio process. This procedure uses a feature-weighted distace metric based o mutual iformatio (MI) to estimate the missig value to improve classificatio task. MI is defied as a measure of depedece betwee radom variables ad has bee used i the past as relevace measure i several selectio algorithms ([21], [31]). This model is referred to as GL-Model i this paper. The primary objective of this model is to provide a solutio to missig data problem by focusig imputatio solutios based o the K Nearest Neighbours (KNN) algorithm ([4], [5], [33]). The KNN algorithm is oe of the most popular approaches for solvig icomplete data problems. This algorithm selects K closest observatios (eighbours) from a icomplete dataset, usig a distace metric. The selected observatios preset kow values o the features to be imputed. The weighted average of these values is calculated ad is used as a estimate for each icomplete feature value. Next, to improve the classificatio performace, a ovel KNN imputatio procedure that uses the feature-weighted distace metric is used. The iput attribute relevace to Mutual Iformatio (MI) for classificatio is cosidered accordig to the distace metric. The cocept of Mutatio might ot work i situatio like follows. Cosider a set of diagostic data of patiets ad ormal people collected from differet hospitals distributed geographically. The missig rate i this type of situatio is very high ad techiques that are based o feature deletio ad imputatio strategy are ot suitable. The reaso behid this is that the missig values may distribute differet from the existig oes ad these techiques either brig i bias or report iaccurate results. For example, the diagosis data of patiets ad ormal people have differet distributios. To solve this problem, Qu et al. [29], proposed a two stage model that avoids mutatio ad deletio. I phase I, the dataset is divided ito disjoit subsets based o the attributes with missig values. I phase II, each subset is used to trai appropriate classificatio algorithms respectively i parallel. This model is referred as QU-Model i this paper. The mai objective of this paper is to compare the performace of the GL-Model ad QU-Model with differet datasets, whose missig mechaism is Missig At Radom (MAR). MAR is the probability of the observed missig patter, give the observed ad uobserved data, does ot deped o the values of the uobserved data. This mechaism is commo i practice ad is geerally cosidered as the default type of missig data. The paper is orgaized as follows. Sectio II explais the GL-Model, while Sectio III explais QU-Model. Sectio IV presets the experimetal results ad performs the compariso. Sectio V presets the summary of the work. II. GL-MODEL The KNN method selects K patters from the full dataset i such a way that they miimize a distace measure. After fidig the k earest eighbours, a replacemet value to substitute the missig attribute must be estimated. The method used for estimatio depeds o the type of data; the mode ca be used for discrete data ad the mea for cotiuous data. A improved alterative is to weight the cotributio of each eighbour accordig to their distace to the icomplete patter whose values will be imputed, givig greater cotributio to close eighbours [21]. A advatage over mea/mode imputatio ad simple hot deck method (i fact, KNNimpute with K = 1) is that the replacemet values are oly iflueced by the most similar cases rather tha by all cases or the most similar oe, respectively. The mai drawback of this approach is that wheever KNNimpute looks for the most similar patters, the algorithm looks for through all traiig data set (i the complete data portio), which implies a high computatioal cost. Oe importat step i KNNImpute is the selectio of distace metric. The distace metric used i the GL-model is the heterogeeous Euclidea-overlap metric (HEOM) [5]. HEOM is described as follows: The distace betwee two iput vectors x a ad x b is deoted as d(x a, x b ) ad is calculated usig Equatio (1). 2 d (x a, x b ) = d j (x aj, (1) j= 1 where dj(x aj, x bj ) is the distace betwee x a, x b o its jth attribute ad is give by Equatio (2). 1 (1 maj)(1 m bj) = 0 d j (x aj, = d 0 (x aj, x j is qualitative (2) d N (x aj, x j is quatitative Whe the iput values are ukow, the it is treated missig data ad a distace value of 1 is retured i these situatios. The overlap fuctio d 0 is assiged a value 0 if the qualitative features are same other wise it is assiged a value 1 (Equatio 3) ad d N is the rage ormalized differece distace fuctio ad is give by Equatio (4). ISSN : Vol. 5 No. 02 Feb

3 0 if x aj = x bj d 0 (x aj, = (3) 1 if x aj x bj d N (x aj, x bj x aj x bj ) = (4) max(x ) mi(x ) j j where max(x j ) ad mi(x j ) are the maximum ad miimum values observed i the N traiig istaces for the jth attribute. To classify a ulabeled patter x, the distaces from x to the labeled istaces are computed, its K earest eighbors are idetified, ad the class labels of these earest eighbors are the used to determie the class label of x. Accordig to the votig KNN rule, x is assiged to the class represeted by a majority of its K earest eighbours [9]. I the stadard KNN algorithm, the K eighbours are implicitly assumed to have equal weight i decisio, regardless of their distaces to the patter to be classified. Some approaches have bee proposed based o assigig differet weights to the K eighbours accordig to their distaces to x, with closer istaces havig greater weights. I GL-model, the weighted KNN algorithm is referred as KNNclassify. Followig the distaceweighted rule proposed by Dudai [12], KNNclassify is implemeted usig the same weightig procedure described i [7]. Thus, a weight a k is assiged to each earest eighbour v k of x, with k = 1, 2,..., K. The earest eighbour receives a weight of 1, the furthest eighbour a weight of 0, ad the remaiig eighbours are scaled liearly betwee 0 ad 1. A ulabeled patter is assiged to the class producig the highest summed weight amog its referece eighbours. I the GL-model, the approach for missig data imputatio ad classificatio is a modified KNNImpute. The KNNImpute method is modified because, the covetioal method s learig process is ot orieted to provide a appropriate imputed dataset for solvig classificatio task. The modified method uses a effective procedure where the eighbourhood is selected by cosiderig the iput attribute relevace for classificatio. For each icomplete patter, its selected K eighbours are used to provide imputed values which ca make the classifier desig easier, ad thus, the classificatio accuracy is icreased. This approach uses a feature-weighted distace metric based o MI, which is a good idicator of depedece betwee radom variables. Feature selectio algorithms, i geeral, assig biary weights to features, that is, a weight equal to 1 for selected relevat attributes; a value of 0 for irrelevat features. This method has the advatage of reducig iput dimesioality ad computatio complexity. I GL-model, the feature-weighted procedure assigs oe weight per feature accordig to the MI estimate betwee each feature ad the target class variable. Details of MI estimatio ca be foud i [21]. For classificatio, the MI measures the amout of iformatio cotaied i a iput feature for predictig the target class variable [10]. A high MI betwee a iput feature ad the target meas that this feature is relevat, regardless of the classificatio algorithm. Otherwise, whe the shared iformatio betwee both variables is small, the iput feature is irrelevat for the classificatio task. I GL-model, the MI cocept to weight the iput features distaces i (1) accordig to their relevace for classificatio is sued. The method assigs a weight λ j (Equatio 5) to each jth iput feature accordig to the amout of iformatio that this attribute cotais about the target class variable. The scalig factors λ j is computed heuristically, as give by Weiberger et al. [34]. The higher λ j the more relevat X j is for classificatio. I(X j,c) λ j = (5) I(X,C) f = 1 j Accordig to the MI cocept, the feature-weighted distace metric (Equatio) betwee two iput vectors x a ad x b is computed usig Equatio (6). 2 d j (xa, xb) = λ jd j(xaj, xbj) (6) j= 1 where d j is the distace defied i Equatio (2). Whe this MI-weighted distace metric is implemeted i KNNImpute, the modified MI-KNNImpute is formed. The advatage of this method is that it selects the K- earest cases by cosiderig the iput attribute relevace to the target class, thus addig useful iformatio about classificatio task durig imputatio stage ad providig missig data estimatio orieted to solve the classificatio task. The MI betwee a attribute ad target class variable is estimated by the Parze widow method [21]. The MI-KNN classify assigs a weight β k to the k-th earest eighbour usig Equatio (7). d (v, x) d (v, x) ( x) I k β I k k = = 1 whe di(vk, x) = di(vk, x) (7) d (v, x) d (v, x) I k I 1 ISSN : Vol. 5 No. 02 Feb

4 where d I (.) is the distace measure based o MI cocept defied by (6) ad V x is the set of K earest eighbours of x arraged i icreasig order of distace. Durig classificatio, x is assiged to the class for which the weights β k of the represetatives amog the K C earest eighbours sum to the largest value. This is performed by cosiderig the relevacy betwee the iput features of the target class variable. III. QU-MODEL The QU-Model proposes a two-stage algorithm for situatios where imputatio methods are ot applicable. I the first stage, the origial dataset is segmeted ito disjoit sets accordig to the iformatio gai of the features with missig values (Equatio 8). I the secod stage, the data i each set obtaied from stage 1 is used to trai the classifier. S = v Gai (S, F) E(S) E(Sv ) (8) v {FN,FA } S where, S is the origial dataset, F is the feature set with missig values, F N ad F A represets the subsets of F depedig o whether the value of F is missig (N) or available (A). E(S) is the iformatio etropy of S (Equatio 9) ad is umber of categories ad p i is the probability that the samples i dataset S belog to the ith class. E (S) = pi log2(pi ) (9) i= ` The method used for traiig durig stage I is give i Figure Give a traiig dataset D, calculate the iformatio gai of each feature with missig data. 2. Select the feature F with the highest iformatio gai. Split D ito two subsets DN ad DA based o the values of F (missig or available). I other words, DA cosists of samples where the values of F are available ad DN cotais the rest samples. Sice the values of F are all missig i DN, F is removed ad the dimesio of DN is reduced by oe. A biary tree T is built i which D is the root ode ad the child odes are DA ad DN. 3. For datasets DA ad DN, repeat step 1 ad step 2 util there is o missig values i each child ode. Figure 1 : Traiig Stage I After Stage 1, the dimesios of most subsets are lower tha that of the origial dataset D, which may help improve the performace of the classifiers used i Phase II. I stage 2, the origial dataset D is split ito several subsets Di. A separate classifier Ci is traied o each subset respectively with the classificatio algorithm give by Libsvm [8]. Whe the umber of features with missig values is high, a miimum missig rate is applied to avoid geeratig too may subsets (the umber of samples i each subset eeds to be maitaied at a reasoable level for classificatio). Testig is performed as below: I Stage 1, for each ukow sample X to be classified fid the correspodig subset Di i the tree T geerated above. I Stage 2, Apply Ci (traied o Di) to assig a class label to X. IV. EXPERIMENTAL RESULTS The experimets were coducted with the objective of testig the GL ad QU models i their efficiecy i solvig classificatio task with icomplete ad complete data. To compare the results, the KNNImute approach, GL-model ad QU-model was tested with the C4.5 classifier. The efficiecy of the algorithm was aalyzed usig the classificatio accuracy ad time take to classify. The real world datasets used are Abaloe, Credit Approval ad Aealig datasets ad were obtaied from UCI repository [28]. The 10-fold cross validatio was used [6] was used i all experimets. Abaloe [3] is a multi-class dataset ad for the sake of clarity, a ew twoclass dataset was created by selectig a subset of samples ad mergig differet classes. The resultig dataset was a relatively balaced oe cotaiig 689 class 1 samples ad 709 class 2 samples. There was a feature sex i the dataset with three possible values: M, F ad I ad all sex values marked with I were regarded as missig data. As a result, the icomplete dataset had 578 samples with missig values. Credit card dataset ( br:443/pakdd2009) cotai customer iformatio ad credit levels (2-class problem). A subset of the origial data was selected based o shops (ID_Shop) ad the the values of the feature Moths_I_Job were discarded for some selected shops to create a icomplete dataset. This was to simulate the real world sceario where certai data from specific shops were missig. Fially, there were samples ad 3171 of them had missig values. The aealig dataset [14] is a multi-class dataset havig 798 istaces havig 38 attributes. I the dataset, the preset of the character - i ay of the attribute deotes missig values, ISSN : Vol. 5 No. 02 Feb

5 resultig i a icomplete dataset. A subset of the origial data was selected based o family type (family), totalig to 70 cases of missig data. Table I shows the classificatio accuracy for the three datasets selected. The experimets were coducted with subset selectio ad full dataset. Table I. Classificatio Accuracy Dataset C4.5 GL Model QU Model Abaloe CreditApproval(subset) CreditApproval (full dataset) Aealig (subset) Aealig (full dataset) While takig the executio time ito cosideratio, the QU-algorithm was fast i producig classificatio results tha GL-Algorithm. But it could be see that the C4.5 algorithm without missig value hadler was the quickest amog all the three. The reaso is that the additioal algorithm requires a egligible time to hadle the icomplete data durig classificatio. But this differece is very miimal (o average 0.09% with GL-Model ad 0.04% with QU-Model) ad the performace of accuracy obtaied by the GL ad QU Models outweighs this fact. Durig experimetatio, while the full dataset with missig data was cosidered ad it could be see that the GL-Model produced better classificatio accuracy. While a subset of the dataset was cosidered, i Credit Approval dataset, the sex attribute took oly two omial values, the mutatio algorithm s performace degraded ad QU-Model raked better tha GL-model. I all the cases, classifier without missig value hadler produced poor result. Figure 2 shows the time complexity of the proposed algorithms Time (Secods) Abaloe CreditApproval (subset) CreditApproval (full dataset) Aealig (subset) Aealig (full dataset) C4.5 GL-Model QU-Model Figure 2 : Performace based o Classificatio Time V. CONCLUSION The problem of missig data i databases creates problems i almost all steps of data miig. I this paper, a method based o KNN combied with imputatio ad a method which uses a two stage approach without imputatio are aalyzed. The first method uses a feature-weighted distace metric combied with KNN classifier to hadle icomplete data for classificatio. The secod method first divides the dataset ito disjoit subsets accordig to the attributes with missig values. Usig these subsets the classificatio process is performed. Several experimets were coducted ad the results prove that while both the models perform well with missig data, whe the attribute value is biary, the QU-model performs better tha GL-model. It would be worthwhile to pursue research ito developig hybrid model that combies the GL ad QU-models by applyig the imputatio method to the subsets i the first stage ad the perform classificatio ad to study ad aalyze their effect o classificatio. REFERENCES [1] Acua, E. ad Rodriguez, C. (2004) The treatmet of missig values ad its effect i the classifier accuracy, i: D. Baks, L. House, F.R. McMorris, P. Arabie, W. Gaul (Eds.), Classificatio, Clusterig ad Data Miig Applicatios, Spriger, Berli, Pp [2] Alliso, P.D. (2001) Missig data, Sage Uiversity Papers Series o Quatitative Applicatios i the Social Scieces, Thousad Oaks, Califoria, USA. [3] Asucio, A. ad Newma, D.J. (2007) UCI Machie Learig Repository, Uiversity of Califoria. [4] Batista, G.E. ad Moard, M.C. (2002) A study of k-earest eighbour as a imputatio method, i: Secod Iteratioal Coferece o Hybrid Itelliget Systems, vol. 87, Satiago, Chile, 2002, pp [5] Batista, G.E. ad Moard, M.C. (2003) A aalysis of four missig data treatmet methods for supervised learig, Applied Artificial Itelligece, Vol. 17, No.5 6, Pp [6] Bishop, C.M. (1995) Neural Networks for Patter Recogitio, Oxford Uiversity Press, Oxford, UK. [7] Brow, J.G. (2002) Usig a multiple imputatio techique to merge data sets, Applied Ecoomics Letters, Vol. 9, No. 5, Pp [8] Chag, C. ad Li, C. (2001) LIBSVM: a library for support vector machies. ISSN : Vol. 5 No. 02 Feb

6 [9] Cover, T.M. ad Hart, P.E. (1967) Nearest eighbor patter classificatio, IEEE Trasactios o Iformatio Theory, Vol. 13, No.1, Pp [10] Cover, T.M. ad Thomas, J.A. (1991) Elemets of Iformatio Theory, Wiley-Itersciece, New York. [11] Duda, R.O., Hart, P.E. ad Stork, D.G. (2000) Patter Classificatio, Wiley-Itersciece, New York. [12] Dudai, S.A. (1976) The distace-weighted k-earest-eighbor rule, IEEE Trasactios o Systems, Ma ad Cyberetics, Vol. 6, No.4, [13] Fracois, D., Rossi, F., Wertz, V. ad Verleyse, M. (2007) Resamplig methods for parameter-free ad robust feature selectio with mutual iformatio, Neurocomputig, Vol. 70, Issue 7 9, Pp [14] Frak, A. ad Asucio, A. (2010) UCI Machie Learig Repository, Irvie, CA: Uiversity of Califoria, School of Iformatio ad Computer Sciece. [15] Gabrys, B. (2002) Neuro-fuzzy approach to processig iputs with missig values i patter recogitio problems, Iteratioal Joural of Approximate Reasoig, Vol. 30, No.3, Pp [16] Garcia-Laecia, P.J., Sacho-Gomez, J., Figueiras-Vidal, A.R.C ad Verleyse, M. (2009) K earest eighbours with mutual iformatio for simultaeous classificatio ad missig data imputatio, Joural Neurocomputig, Elsevier Publicatios, Vol. 72. Issue 7-9, Pp [17] Ghahramai, Z. ad Jorda, M.I. (1994) Supervised learig from icomplete data via a EM approach, J.D. Cowa, G. Tesauro, J. Alspector (Eds.), Advaces i NIPS, Vol. 6, Morga Kaufma Publishers, Ic., Los Altos, CA, Pp [18] Hammer, B. ad Villma, T. (2002) Geeralized relevace learig vector quatizatio, Neural Networks, Vol. 15, Issues 8 9, Pp [19] Hechebichler, K. ad Schliep, K. (2007) Weighted k-earest-eighbor techiques ad ordial classificatio. Techical Report, Ludwig Maximilias Uiversity Muich. [20] Jerez, J.M., Molia, I., Subirats, J.L. ad Fraco, L. (2006) Missig data imputatio i breast cacer progosis, i: BioMed 06: Proceedigs of the 24th IASTED Itera- tioal Coferece o Biomedical Egieerig, ACTA Press, Aaheim, CA, USA, Pp [21] Kwak, N. ad Choi, C.H. (2002) Iput feature selectio by mutual iformatio based o Parze widow, IEEE Trasactios o Patter Aalysis ad Machie Itelligece, Vol. 24, No.12, Pp [22] Lall, U. ad Sharma, A., (1996) A earest-eighbor bootstrap for resamplig hydrologic time series, Water Resource. Res., Vol.32, Pp [23] Little, R.J.A. (1999) Methods for hadlig missig values i cliical trials, Joural of Rheumatology, Vol. 26, No. 8, Pp [24] Little, R.J.A. ad Rubi, D.B. (2002) Statistical Aalysis with Missig Data, 2d Editio, Joh Wiley ad Sos, New York. [25] Markey, M.K. ad Patel, A. (2004) Impact of missig data i traiig artificial eural etworks for computer-aided diagosis, i: Iteratioal Coferece o Machie Learig ad Applicatios, Pp [26] Marti, T. (2003) A day i the life of a Data Mier, Bulleti of the Iteratioal Statistical Istitute, 54th Sessio, Vol. LX, Ivited Papers, August 2003, Berli, Germay. Pp [27] McLachla, G.J. ad Krisha, T. (1997) The EM Algorithm ad Extesios, Wiley, New York. [28] Newma, D.J., Hettich, S., Blake, C.L. ad Merz, C.J. (1998) UCI repository of machie learig databases. [29] Qu, X., Yua, B. ad Liu, W. (2009) A Novel Two-Phase Method for the Classificatio of Icomplete Data, Iteratioal Coferece o Iformatio Maagemet, Iovatio Maagemet ad Idustrial Egieerig, Pp [30] Quila, J.R. (1993) C4.5: Programs for Machie Learig, Morga Kaufma, Los Altos, CA. [31] Rossi, F., Ledasse, A., Fracois, D., Wertz, V. ad Verleyse, M. (2006) Mutual iformatio for the selectio of relevat variables i spectrometric oliear modelig, Chemometrics ad Itelliget Laboratory Systems, Vol. 80, Pp [32] Schafer, J.L. (1997) Aalysis of Icomplete Multivariate Data, Chapma & Hall, Florida, USA. [33] Troyaskaya, O., Cator, M., Alter, O., Sherlock, G., Brow, P., Botstei, D., Tibshirai, R., Hastie, T. ad Altma, R. (2001) Missig value estimatio methods for DNA microarrays, Bioiformatics, Vol. 17, No.6, Pp [34] Weiberger, K., Blitzer, J. ad Saul, L. (2006) Distace metric learig for large margi earest eighbor classificatio, i: Y. Weiss, B. Scholkopf, J. Platt (Eds.), Advaces i NIPS 18, MIT Press, Cambridge, MA, Pp ISSN : Vol. 5 No. 02 Feb

3D Model Retrieval Method Based on Sample Prediction

3D Model Retrieval Method Based on Sample Prediction 20 Iteratioal Coferece o Computer Commuicatio ad Maagemet Proc.of CSIT vol.5 (20) (20) IACSIT Press, Sigapore 3D Model Retrieval Method Based o Sample Predictio Qigche Zhag, Ya Tag* School of Computer

More information

Ones Assignment Method for Solving Traveling Salesman Problem

Ones Assignment Method for Solving Traveling Salesman Problem Joural of mathematics ad computer sciece 0 (0), 58-65 Oes Assigmet Method for Solvig Travelig Salesma Problem Hadi Basirzadeh Departmet of Mathematics, Shahid Chamra Uiversity, Ahvaz, Ira Article history:

More information

Euclidean Distance Based Feature Selection for Fault Detection Prediction Model in Semiconductor Manufacturing Process

Euclidean Distance Based Feature Selection for Fault Detection Prediction Model in Semiconductor Manufacturing Process Vol.133 (Iformatio Techology ad Computer Sciece 016), pp.85-89 http://dx.doi.org/10.1457/astl.016. Euclidea Distace Based Feature Selectio for Fault Detectio Predictio Model i Semicoductor Maufacturig

More information

Pruning and Summarizing the Discovered Time Series Association Rules from Mechanical Sensor Data Qing YANG1,a,*, Shao-Yu WANG1,b, Ting-Ting ZHANG2,c

Pruning and Summarizing the Discovered Time Series Association Rules from Mechanical Sensor Data Qing YANG1,a,*, Shao-Yu WANG1,b, Ting-Ting ZHANG2,c Advaces i Egieerig Research (AER), volume 131 3rd Aual Iteratioal Coferece o Electroics, Electrical Egieerig ad Iformatio Sciece (EEEIS 2017) Pruig ad Summarizig the Discovered Time Series Associatio Rules

More information

Investigating methods for improving Bagged k-nn classifiers

Investigating methods for improving Bagged k-nn classifiers Ivestigatig methods for improvig Bagged k-nn classifiers Fuad M. Alkoot Telecommuicatio & Navigatio Istitute, P.A.A.E.T. P.O.Box 4575, Alsalmia, 22046 Kuwait Abstract- We experimet with baggig knn classifiers

More information

Mining from Quantitative Data with Linguistic Minimum Supports and Confidences

Mining from Quantitative Data with Linguistic Minimum Supports and Confidences Miig from Quatitative Data with Liguistic Miimum Supports ad Cofideces Tzug-Pei Hog, Mig-Jer Chiag ad Shyue-Liag Wag Departmet of Electrical Egieerig Natioal Uiversity of Kaohsiug Kaohsiug, 8, Taiwa, R.O.C.

More information

A SOFTWARE MODEL FOR THE MULTILAYER PERCEPTRON

A SOFTWARE MODEL FOR THE MULTILAYER PERCEPTRON A SOFTWARE MODEL FOR THE MULTILAYER PERCEPTRON Roberto Lopez ad Eugeio Oñate Iteratioal Ceter for Numerical Methods i Egieerig (CIMNE) Edificio C1, Gra Capitá s/, 08034 Barceloa, Spai ABSTRACT I this work

More information

Administrative UNSUPERVISED LEARNING. Unsupervised learning. Supervised learning 11/25/13. Final project. No office hours today

Administrative UNSUPERVISED LEARNING. Unsupervised learning. Supervised learning 11/25/13. Final project. No office hours today Admiistrative Fial project No office hours today UNSUPERVISED LEARNING David Kauchak CS 451 Fall 2013 Supervised learig Usupervised learig label label 1 label 3 model/ predictor label 4 label 5 Supervised

More information

arxiv: v2 [cs.ds] 24 Mar 2018

arxiv: v2 [cs.ds] 24 Mar 2018 Similar Elemets ad Metric Labelig o Complete Graphs arxiv:1803.08037v [cs.ds] 4 Mar 018 Pedro F. Felzeszwalb Brow Uiversity Providece, RI, USA pff@brow.edu March 8, 018 We cosider a problem that ivolves

More information

The isoperimetric problem on the hypercube

The isoperimetric problem on the hypercube The isoperimetric problem o the hypercube Prepared by: Steve Butler November 2, 2005 1 The isoperimetric problem We will cosider the -dimesioal hypercube Q Recall that the hypercube Q is a graph whose

More information

Clustering and Classifying Diabetic Data Sets Using K-Means Algorithm

Clustering and Classifying Diabetic Data Sets Using K-Means Algorithm Article ca be accessed olie at http://www.publishigidia.com Clusterig ad Classifyig Diabetic Data Sets Usig K-Meas Algorithm M. Kothaiayaki*, P. Thagaraj** Abstract The k-meas algorithm is well kow for

More information

A new algorithm to build feed forward neural networks.

A new algorithm to build feed forward neural networks. A ew algorithm to build feed forward eural etworks. Amit Thombre Cetre of Excellece, Software Techologies ad Kowledge Maagemet, Tech Mahidra, Pue, Idia Abstract The paper presets a ew algorithm to build

More information

Fundamentals of Media Processing. Shin'ichi Satoh Kazuya Kodama Hiroshi Mo Duy-Dinh Le

Fundamentals of Media Processing. Shin'ichi Satoh Kazuya Kodama Hiroshi Mo Duy-Dinh Le Fudametals of Media Processig Shi'ichi Satoh Kazuya Kodama Hiroshi Mo Duy-Dih Le Today's topics Noparametric Methods Parze Widow k-nearest Neighbor Estimatio Clusterig Techiques k-meas Agglomerative Hierarchical

More information

Our second algorithm. Comp 135 Machine Learning Computer Science Tufts University. Decision Trees. Decision Trees. Decision Trees.

Our second algorithm. Comp 135 Machine Learning Computer Science Tufts University. Decision Trees. Decision Trees. Decision Trees. Comp 135 Machie Learig Computer Sciece Tufts Uiversity Fall 2017 Roi Khardo Some of these slides were adapted from previous slides by Carla Brodley Our secod algorithm Let s look at a simple dataset for

More information

New HSL Distance Based Colour Clustering Algorithm

New HSL Distance Based Colour Clustering Algorithm The 4th Midwest Artificial Itelligece ad Cogitive Scieces Coferece (MAICS 03 pp 85-9 New Albay Idiaa USA April 3-4 03 New HSL Distace Based Colour Clusterig Algorithm Vasile Patrascu Departemet of Iformatics

More information

Criterion in selecting the clustering algorithm in Radial Basis Functional Link Nets

Criterion in selecting the clustering algorithm in Radial Basis Functional Link Nets WSEAS TRANSACTIONS o SYSTEMS Ag Sau Loog, Og Hog Choo, Low Heg Chi Criterio i selectig the clusterig algorithm i Radial Basis Fuctioal Lik Nets ANG SAU LOONG 1, ONG HONG CHOON 2 & LOW HENG CHIN 3 Departmet

More information

HADOOP: A NEW APPROACH FOR DOCUMENT CLUSTERING

HADOOP: A NEW APPROACH FOR DOCUMENT CLUSTERING Y.K. Patil* Iteratioal Joural of Advaced Research i ISSN: 2278-6244 IT ad Egieerig Impact Factor: 4.54 HADOOP: A NEW APPROACH FOR DOCUMENT CLUSTERING Prof. V.S. Nadedkar** Abstract: Documet clusterig is

More information

Study on effective detection method for specific data of large database LI Jin-feng

Study on effective detection method for specific data of large database LI Jin-feng Iteratioal Coferece o Automatio, Mechaical Cotrol ad Computatioal Egieerig (AMCCE 205) Study o effective detectio method for specific data of large database LI Ji-feg (Vocatioal College of DogYig, Shadog

More information

Bayesian Network Structure Learning from Attribute Uncertain Data

Bayesian Network Structure Learning from Attribute Uncertain Data Bayesia Network Structure Learig from Attribute Ucertai Data Wetig Sog 1,2, Jeffrey Xu Yu 3, Hog Cheg 3, Hogya Liu 4, Ju He 1,2,*, ad Xiaoyog Du 1,2 1 Key Labs of Data Egieerig ad Kowledge Egieerig, Miistry

More information

An Improved Shuffled Frog-Leaping Algorithm for Knapsack Problem

An Improved Shuffled Frog-Leaping Algorithm for Knapsack Problem A Improved Shuffled Frog-Leapig Algorithm for Kapsack Problem Zhoufag Li, Ya Zhou, ad Peg Cheg School of Iformatio Sciece ad Egieerig Hea Uiversity of Techology ZhegZhou, Chia lzhf1978@126.com Abstract.

More information

A Wrapper-Based Combined Recursive Orthogonal Array and Support Vector Machine for Classification and Feature Selection

A Wrapper-Based Combined Recursive Orthogonal Array and Support Vector Machine for Classification and Feature Selection Moder Applied Sciece; Vol. 8, No. ; 24 ISSN 93-844 E-ISSN 93-852 Published by Caadia Ceter of Sciece ad Educatio A Wrapper-Based Combied Recursive Orthogoal Array ad Support Vector Machie for Classificatio

More information

Image Segmentation EEE 508

Image Segmentation EEE 508 Image Segmetatio Objective: to determie (etract) object boudaries. It is a process of partitioig a image ito distict regios by groupig together eighborig piels based o some predefied similarity criterio.

More information

Fuzzy Rule Selection by Data Mining Criteria and Genetic Algorithms

Fuzzy Rule Selection by Data Mining Criteria and Genetic Algorithms Fuzzy Rule Selectio by Data Miig Criteria ad Geetic Algorithms Hisao Ishibuchi Dept. of Idustrial Egieerig Osaka Prefecture Uiversity 1-1 Gakue-cho, Sakai, Osaka 599-8531, JAPAN E-mail: hisaoi@ie.osakafu-u.ac.jp

More information

EMPIRICAL ANALYSIS OF FAULT PREDICATION TECHNIQUES FOR IMPROVING SOFTWARE PROCESS CONTROL

EMPIRICAL ANALYSIS OF FAULT PREDICATION TECHNIQUES FOR IMPROVING SOFTWARE PROCESS CONTROL Iteratioal Joural of Iformatio Techology ad Kowledge Maagemet July-December 2012, Volume 5, No. 2, pp. 371-375 EMPIRICAL ANALYSIS OF FAULT PREDICATION TECHNIQUES FOR IMPROVING SOFTWARE PROCESS CONTROL

More information

BASED ON ITERATIVE ERROR-CORRECTION

BASED ON ITERATIVE ERROR-CORRECTION A COHPARISO OF CRYPTAALYTIC PRICIPLES BASED O ITERATIVE ERROR-CORRECTIO Miodrag J. MihaljeviC ad Jova Dj. GoliC Istitute of Applied Mathematics ad Electroics. Belgrade School of Electrical Egieerig. Uiversity

More information

Neural Networks A Model of Boolean Functions

Neural Networks A Model of Boolean Functions Neural Networks A Model of Boolea Fuctios Berd Steibach, Roma Kohut Freiberg Uiversity of Miig ad Techology Istitute of Computer Sciece D-09596 Freiberg, Germay e-mails: steib@iformatik.tu-freiberg.de

More information

A Novel Feature Extraction Algorithm for Haar Local Binary Pattern Texture Based on Human Vision System

A Novel Feature Extraction Algorithm for Haar Local Binary Pattern Texture Based on Human Vision System A Novel Feature Extractio Algorithm for Haar Local Biary Patter Texture Based o Huma Visio System Liu Tao 1,* 1 Departmet of Electroic Egieerig Shaaxi Eergy Istitute Xiayag, Shaaxi, Chia Abstract The locality

More information

DATA MINING II - 1DL460

DATA MINING II - 1DL460 DATA MINING II - 1DL460 Sprig 2017 A secod course i data miig http://www.it.uu.se/edu/course/homepage/ifoutv2/vt17/ Kjell Orsbor Uppsala Database Laboratory Departmet of Iformatio Techology, Uppsala Uiversity,

More information

x x 2 x Iput layer = quatity of classificatio mode X T = traspositio matrix The core of such coditioal probability estimatig method is calculatig the

x x 2 x Iput layer = quatity of classificatio mode X T = traspositio matrix The core of such coditioal probability estimatig method is calculatig the COMPARATIVE RESEARCHES ON PROBABILISTIC NEURAL NETWORKS AND MULTI-LAYER PERCEPTRON NETWORKS FOR REMOTE SENSING IMAGE SEGMENTATION Liu Gag a, b, * a School of Electroic Iformatio, Wuha Uiversity, 430079,

More information

4.2.1 Bayesian Principal Component Analysis Weighted K Nearest Neighbor Regularized Expectation Maximization

4.2.1 Bayesian Principal Component Analysis Weighted K Nearest Neighbor Regularized Expectation Maximization 4 DATA PREPROCESSING 4.1 Data Normalizatio 4.1.1 Mi-Max 4.1.2 Z-Score 4.1.3 Decimal Scalig 4.2 Data Imputatio 4.2.1 Bayesia Pricipal Compoet Aalysis 4.2.2 K Nearest Neighbor 4.2.3 Weighted K Nearest Neighbor

More information

Evaluation scheme for Tracking in AMI

Evaluation scheme for Tracking in AMI A M I C o m m u i c a t i o A U G M E N T E D M U L T I - P A R T Y I N T E R A C T I O N http://www.amiproject.org/ Evaluatio scheme for Trackig i AMI S. Schreiber a D. Gatica-Perez b AMI WP4 Trackig:

More information

Accuracy Improvement in Camera Calibration

Accuracy Improvement in Camera Calibration Accuracy Improvemet i Camera Calibratio FaJie L Qi Zag ad Reihard Klette CITR, Computer Sciece Departmet The Uiversity of Aucklad Tamaki Campus, Aucklad, New Zealad fli006, qza001@ec.aucklad.ac.z r.klette@aucklad.ac.z

More information

An Estimation of Distribution Algorithm for solving the Knapsack problem

An Estimation of Distribution Algorithm for solving the Knapsack problem Vol.4,No.5, 214 Published olie: May 25, 214 DOI: 1.7321/jscse.v4.5.1 A Estimatio of Distributio Algorithm for solvig the Kapsack problem 1 Ricardo Pérez, 2 S. Jös, 3 Arturo Herádez, 4 Carlos A. Ochoa *1,

More information

Journal of Chemical and Pharmaceutical Research, 2013, 5(12): Research Article

Journal of Chemical and Pharmaceutical Research, 2013, 5(12): Research Article Available olie www.jocpr.com Joural of Chemical ad Pharmaceutical Research, 2013, 5(12):745-749 Research Article ISSN : 0975-7384 CODEN(USA) : JCPRC5 K-meas algorithm i the optimal iitial cetroids based

More information

A New Morphological 3D Shape Decomposition: Grayscale Interframe Interpolation Method

A New Morphological 3D Shape Decomposition: Grayscale Interframe Interpolation Method A ew Morphological 3D Shape Decompositio: Grayscale Iterframe Iterpolatio Method D.. Vizireau Politehica Uiversity Bucharest, Romaia ae@comm.pub.ro R. M. Udrea Politehica Uiversity Bucharest, Romaia mihea@comm.pub.ro

More information

LU Decomposition Method

LU Decomposition Method SOLUTION OF SIMULTANEOUS LINEAR EQUATIONS LU Decompositio Method Jamie Traha, Autar Kaw, Kevi Marti Uiversity of South Florida Uited States of America kaw@eg.usf.edu http://umericalmethods.eg.usf.edu Itroductio

More information

CSCI 5090/7090- Machine Learning. Spring Mehdi Allahyari Georgia Southern University

CSCI 5090/7090- Machine Learning. Spring Mehdi Allahyari Georgia Southern University CSCI 5090/7090- Machie Learig Sprig 018 Mehdi Allahyari Georgia Souther Uiversity Clusterig (slides borrowed from Tom Mitchell, Maria Floria Balca, Ali Borji, Ke Che) 1 Clusterig, Iformal Goals Goal: Automatically

More information

Lecture 13: Validation

Lecture 13: Validation Lecture 3: Validatio Resampli methods Holdout Cross Validatio Radom Subsampli -Fold Cross-Validatio Leave-oe-out The Bootstrap Bias ad variace estimatio Three-way data partitioi Itroductio to Patter Recoitio

More information

Analysis of Documents Clustering Using Sampled Agglomerative Technique

Analysis of Documents Clustering Using Sampled Agglomerative Technique Aalysis of Documets Clusterig Usig Sampled Agglomerative Techique Omar H. Karam, Ahmed M. Hamad, ad Sheri M. Moussa Abstract I this paper a clusterig algorithm for documets is proposed that adapts a samplig-based

More information

Software Fault Prediction of Unlabeled Program Modules

Software Fault Prediction of Unlabeled Program Modules Software Fault Predictio of Ulabeled Program Modules C. Catal, U. Sevim, ad B. Diri, Member, IAENG Abstract Software metrics ad fault data belogig to a previous software versio are used to build the software

More information

Cubic Polynomial Curves with a Shape Parameter

Cubic Polynomial Curves with a Shape Parameter roceedigs of the th WSEAS Iteratioal Coferece o Robotics Cotrol ad Maufacturig Techology Hagzhou Chia April -8 00 (pp5-70) Cubic olyomial Curves with a Shape arameter MO GUOLIANG ZHAO YANAN Iformatio ad

More information

Performance Comparisons of PSO based Clustering

Performance Comparisons of PSO based Clustering Performace Comparisos of PSO based Clusterig Suresh Chadra Satapathy, 2 Guaidhi Pradha, 3 Sabyasachi Pattai, 4 JVR Murthy, 5 PVGD Prasad Reddy Ail Neeruoda Istitute of Techology ad Scieces, Sagivalas,Vishaapatam

More information

Lecture Notes 6 Introduction to algorithm analysis CSS 501 Data Structures and Object-Oriented Programming

Lecture Notes 6 Introduction to algorithm analysis CSS 501 Data Structures and Object-Oriented Programming Lecture Notes 6 Itroductio to algorithm aalysis CSS 501 Data Structures ad Object-Orieted Programmig Readig for this lecture: Carrao, Chapter 10 To be covered i this lecture: Itroductio to algorithm aalysis

More information

Improving Template Based Spike Detection

Improving Template Based Spike Detection Improvig Template Based Spike Detectio Kirk Smith, Member - IEEE Portlad State Uiversity petra@ee.pdx.edu Abstract Template matchig algorithms like SSE, Covolutio ad Maximum Likelihood are well kow for

More information

Cluster Analysis. Andrew Kusiak Intelligent Systems Laboratory

Cluster Analysis. Andrew Kusiak Intelligent Systems Laboratory Cluster Aalysis Adrew Kusiak Itelliget Systems Laboratory 2139 Seamas Ceter The Uiversity of Iowa Iowa City, Iowa 52242-1527 adrew-kusiak@uiowa.edu http://www.icae.uiowa.edu/~akusiak Two geeric modes of

More information

Intrusion Detection using Fuzzy Clustering and Artificial Neural Network

Intrusion Detection using Fuzzy Clustering and Artificial Neural Network Itrusio Detectio usig Fuzzy Clusterig ad Artificial Neural Network Shraddha Suraa Research Scholar, Departmet of Computer Egieerig, Vishwakarma Istitute of Techology, Pue Idia shraddha.suraa@gmail.com

More information

Image based Cats and Possums Identification for Intelligent Trapping Systems

Image based Cats and Possums Identification for Intelligent Trapping Systems Volume 159 No, February 017 Image based Cats ad Possums Idetificatio for Itelliget Trappig Systems T. A. S. Achala Perera School of Egieerig Aucklad Uiversity of Techology New Zealad Joh Collis School

More information

Text Feature Selection based on Feature Dispersion Degree and Feature Concentration Degree

Text Feature Selection based on Feature Dispersion Degree and Feature Concentration Degree Available olie at www.ijpe-olie.com vol. 13, o. 7, November 017, pp. 1159-1164 DOI: 10.3940/ijpe.17.07.p19.11591164 Text Feature Selectio based o Feature Dispersio Degree ad Feature Cocetratio Degree Zhifeg

More information

Analysis of Different Similarity Measure Functions and their Impacts on Shared Nearest Neighbor Clustering Approach

Analysis of Different Similarity Measure Functions and their Impacts on Shared Nearest Neighbor Clustering Approach Aalysis of Differet Similarity Measure Fuctios ad their Impacts o Shared Nearest Neighbor Clusterig Approach Ail Kumar Patidar School of IT, Rajiv Gadhi Techical Uiversity, Bhopal (M.P.), Idia Jitedra

More information

BAYESIAN WITH FULL CONDITIONAL POSTERIOR DISTRIBUTION APPROACH FOR SOLUTION OF COMPLEX MODELS. Pudji Ismartini

BAYESIAN WITH FULL CONDITIONAL POSTERIOR DISTRIBUTION APPROACH FOR SOLUTION OF COMPLEX MODELS. Pudji Ismartini Proceedig of Iteratioal Coferece O Research, Implemetatio Ad Educatio Of Mathematics Ad Scieces 014, Yogyakarta State Uiversity, 18-0 May 014 BAYESIAN WIH FULL CONDIIONAL POSERIOR DISRIBUION APPROACH FOR

More information

An Efficient Algorithm for Graph Bisection of Triangularizations

An Efficient Algorithm for Graph Bisection of Triangularizations A Efficiet Algorithm for Graph Bisectio of Triagularizatios Gerold Jäger Departmet of Computer Sciece Washigto Uiversity Campus Box 1045 Oe Brookigs Drive St. Louis, Missouri 63130-4899, USA jaegerg@cse.wustl.edu

More information

ANN WHICH COVERS MLP AND RBF

ANN WHICH COVERS MLP AND RBF ANN WHICH COVERS MLP AND RBF Josef Boští, Jaromír Kual Faculty of Nuclear Scieces ad Physical Egieerig, CTU i Prague Departmet of Software Egieerig Abstract Two basic types of artificial eural etwors Multi

More information

Algorithms for Disk Covering Problems with the Most Points

Algorithms for Disk Covering Problems with the Most Points Algorithms for Disk Coverig Problems with the Most Poits Bi Xiao Departmet of Computig Hog Kog Polytechic Uiversity Hug Hom, Kowloo, Hog Kog csbxiao@comp.polyu.edu.hk Qigfeg Zhuge, Yi He, Zili Shao, Edwi

More information

Redundancy Allocation for Series Parallel Systems with Multiple Constraints and Sensitivity Analysis

Redundancy Allocation for Series Parallel Systems with Multiple Constraints and Sensitivity Analysis IOSR Joural of Egieerig Redudacy Allocatio for Series Parallel Systems with Multiple Costraits ad Sesitivity Aalysis S. V. Suresh Babu, D.Maheswar 2, G. Ragaath 3 Y.Viaya Kumar d G.Sakaraiah e (Mechaical

More information

Pattern Recognition Systems Lab 1 Least Mean Squares

Pattern Recognition Systems Lab 1 Least Mean Squares Patter Recogitio Systems Lab 1 Least Mea Squares 1. Objectives This laboratory work itroduces the OpeCV-based framework used throughout the course. I this assigmet a lie is fitted to a set of poits usig

More information

Our Learning Problem, Again

Our Learning Problem, Again Noparametric Desity Estimatio Matthew Stoe CS 520, Sprig 2000 Lecture 6 Our Learig Problem, Agai Use traiig data to estimate ukow probabilities ad probability desity fuctios So far, we have depeded o describig

More information

Analysis of Server Resource Consumption of Meteorological Satellite Application System Based on Contour Curve

Analysis of Server Resource Consumption of Meteorological Satellite Application System Based on Contour Curve Advaces i Computer, Sigals ad Systems (2018) 2: 19-25 Clausius Scietific Press, Caada Aalysis of Server Resource Cosumptio of Meteorological Satellite Applicatio System Based o Cotour Curve Xiagag Zhao

More information

Task scenarios Outline. Scenarios in Knowledge Extraction. Proposed Framework for Scenario to Design Diagram Transformation

Task scenarios Outline. Scenarios in Knowledge Extraction. Proposed Framework for Scenario to Design Diagram Transformation 6-0-0 Kowledge Trasformatio from Task Scearios to View-based Desig Diagrams Nima Dezhkam Kamra Sartipi {dezhka, sartipi}@mcmaster.ca Departmet of Computig ad Software McMaster Uiversity CANADA SEKE 08

More information

Feature Selection for Change Detection in Multivariate Time-Series

Feature Selection for Change Detection in Multivariate Time-Series Feature Selectio for Chage Detectio i Multivariate Time-Series Michael Botsch Istitute for Circuit Theory ad Sigal Processig Techical Uiversity Muich 80333 Muich, Germay Email: botsch@tum.de Josef A. Nossek

More information

A New Bit Wise Technique for 3-Partitioning Algorithm

A New Bit Wise Technique for 3-Partitioning Algorithm Special Issue of Iteratioal Joural of Computer Applicatios (0975 8887) o Optimizatio ad O-chip Commuicatio, No.1. Feb.2012, ww.ijcaolie.org A New Bit Wise Techique for 3-Partitioig Algorithm Rajumar Jai

More information

Evaluation of Support Vector Machine Kernels for Detecting Network Anomalies

Evaluation of Support Vector Machine Kernels for Detecting Network Anomalies Evaluatio of Support Vector Machie Kerels for Detectig Network Aomalies Prera Batta, Maider Sigh, Zhida Li, Qigye Dig, ad Ljiljaa Trajković Commuicatio Networks Laboratory http://www.esc.sfu.ca/~ljilja/cl/

More information

Probabilistic Fuzzy Time Series Method Based on Artificial Neural Network

Probabilistic Fuzzy Time Series Method Based on Artificial Neural Network America Joural of Itelliget Systems 206, 6(2): 42-47 DOI: 0.5923/j.ajis.2060602.02 Probabilistic Fuzzy Time Series Method Based o Artificial Neural Network Erol Egrioglu,*, Ere Bas, Cagdas Haka Aladag

More information

Keywords Software Architecture, Object-oriented metrics, Reliability, Reusability, Coupling evaluator, Cohesion, efficiency

Keywords Software Architecture, Object-oriented metrics, Reliability, Reusability, Coupling evaluator, Cohesion, efficiency Volume 3, Issue 9, September 2013 ISSN: 2277 128X Iteratioal Joural of Advaced Research i Computer Sciece ad Software Egieerig Research Paper Available olie at: www.ijarcsse.com Couplig Evaluator to Ehace

More information

Optimization of Multiple Input Single Output Fuzzy Membership Functions Using Clonal Selection Algorithm

Optimization of Multiple Input Single Output Fuzzy Membership Functions Using Clonal Selection Algorithm Optimizatio of Multiple Iput Sigle Output Fuzzy Membership Fuctios Usig Cloal Selectio Algorithm AYŞE MERVE ACILAR, AHMET ARSLAN Computer Egieerig Departmet Selcuk Uiversity Selcuk Uiversity, Eg.-Arch.

More information

Chapter 1. Introduction to Computers and C++ Programming. Copyright 2015 Pearson Education, Ltd.. All rights reserved.

Chapter 1. Introduction to Computers and C++ Programming. Copyright 2015 Pearson Education, Ltd.. All rights reserved. Chapter 1 Itroductio to Computers ad C++ Programmig Copyright 2015 Pearso Educatio, Ltd.. All rights reserved. Overview 1.1 Computer Systems 1.2 Programmig ad Problem Solvig 1.3 Itroductio to C++ 1.4 Testig

More information

A Novel Hybrid Algorithm for Software Cost Estimation Based on Cuckoo Optimization and K-Nearest Neighbors Algorithms

A Novel Hybrid Algorithm for Software Cost Estimation Based on Cuckoo Optimization and K-Nearest Neighbors Algorithms Egieerig, Techology & Applied Sciece Research Vol. 6, No. 3, 26, 8-22 8 A Novel Hybrid Algorithm for Software Cost Estimatio Based o Optimizatio ad K-Nearest Neighbors Algorithms Elaz Eskadaria Miadoab

More information

Research on Identification Model of Financial Fraud of Listed Company Based on Data Mining Technology

Research on Identification Model of Financial Fraud of Listed Company Based on Data Mining Technology 208 2d Iteratioal Coferece o Systems, Computig, ad Applicatios (SYSTCA 208) Research o Idetificatio Model of Fiacial Fraud of Listed Compay Based o Data Miig Techology Jiaqi Hu, Xiao Che School of Busiess,

More information

New Fuzzy Color Clustering Algorithm Based on hsl Similarity

New Fuzzy Color Clustering Algorithm Based on hsl Similarity IFSA-EUSFLAT 009 New Fuzzy Color Clusterig Algorithm Based o hsl Similarity Vasile Ptracu Departmet of Iformatics Techology Tarom Compay Bucharest Romaia Email: patrascu.v@gmail.com Abstract I this paper

More information

Designing a learning system

Designing a learning system CS 75 Machie Learig Lecture Desigig a learig system Milos Hauskrecht milos@cs.pitt.edu 539 Seott Square, x-5 people.cs.pitt.edu/~milos/courses/cs75/ Admiistrivia No homework assigmet this week Please try

More information

Description of some supervised learning algorithms

Description of some supervised learning algorithms Descriptio of some supervised learig algorithms Patrick Keekayoro patrick.keekayoro@outlook.com Statistical Cybermetrics Research Group Uiversity of Wolverhampto 1. Supervised learig Supervised machie

More information

Improvement of the Orthogonal Code Convolution Capabilities Using FPGA Implementation

Improvement of the Orthogonal Code Convolution Capabilities Using FPGA Implementation Improvemet of the Orthogoal Code Covolutio Capabilities Usig FPGA Implemetatio Naima Kaabouch, Member, IEEE, Apara Dhirde, Member, IEEE, Saleh Faruque, Member, IEEE Departmet of Electrical Egieerig, Uiversity

More information

CS 683: Advanced Design and Analysis of Algorithms

CS 683: Advanced Design and Analysis of Algorithms CS 683: Advaced Desig ad Aalysis of Algorithms Lecture 6, February 1, 2008 Lecturer: Joh Hopcroft Scribes: Shaomei Wu, Etha Feldma February 7, 2008 1 Threshold for k CNF Satisfiability I the previous lecture,

More information

A Note on Least-norm Solution of Global WireWarping

A Note on Least-norm Solution of Global WireWarping A Note o Least-orm Solutio of Global WireWarpig Charlie C. L. Wag Departmet of Mechaical ad Automatio Egieerig The Chiese Uiversity of Hog Kog Shati, N.T., Hog Kog E-mail: cwag@mae.cuhk.edu.hk Abstract

More information

Chapter 3 Classification of FFT Processor Algorithms

Chapter 3 Classification of FFT Processor Algorithms Chapter Classificatio of FFT Processor Algorithms The computatioal complexity of the Discrete Fourier trasform (DFT) is very high. It requires () 2 complex multiplicatios ad () complex additios [5]. As

More information

Optimization for framework design of new product introduction management system Ma Ying, Wu Hongcui

Optimization for framework design of new product introduction management system Ma Ying, Wu Hongcui 2d Iteratioal Coferece o Electrical, Computer Egieerig ad Electroics (ICECEE 2015) Optimizatio for framework desig of ew product itroductio maagemet system Ma Yig, Wu Hogcui Tiaji Electroic Iformatio Vocatioal

More information

The Counterchanged Crossed Cube Interconnection Network and Its Topology Properties

The Counterchanged Crossed Cube Interconnection Network and Its Topology Properties WSEAS TRANSACTIONS o COMMUNICATIONS Wag Xiyag The Couterchaged Crossed Cube Itercoectio Network ad Its Topology Properties WANG XINYANG School of Computer Sciece ad Egieerig South Chia Uiversity of Techology

More information

Solving Fuzzy Assignment Problem Using Fourier Elimination Method

Solving Fuzzy Assignment Problem Using Fourier Elimination Method Global Joural of Pure ad Applied Mathematics. ISSN 0973-768 Volume 3, Number 2 (207), pp. 453-462 Research Idia Publicatios http://www.ripublicatio.com Solvig Fuzzy Assigmet Problem Usig Fourier Elimiatio

More information

ISSN (Print) Research Article. *Corresponding author Nengfa Hu

ISSN (Print) Research Article. *Corresponding author Nengfa Hu Scholars Joural of Egieerig ad Techology (SJET) Sch. J. Eg. Tech., 2016; 4(5):249-253 Scholars Academic ad Scietific Publisher (A Iteratioal Publisher for Academic ad Scietific Resources) www.saspublisher.com

More information

Counting the Number of Minimum Roman Dominating Functions of a Graph

Counting the Number of Minimum Roman Dominating Functions of a Graph Coutig the Number of Miimum Roma Domiatig Fuctios of a Graph SHI ZHENG ad KOH KHEE MENG, Natioal Uiversity of Sigapore We provide two algorithms coutig the umber of miimum Roma domiatig fuctios of a graph

More information

Dynamic Programming and Curve Fitting Based Road Boundary Detection

Dynamic Programming and Curve Fitting Based Road Boundary Detection Dyamic Programmig ad Curve Fittig Based Road Boudary Detectio SHYAM PRASAD ADHIKARI, HYONGSUK KIM, Divisio of Electroics ad Iformatio Egieerig Chobuk Natioal Uiversity 664-4 Ga Deokji-Dog Jeoju-City Jeobuk

More information

c-dominating Sets for Families of Graphs

c-dominating Sets for Families of Graphs c-domiatig Sets for Families of Graphs Kelsie Syder Mathematics Uiversity of Mary Washigto April 6, 011 1 Abstract The topic of domiatio i graphs has a rich history, begiig with chess ethusiasts i the

More information

An Efficient Algorithm for Graph Bisection of Triangularizations

An Efficient Algorithm for Graph Bisection of Triangularizations Applied Mathematical Scieces, Vol. 1, 2007, o. 25, 1203-1215 A Efficiet Algorithm for Graph Bisectio of Triagularizatios Gerold Jäger Departmet of Computer Sciece Washigto Uiversity Campus Box 1045, Oe

More information

Morgan Kaufmann Publishers 26 February, COMPUTER ORGANIZATION AND DESIGN The Hardware/Software Interface. Chapter 5

Morgan Kaufmann Publishers 26 February, COMPUTER ORGANIZATION AND DESIGN The Hardware/Software Interface. Chapter 5 Morga Kaufma Publishers 26 February, 28 COMPUTER ORGANIZATION AND DESIGN The Hardware/Software Iterface 5 th Editio Chapter 5 Set-Associative Cache Architecture Performace Summary Whe CPU performace icreases:

More information

Octahedral Graph Scaling

Octahedral Graph Scaling Octahedral Graph Scalig Peter Russell Jauary 1, 2015 Abstract There is presetly o strog iterpretatio for the otio of -vertex graph scalig. This paper presets a ew defiitio for the term i the cotext of

More information

Sectio 4, a prototype project of settig field weight with AHP method is developed ad the experimetal results are aalyzed. Fially, we coclude our work

Sectio 4, a prototype project of settig field weight with AHP method is developed ad the experimetal results are aalyzed. Fially, we coclude our work 200 2d Iteratioal Coferece o Iformatio ad Multimedia Techology (ICIMT 200) IPCSIT vol. 42 (202) (202) IACSIT Press, Sigapore DOI: 0.7763/IPCSIT.202.V42.0 Idex Weight Decisio Based o AHP for Iformatio Retrieval

More information

Performance Plus Software Parameter Definitions

Performance Plus Software Parameter Definitions Performace Plus+ Software Parameter Defiitios/ Performace Plus Software Parameter Defiitios Chapma Techical Note-TG-5 paramete.doc ev-0-03 Performace Plus+ Software Parameter Defiitios/2 Backgroud ad Defiitios

More information

Fuzzy Minimal Solution of Dual Fully Fuzzy Matrix Equations

Fuzzy Minimal Solution of Dual Fully Fuzzy Matrix Equations Iteratioal Coferece o Applied Mathematics, Simulatio ad Modellig (AMSM 2016) Fuzzy Miimal Solutio of Dual Fully Fuzzy Matrix Equatios Dequa Shag1 ad Xiaobi Guo2,* 1 Sciece Courses eachig Departmet, Gasu

More information

Comparison of classification algorithms in the task of object recognition on radar images of the MSTAR base

Comparison of classification algorithms in the task of object recognition on radar images of the MSTAR base Compariso of classificatio algorithms i the task of object recogitio o radar images of the MSTAR base A.A. Borodiov 1, V.V. Myasikov 1,2 1 Samara Natioal Research Uiversity, 34 Moskovskoe Shosse, 443086,

More information

Extending The Sleuth Kit and its Underlying Model for Pooled Storage File System Forensic Analysis

Extending The Sleuth Kit and its Underlying Model for Pooled Storage File System Forensic Analysis Extedig The Sleuth Kit ad its Uderlyig Model for Pooled File System Foresic Aalysis Frauhofer Istitute for Commuicatio, Iformatio Processig ad Ergoomics Ja-Niclas Hilgert* Marti Lambertz Daiel Plohma ja-iclas.hilgert@fkie.frauhofer.de

More information

Arithmetic Sequences

Arithmetic Sequences . Arithmetic Sequeces COMMON CORE Learig Stadards HSF-IF.A. HSF-BF.A.1a HSF-BF.A. HSF-LE.A. Essetial Questio How ca you use a arithmetic sequece to describe a patter? A arithmetic sequece is a ordered

More information

A Parallel DFA Minimization Algorithm

A Parallel DFA Minimization Algorithm A Parallel DFA Miimizatio Algorithm Ambuj Tewari, Utkarsh Srivastava, ad P. Gupta Departmet of Computer Sciece & Egieerig Idia Istitute of Techology Kapur Kapur 208 016,INDIA pg@iitk.ac.i Abstract. I this

More information

Data-Driven Nonlinear Hebbian Learning Method for Fuzzy Cognitive Maps

Data-Driven Nonlinear Hebbian Learning Method for Fuzzy Cognitive Maps Data-Drive Noliear Hebbia Learig Method for Fuzzy Cogitive Maps Wociech Stach, Lukasz Kurga, ad Witold Pedrycz Abstract Fuzzy Cogitive Maps (FCMs) are a coveiet tool for modelig of dyamic systems by meas

More information

Assignment Problems with fuzzy costs using Ones Assignment Method

Assignment Problems with fuzzy costs using Ones Assignment Method IOSR Joural of Mathematics (IOSR-JM) e-issn: 8-8, p-issn: 9-6. Volume, Issue Ver. V (Sep. - Oct.06), PP 8-89 www.iosrjourals.org Assigmet Problems with fuzzy costs usig Oes Assigmet Method S.Vimala, S.Krisha

More information

A Study on the Performance of Cholesky-Factorization using MPI

A Study on the Performance of Cholesky-Factorization using MPI A Study o the Performace of Cholesky-Factorizatio usig MPI Ha S. Kim Scott B. Bade Departmet of Computer Sciece ad Egieerig Uiversity of Califoria Sa Diego {hskim, bade}@cs.ucsd.edu Abstract Cholesky-factorizatio

More information

The Closest Line to a Data Set in the Plane. David Gurney Southeastern Louisiana University Hammond, Louisiana

The Closest Line to a Data Set in the Plane. David Gurney Southeastern Louisiana University Hammond, Louisiana The Closest Lie to a Data Set i the Plae David Gurey Southeaster Louisiaa Uiversity Hammod, Louisiaa ABSTRACT This paper looks at three differet measures of distace betwee a lie ad a data set i the plae:

More information

performance to the performance they can experience when they use the services from a xed location.

performance to the performance they can experience when they use the services from a xed location. I the Proceedigs of The First Aual Iteratioal Coferece o Mobile Computig ad Networkig (MobiCom 9) November -, 99, Berkeley, Califoria USA Performace Compariso of Mobile Support Strategies Rieko Kadobayashi

More information

Effect of control points distribution on the orthorectification accuracy of an Ikonos II image through rational polynomial functions

Effect of control points distribution on the orthorectification accuracy of an Ikonos II image through rational polynomial functions Effect of cotrol poits distributio o the orthorectificatio accuracy of a Ikoos II image through ratioal polyomial fuctios Marcela do Valle Machado 1, Mauro Homem Atues 1 ad Paula Debiasi 1 1 Federal Rural

More information

Using The Central Limit Theorem for Belief Network Learning

Using The Central Limit Theorem for Belief Network Learning Usig The Cetral Limit Theorem for Belief Network Learig Ia Davidso, Mioo Amiia Computer Sciece Dept, SUNY Albay Albay, NY, USA,. davidso@cs.albay.edu Abstract. Learig the parameters (coditioal ad margial

More information

CLASSIFICATION MODELS BASED FORWARD SELECTION FOR BUSINESS PERFORMANCE PREDICTION

CLASSIFICATION MODELS BASED FORWARD SELECTION FOR BUSINESS PERFORMANCE PREDICTION Joural of Theoretical ad Applied Iformatio Techology 20 th September 204. Vol. 67 No.2 2005-204 JATIT & LLS. All rights reserved. ISSN: 992-8645 www.atit.org E-ISSN: 87-395 CLASSIFICATION MODELS BASED

More information

Stone Images Retrieval Based on Color Histogram

Stone Images Retrieval Based on Color Histogram Stoe Images Retrieval Based o Color Histogram Qiag Zhao, Jie Yag, Jigyi Yag, Hogxig Liu School of Iformatio Egieerig, Wuha Uiversity of Techology Wuha, Chia Abstract Stoe images color features are chose

More information