A Post Randomization Framework for Privacy-Preserving Bayesian. Network Parameter Learning

Size: px
Start display at page:

Download "A Post Randomization Framework for Privacy-Preserving Bayesian. Network Parameter Learning"

Transcription

1 A Post Randomzaton Framework for Prvacy-Preservng Bayesan Network Parameter Learnng JIANJIE MA K.SIVAKUMAR School Electrcal Engneerng and Computer Scence, Washngton State Unversty Pullman, WA {jma, Abstract: - Post Randomzaton technque has been successfully used n statstcal dsclosure lmtaton. The applcaton Post Randomzaton technque to Prvacy-Preservng data mnng s explored n ths paper. The problem prvacy-preservng Bayesan network parameter learnng s consdered as a specfc example. We propose to use post randomzaton technque to randomze the prvacy-senstve varables n learnng Bayesan network parameters from dstrbuted heterogeneous databases. The only requred nformaton from the data set s a set suffcent statstcs for learnng Bayesan network parameters. The proposed method estmates the suffcent statstcs from the randomzed data. We show both theoretcally and expermentally that ths method learns a set accurate parameters, even under large levels randomzaton. We also llustrate the trade f between prvacy and accuracy by smulatons. Key-Words: - Bayesan Network, Prvacy-Preservng Data Mnng, Dstrbuted Heterogeneous Databases, Post Randomzaton. Introducton Prvacy-preservng data mnng deals wth the problem buldng accurate data mnng models over aggregate data, whle protectng prvacy at the level ndvdual records. There are two man approaches to prvacy-preservng data mnng. One approach s to perturb or randomze the data before sendng t to the data mner. The perturbed or randomzed data are then used to learn or mne the models and patterns [,]. The other approach s to use secure multparty computaton (SMC) to enable two or more partes to buld data models wthout every party learnng anythng about the other party s data [4]. Prvacy-preservng Bayesan network (BN) learnng s a more recent topc. Wrght and Yang [] dscuss prvacy-preservng BN structure computaton on dstrbuted heterogeneous databases whle Meng et al. [8] have consdered the prvacy-senstve BN parameter learnng problem. The underlyng method used n both works s to convert the computatons requred for BN learnng nto a seres nner product computatons and then to use a secure nner product computaton method proposed elsewhere. The number secure computaton operatons ncreases exponentally wth the possble confguratons the problem varables. The current work on prvacy-preservng BN learnng focuses on the multparty models, whch requres that every party have some computatonal capablty. Besdes ths model, our work consders a model where there s a data mner who actually does all the computatons for the partcpatng partes. SMC method has the followng two drawbacks: () t assumes a sem-honest model, whch s ten unrealstc n the real world () t requres large volumes synchronzed computatons among

2 partcpatng partes. Most the synchronzed computatons are overheads due to prvacy requrement. Post randomzaton overcomes the drawbacks SMC method by a trade f between accuracy and prvacy. A malcous party who does not obey the protocol n SMC method can easly get some prvate nformaton other partes whch he wll not be able to f post randomzatons are mplemented to ndvdual data records.. Problem Formulaton Prvacy-Preservng BN learnng nvolves dstrbuted databases, where the database s owned by several partes. If the database s homogeneously dstrbuted, prvacy-preservng BN Learnng s relatvely easy snce every party can send data mner (or other partes) a set suffcent statstcs from hs part the database. Prvacy ndvdual records wll not be breached by sendng suffcent statstcs to other partes or data mner. The problem prvacy-preservng BN learnng from heterogeneous database s that several partes who each own a vertcal porton the database want to learn a global BN for ther mutual benefts but they are concerned about the prvacy ther senstve varables. In ths paper, we consder the problem BN parameter learnng for the case dscrete varables. We consder the followng two models. Model I: There s no data mner; every party has to do some porton the learnng computatons. Every party sends ther randomzed data to those partes who need those data. Model II: There s a data mner who s does all computatons for the partcpatng partes. Every party smply sends all ther randomzed data to the data mner. 3. Prvacy Analyss for Post Randomzaton Consder a database D wth n varables { X X n }, where X takes dscrete values from the set S. The post randomzaton for varable X s a (random) mappng R : S S, based on a set transton probabltes p lm p( X km X kl ), where km, kl S and X denotes the (randomzed) varable value correspondng to varable probablty X. The transton plm s the probablty that a varable X wth orgnal value k l s randomzed to the value k m. Post Randomzaton s so named because the randomzaton happens after data have been collected. Let P { p lm} denote the K K matrx that has plm as ts ( l, m) th entry, where K s the cardnalty the set S. The condton that P s nonsngular has to be mposed f we want to estmate the frequency dstrbuton varable from the randomzed varables. In the followng, we gve out some smple but effectve post randomzaton schemes on whch our experments are based. If varable X takes bnary values, we can use Bnary Randomzaton as shown n Fg. (a). If the varable s ternary, ternary symmetrc channel as shown n Fg.(b) can be used. p p p p (a) Fg. : X p p p Randomzaton Schemes We can apply the same randomzaton schemes ndependently to all the varables: unform randomzaton to the data set. Alternatvely, we can use a non-unform randomzaton where dfferent post randomzaton schemes are appled to dfferent varables ndependently. The non-unform randomzaton s effectve when dfferent varables have dfferent senstvty levels. For example, we can choose dfferent randomzaton parameters p 3 p (b) 3

3 and p to dfferent bnary varables for non-unform randomzaton f the prvacy requrement the two varables are dfferent. The non-unform randomzaton ncludes the specal case when there s no prvacy requrement for some the varables. From the above, we can see that f varable X takes K values (or categores), the dmenson P wll be K K. Wth larger K, more randomzaton s ntroduced nto varable general. Ths s good from a prvacy pont vew. However, the varances the estmators for frequency counts wll also be lager under the same sample sze. One soluton for ths problem s to partton the K categores nto several groups such that a value n one group can only be randomzed to a value n the same group. In ths case, Matrx P becomes a block dagonal matrx. The problem how many groups should the K values be parttoned nto s a matter desgn choce. The post randomzaton can also be mplemented to several varables smultaneously. For example, the varables X X and X j can be randomzed smultaneously accordng to transton probablty p X l, X l X k, X ). ( j j k Randomzng varables smultaneously can avod the possble nconsstency the database caused by randomzaton. We consder the noton prvacy ntroduced by Evfmevsk et al. [5] n terms an amplfcaton factor γ. The amplfcaton γ n [5] s proposed n the framework where every data record should be randomzed wth a factor greater than γ, before the data are sent to the data mner, to lmt prvacy breach. However, n ths paper, we use the amplfcaton γ purely as a worst-case quantfcaton prvacy for a desgned post randomzaton scheme. It s proved n [5] that f the randomzaton operator s at most γ amplfyng, revealng n X k wll cause nether an upward ρ -to- ρ prvacy breach nor a downward ρ -to- ρ prvacy breach f ρ ρ ρ ρ > γ. Clearly, the smaller the value γ, the better s the worst case prvacy. Ideally we would lke to have γ. The at most γ amplfcaton provdes a worst case quantfcaton prvacy. However, t does not provde any nformaton about prvacy n general. Besdes γ, we use K mn #{ k P( X k' X k) > }, whch s k the mnmum number possble categores that can be randomzed to category k' n a desgned post randomzaton, as another quantfcaton prvacy. Ths K ndcates the prvacy preserved n general. It s smlar to the K defned n K-anonymty n [9] but n probablstc sense. If we group the categores a varable nto several group, then K become smaller n general 4. Post Randomzaton Framework for Parameter Learnng For parameter learnng, we assume the structure G s fxed and known to every partcpatng party. For Model I, we use the defnton cross varable and cross parents defned n [3]. N jk s the number records such that X s n k th category whle ts parents are n j th category. For each party a () Randomze cross parents at same ste accordng to ther respectve prvacy requrements usng post randomzaton descrbed n Secton 3. Randomzatons are done ndependently for each (combned) varable and each record. () Send randomzed cross parents party a for party a j to party a j together wth the probablty transton matrx used. (3) Learn parameters for local varables n party a. Ths step does not nvolve randomzed data. (4) Estmate the suffcent statstcs Njk s for each

4 cross varable at same ste usng local data and randomzed parent data from other partes. (5) Compute the parameters for cross varables usng the estmated suffcent statstcs Nˆ s. (6) Share the parameters wth all other partes. Local varables at each ste are not randomzed for local calculatons. Steps learnng parameters for model II: For each party a : () Randomze all senstve varables accordng to ther respectve prvacy requrements usng post randomzaton descrbed n Secton 3. Randomzatons are done ndependently for each (combned) varable and each record. () Send randomzed data and ther correspondng probablty transton matrces to the data mner. For the data mner: () Estmate the suffcent statstcs Njk jk for each node X usng the randomzed data from partcpatng partes. () Estmate the parameters usng the estmated suffcent statstcs Nˆ. jk (3) Broadcast the parameters to all partes. The detals estmaton suffcent statstcs and parameter (step 4 and 5 for Model I, Step and for data mner n Model II) from randomzed data are descrbed n Secton Estmaton Suffcent Statstcs & Parameters from Randomzed Data The problem prvacy-preservng BN Parameter learnng can be decomposed nto a seres estmaton N jk s for each node X and a gven fxed structure G from the randomzed data D. Consder the followng general case: Varable wth cardnalty X K has Q parent nodes Pa ( ),, Pa ( Q). The cardnalty Pa (q) s K (q) Pa. These varables can be arbtrary vertcally parttoned to dfferent partes n both models. The randomzaton each (combned) varable can also be done by groupng the categores the varables nto groups. We have the followng dfferent cases for estmatng Njk s from the randomzed data D due to smultaneous randomzaton. (a) X and ts parents are all randomzed ndependently each other. (b) Some parents X are randomzed smultaneously. (c) X s randomzed smultaneously wth some ts parents. (d) X s randomzed smultaneously wth non-parent varables. For (b) and (c) above, we can consder the smultaneously randomzed varables as combned varables n estmatng the suffcent statstcs. For example, f varable X s randomzed smultaneously wth one ts parents Pa (), N jk s equal to the number records such Q that ( X ; Pa ()) ( k, j ), Pa ( ) j,, Pa ( Q) j, where ( X; Pa ()) s a combned varable. Thus, we can estmate the Njk s from the randomzed data by consderng ( X; Pa ()) as a sngle varable wth cardnalty K (). For case (d), snce the current K Pa Njk doesn't nvolve the varable randomzed smultaneously wth X, the data mner can get the margnal transton probablty matrx from the gven transton matrx the combned varable. From the above arguments, we conclude that the cases (b), (c), and (d) above can effectvely be consdered to be equvalent to case (a). Hence, wthout loss generalty, we can dscuss case (a) only. We denote by Pa X ) as a compound varable ( for all the parents Varable Pa( X ) takes J K Q q Pa q) X. Hence ( dfferent values.

5 K N j N jk k and N s J K dmensonal vector N jk values, that s t ( N, N,, N K, N, NJ K, where N ) superscrpt t denotes transpose. N (l) s an element N. N jk, N j and N are defned smlarly as N jk, Nj and N respectvely but for the randomzed data D. N jk, Nj and Nˆ jk, Nˆ j and Nˆ are estmators N respectvely. Gven the tranng data D wth N records varables X and ts Q parents n the above general case, f they are post-randomzed wth transton probablty Pa () Pa (Q) matrces P, P,, P, respectvely, we have the followng theorem. Theorem : E t [ N D] P N, where We can use the estmated suffcent statstcs to get ML estmate the parameters as Nˆ jk Nˆ jk θˆ jk K N and the MAP estmate the ˆ j Nˆ k parameters as θ ˆ dstrbuton jk jk α α jk j + Nˆ + Nˆ jk j, where the pror θj s assumed to be Drchlet wth parameter α, α,, α }. The dstrbuton estmator { j j jr θˆ jk s dscussed n [7]. One mportant result from [7] s that the dstrbuton the estmator θˆ jk can be approxmated as a normal dstrbuton wth mean θ jk and wth a varance the order, where N N s the tranng sample sze. P pa pa pa ( ) pa () pa ( Q) and P P P P P P, denotes Kronecker matrx product. Moreover, J K Cov[ N D] N ( l) Vl where Vl s a l covarance matrx such that ts P( l, l)( P( l, l)) f l l V l ( l, l). P( l, l ) P( l, l) f l l K J K J ( l, l) th element s Pros are omtted here due to the page lmtatons. Interested readers can refer to a longer verson ths paper for detals [7]. The followng theorem establshes the bas and varance the estmator N N. Its pro s straght-forward ˆ t ( P ) and s omtted. Theorem : for ˆ t ( P ) N N s an unbased estmator N and ˆ t Cov{ N D} ( P ) Cov{ N D}( P ), where P and Cov{ N D} are defned n Theorem. 6. Expermental Results 6. Non-unform Randomzaton In ths experment, we use the Bayesan Network shown n Fg., where the varables are dstrbuted over three stes. All varables are bnary except varables L and B whch are ternary. The condtonal probabltes the dfferent nodes are also shown., samples were generated from ths Bayesan Network to form the dataset D. Ths data was then randomzed accordng to the scheme descrbed n Table, where varables T, S, and G were consdered not senstve and hence not randomzed. The correspondng at most γ amplfcaton s also shown n Table. K for Bnary randomzaton whereas K3 for ternary randomzaton. Table shows a part parameters learnt from the randomzed data usng the algorthm descrbed n Secton 4 for Model II. Less randomzaton occurs n Model I, so the results for Model I are better than

6 those for Model II. The remanng part can be calculated by one mnus the gven part. All the values n the Table are average over 5 ndependent runs, wth the correspondng standard devaton ndcated n parenthess. It s clear from the Table that the proposed algorthms can accurately learn the BN parameters for both scenaros, even for moderate levels randomzaton. A T X Ste A.7,.3 T.,.9,.9,. S.5,.5 L.3,.7,.4,.5,.3,.5 X.,.6,.8,.4 F.5,.9,.75,. E.5,.8,.5,.5,.3,.4,.75,.,.85,.5,.7,.6 D.7,.65,.,.4,.8,.35,.3,.35,.9,.6,.,.65 C.9,.4,.6,.5,.,.6,.4,.75 B.8,.5,.,.5,.,.35 F E G.,.4,.8,.6 Fg. : A Bayesan Network for experment 6. A,D Bnary symmetrc p p. 5, γ 3 L,B Ternary symmetrc p p. 5 γ E Bnary symmetrc p p. γ 4 X Bnary symmetrc p p. γ 4 C,F Bnary p. p. 5 γ 9 C L Ste 3 Ste Table : Randomzaton performed 6. Trade f between Prvacy and Accuracy In ths experment, we use the Bayesan network shown n Fg. 3, where varables are dstrbuted over two stes. All Varables are bnary. We generated, samples from ths Bayesan Network. In order to see the trade f between prvacy and accuracy, we randomze the samples usng bnary symmetrc randomzaton wth dfferent levels p p p and learn parameters from randomzed samples usng the method dscussed n Secton 4. As n prevous experment, we only present the results G D S B usng Model II. In the experment usng Model II, every varable s randomzed usng the Symmetrc Bnary Randomzaton wth the same randomzaton level p. Snce parameters assocated wth a node s nothng but the condtonal probablty gven ts parents, the accuracy parameters assocated wth a node can be measured by condtonal Kullback-Lebler (CKL) dstance between the parameters learnt from randomzed data and those learnt from non-randomzed data. The CKL dstance for node n our case s J () ( p) D( X, p) P( pa j) DKL( P ( X pa j), P ( X pa j)), j () ( p) where P ( X pa j) and P ( X pa j) are the parameters learnt from non-randomzed data and A.7(.7) T.(.5).9(.77) S.5(.) X.(.8).6(.) L.3(.49).7(.57).39(.64).4(.55) B.8(.77).6(.4).94(.7).49(.73) E.5(.).8(.9).4(.7).5(.).3(.6).4(.34) D.69(.).65(.3).(3.3).38(.77).79(.7).39(5.65) C.9(.).38(.6).6(.6).5(.) F.4(.73).9(.) G.(.3).4(.9) Table : Mean and standard devaton ( - ) over 5 runs parameters learnt from the randomzed data. those learnt from randomzed data wth randomzaton level p respectvely and D denotes the ordnary KL dstance between two dstrbutons. We present those dstances assocated wth node C, node D and node F n Fg. 4. Those nodes are typcal nodes for the gven Bayesan network. The averages are over ndependent runs. Average plus one standard devaton runs s also depcted (wth dotted lne). From the Fg. 4, we can clearly see the trade f between accuracy and prvacy. Snce we use the symmetrc bnary randomzaton, more prvacy s preserved wth bgger p when p <. 5. Wth, tranng samples, the method stll gets KL

7 good accuracy when p.3. condtonal KL dstance condtonal KL dstance condtonal KL dstance A.5.5 G.3.7 B E C D F Fg. 3: A Bayesan Network for experment Avg Avg+Std Node C randomzaton parameter p Avg Avg+Std Node D randomzaton parameter p..5 A B Ste C Avg Avg+Std Node F randomzaton parameter p Fg. 4: CKL dstance vs. randomzaton level p 6.3 Tranng Sample sze As ponted out n Secton 4, the varance the D E Ste F G estmator parameter θ jk s the order one over the sample sze N. Thus, under the same accuracy requrement, more prvacy can be preserved f there are more tranng samples. Ths experment s performed to llustrate the effect tranng sample 8 sze. We generated 5 tranng samples usng Bayesan network n Fg. 3. The proposed method n Secton 4 s used to learn the Bayesan parameters from randomzed data wth randomzaton levels p., p., p. 3, and p. 4 wth tranng sample sze k 5 ( k 8 ) respectvely. The experment results are shown n Fg. 5. As n experment 6., the average s over ndependent runs and the average plus one standard devaton s also shown. The experment results for randomzaton level p. 4 are shown separately. Those Condtonal dstances out the scale vertcal axs are not shown n the Fgure. From ths experment, we can clearly see that tranng sample sze play a key role n the trade f between accuracy and prvacy. We can see that when the tranng sample sze s very large, we can have both good prvacy and good accuracy. 7. Concluson We have proposed a post randomzaton technque to learn parameters a Bayesan network from dstrbuted heterogeneous data. Our method estmates the suffcent statstcs from the randomzed data, whch are subsequently used to learn the parameters. Our experments show that post randomzaton s an effcent, flexble, and easy-to-use method to learn Bayesan network parameters from prvacy senstve data. Currently, we are explorng the extenson post randomzaton technques to learn BN structure from senstve data. The dea estmatng suffcent statstcs from randomzed data can be used to learn other data mnng models lke decson trees. We plan to report these extensons and applcatons n a future publcaton.

8 Condtonal KL dstance Condtonal KL dstance Condtonal KL dstance Condtonal KL dstance node C when p.,. and.3 Avg(p.) Avg+Std(p.) Avg(p.) Avg+Std(p.) Avg(p.3) Avg+Std(p.3) tranng sample sze5* k node D when p.,. and.3 Avg(p.) Avg+Std(p.) Avg(p.) Avg+Std(p.) Avg(p.3) Avg+Std(p.3) tranng sample sze5* k node F when p.,. and.3 Avg(p.) Avg+Std(p.) Avg(p.) Avg+Std(p.) Avg(p.3) Avg+Std(p.3) tranng sample sze5* k node C, D, F when p.4 Avg(C) Avg+Std(C) Avg(D) Avg+Std(D) Avg(F) Avg+Std(F) tranng sample sze5* k Fg. 5: CKL dstance vs. tranng sample sze References: [] D. Agrawal and C. C. Aggarwal. On the Desgn and Quantfcaton Prvacy Preservng Data Mnng Algorthm, SIGMOD [] R. Agrawal and R. Srkant. Prvacy-preservng data mnng. In Proceedngs SIGMOD Conference on Management Data, pages , May. [3] R. Chen, K. Svakumar, and H. Kargupta, Collectve Mnng Bayesan Networks from Dstrbuted Heterogeneous Data, Knowledge and Informaton Systems Journal, vol. 6, 4. [4] C. Clfton, M. Kantarcoglu, J. Vadya, X. Ln, and M. Zhu. Tools for Prvacy Preservng Dstrbuted Data mnng. ACM SIGKDD Exploratons, 4():8-34, 3. [5] A. Evfmevsk, J. Gehrke, and R. Srkant. Lmtng prvacy breaches n prvacy preservng data mnng. In proceedngs the ACM SIGMOD/POD Conference, pages -, San Dego, CA, June 3. [6] J. M. Gouweleeuw, P. Kooman, L.C.R.J. Wllenborg, and P.-P. de Wolf. Post Randomsaton for Statstcal Dsclosure Control: Theory and Implementaton. Journal fcal Statstcs, Vol pages [7] J. Ma and K. Svakumar, Prvacy-Preservng Bayesan Network Learnng Usng Post Randomzaton, (n preparaton), 5. [8] D. Meng, K. Svakumar and H. Kargupta. Prvacy-Senstve Bayesan Network Parameter Learnng. In the Fourth IEEE Internatonal Conference on Data Mnng. Brghton, UK. November 4. [9] L.Sweeney. k-anonymty: a model for protectng prvacy. Internatonal Journal on uncertanty, Fuzzness and Knowledge-based Systems, (5):557-57,. [] R. Wrght and Z. Yang. Prvacy Preservng Bayesan Network Structure Computaton on Dstrbuted Heterogeneous Data. In Proceedngs the 4 ACM SIGKDD nternatonal conference on Knowledge dscovery and data mnng.

A Binarization Algorithm specialized on Document Images and Photos

A Binarization Algorithm specialized on Document Images and Photos A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a

More information

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

Parallelism for Nested Loops with Non-uniform and Flow Dependences

Parallelism for Nested Loops with Non-uniform and Flow Dependences Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr

More information

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth

More information

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

An Optimal Algorithm for Prufer Codes *

An Optimal Algorithm for Prufer Codes * J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,

More information

A NOTE ON FUZZY CLOSURE OF A FUZZY SET

A NOTE ON FUZZY CLOSURE OF A FUZZY SET (JPMNT) Journal of Process Management New Technologes, Internatonal A NOTE ON FUZZY CLOSURE OF A FUZZY SET Bhmraj Basumatary Department of Mathematcal Scences, Bodoland Unversty, Kokrajhar, Assam, Inda,

More information

Cluster Analysis of Electrical Behavior

Cluster Analysis of Electrical Behavior Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School

More information

Module Management Tool in Software Development Organizations

Module Management Tool in Software Development Organizations Journal of Computer Scence (5): 8-, 7 ISSN 59-66 7 Scence Publcatons Management Tool n Software Development Organzatons Ahmad A. Al-Rababah and Mohammad A. Al-Rababah Faculty of IT, Al-Ahlyyah Amman Unversty,

More information

Determining the Optimal Bandwidth Based on Multi-criterion Fusion

Determining the Optimal Bandwidth Based on Multi-criterion Fusion Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn

More information

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance Tsnghua Unversty at TAC 2009: Summarzng Mult-documents by Informaton Dstance Chong Long, Mnle Huang, Xaoyan Zhu State Key Laboratory of Intellgent Technology and Systems, Tsnghua Natonal Laboratory for

More information

For instance, ; the five basic number-sets are increasingly more n A B & B A A = B (1)

For instance, ; the five basic number-sets are increasingly more n A B & B A A = B (1) Secton 1.2 Subsets and the Boolean operatons on sets If every element of the set A s an element of the set B, we say that A s a subset of B, or that A s contaned n B, or that B contans A, and we wrte A

More 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

Proper Choice of Data Used for the Estimation of Datum Transformation Parameters

Proper Choice of Data Used for the Estimation of Datum Transformation Parameters Proper Choce of Data Used for the Estmaton of Datum Transformaton Parameters Hakan S. KUTOGLU, Turkey Key words: Coordnate systems; transformaton; estmaton, relablty. SUMMARY Advances n technologes and

More information

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data

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

More information

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

X- Chart Using ANOM Approach

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

More information

Learning-Based Top-N Selection Query Evaluation over Relational Databases

Learning-Based Top-N Selection Query Evaluation over Relational Databases Learnng-Based Top-N Selecton Query Evaluaton over Relatonal Databases Lang Zhu *, Wey Meng ** * School of Mathematcs and Computer Scence, Hebe Unversty, Baodng, Hebe 071002, Chna, zhu@mal.hbu.edu.cn **

More information

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 A mathematcal programmng approach to the analyss, desgn and

More information

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

Learning from Multiple Related Data Streams with Asynchronous Flowing Speeds

Learning from Multiple Related Data Streams with Asynchronous Flowing Speeds Learnng from Multple Related Data Streams wth Asynchronous Flowng Speeds Zh Qao, Peng Zhang, Jng He, Jnghua Yan, L Guo Insttute of Computng Technology, Chnese Academy of Scences, Bejng, 100190, Chna. School

More information

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

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

More information

R s s f. m y s. SPH3UW Unit 7.3 Spherical Concave Mirrors Page 1 of 12. Notes

R s s f. m y s. SPH3UW Unit 7.3 Spherical Concave Mirrors Page 1 of 12. Notes SPH3UW Unt 7.3 Sphercal Concave Mrrors Page 1 of 1 Notes Physcs Tool box Concave Mrror If the reflectng surface takes place on the nner surface of the sphercal shape so that the centre of the mrror bulges

More information

Simulation Based Analysis of FAST TCP using OMNET++

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

More information

The Research of Support Vector Machine in Agricultural Data Classification

The Research of Support Vector Machine in Agricultural Data Classification The Research of Support Vector Machne n Agrcultural Data Classfcaton Le Sh, Qguo Duan, Xnmng Ma, Me Weng College of Informaton and Management Scence, HeNan Agrcultural Unversty, Zhengzhou 45000 Chna Zhengzhou

More information

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

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

Lecture 4: Principal components

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

More information

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana

More information

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour 6.854 Advanced Algorthms Petar Maymounkov Problem Set 11 (November 23, 2005) Wth: Benjamn Rossman, Oren Wemann, and Pouya Kheradpour Problem 1. We reduce vertex cover to MAX-SAT wth weghts, such that the

More information

Empirical Distributions of Parameter Estimates. in Binary Logistic Regression Using Bootstrap

Empirical Distributions of Parameter Estimates. in Binary Logistic Regression Using Bootstrap Int. Journal of Math. Analyss, Vol. 8, 4, no. 5, 7-7 HIKARI Ltd, www.m-hkar.com http://dx.do.org/.988/jma.4.494 Emprcal Dstrbutons of Parameter Estmates n Bnary Logstc Regresson Usng Bootstrap Anwar Ftranto*

More information

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching A Fast Vsual Trackng Algorthm Based on Crcle Pxels Matchng Zhqang Hou hou_zhq@sohu.com Chongzhao Han czhan@mal.xjtu.edu.cn Ln Zheng Abstract: A fast vsual trackng algorthm based on crcle pxels matchng

More information

Concurrent Apriori Data Mining Algorithms

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

More information

Classifier Selection Based on Data Complexity Measures *

Classifier Selection Based on Data Complexity Measures * Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.

More information

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS

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

More information

Solving two-person zero-sum game by Matlab

Solving two-person zero-sum game by Matlab Appled Mechancs and Materals Onlne: 2011-02-02 ISSN: 1662-7482, Vols. 50-51, pp 262-265 do:10.4028/www.scentfc.net/amm.50-51.262 2011 Trans Tech Publcatons, Swtzerland Solvng two-person zero-sum game by

More information

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

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

More information

Lecture 5: Probability Distributions. Random Variables

Lecture 5: Probability Distributions. Random Variables Lecture 5: Probablty Dstrbutons Random Varables Probablty Dstrbutons Dscrete Random Varables Contnuous Random Varables and ther Dstrbutons Dscrete Jont Dstrbutons Contnuous Jont Dstrbutons Independent

More information

Learning the Kernel Parameters in Kernel Minimum Distance Classifier

Learning the Kernel Parameters in Kernel Minimum Distance Classifier Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department

More information

Analysis of Continuous Beams in General

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

More information

Reducing Frame Rate for Object Tracking

Reducing Frame Rate for Object Tracking Reducng Frame Rate for Object Trackng Pavel Korshunov 1 and We Tsang Oo 2 1 Natonal Unversty of Sngapore, Sngapore 11977, pavelkor@comp.nus.edu.sg 2 Natonal Unversty of Sngapore, Sngapore 11977, oowt@comp.nus.edu.sg

More information

Machine Learning: Algorithms and Applications

Machine Learning: Algorithms and Applications 14/05/1 Machne Learnng: Algorthms and Applcatons Florano Zn Free Unversty of Bozen-Bolzano Faculty of Computer Scence Academc Year 011-01 Lecture 10: 14 May 01 Unsupervsed Learnng cont Sldes courtesy of

More information

MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION

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

More information

Private Information Retrieval (PIR)

Private Information Retrieval (PIR) 2 Levente Buttyán Problem formulaton Alce wants to obtan nformaton from a database, but she does not want the database to learn whch nformaton she wanted e.g., Alce s an nvestor queryng a stock-market

More information

Smoothing Spline ANOVA for variable screening

Smoothing Spline ANOVA for variable screening Smoothng Splne ANOVA for varable screenng a useful tool for metamodels tranng and mult-objectve optmzaton L. Rcco, E. Rgon, A. Turco Outlne RSM Introducton Possble couplng Test case MOO MOO wth Game Theory

More information

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

Synthesizer 1.0. User s Guide. A Varying Coefficient Meta. nalytic Tool. Z. Krizan Employing Microsoft Excel 2007

Synthesizer 1.0. User s Guide. A Varying Coefficient Meta. nalytic Tool. Z. Krizan Employing Microsoft Excel 2007 Syntheszer 1.0 A Varyng Coeffcent Meta Meta-Analytc nalytc Tool Employng Mcrosoft Excel 007.38.17.5 User s Gude Z. Krzan 009 Table of Contents 1. Introducton and Acknowledgments 3. Operatonal Functons

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

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,

More information

Fusion Performance Model for Distributed Tracking and Classification

Fusion Performance Model for Distributed Tracking and Classification Fuson Performance Model for Dstrbuted rackng and Classfcaton K.C. Chang and Yng Song Dept. of SEOR, School of I&E George Mason Unversty FAIRFAX, VA kchang@gmu.edu Martn Lggns Verdan Systems Dvson, Inc.

More information

TESTING AND IMPROVING LOCAL ADAPTIVE IMPORTANCE SAMPLING IN LJF LOCAL-JT IN MULTIPLY SECTIONED BAYESIAN NETWORKS

TESTING AND IMPROVING LOCAL ADAPTIVE IMPORTANCE SAMPLING IN LJF LOCAL-JT IN MULTIPLY SECTIONED BAYESIAN NETWORKS TESTING AND IMPROVING LOCAL ADAPTIVE IMPORTANCE SAMPLING IN LJF LOCAL-JT IN MULTIPLY SECTIONED BAYESIAN NETWORKS Dan Wu 1 and Sona Bhatt 2 1 School of Computer Scence Unversty of Wndsor, Wndsor, Ontaro

More information

TN348: Openlab Module - Colocalization

TN348: Openlab Module - Colocalization TN348: Openlab Module - Colocalzaton Topc The Colocalzaton module provdes the faclty to vsualze and quantfy colocalzaton between pars of mages. The Colocalzaton wndow contans a prevew of the two mages

More information

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

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

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

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

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task Proceedngs of NTCIR-6 Workshop Meetng, May 15-18, 2007, Tokyo, Japan Term Weghtng Classfcaton System Usng the Ch-square Statstc for the Classfcaton Subtask at NTCIR-6 Patent Retreval Task Kotaro Hashmoto

More information

Implementation Naïve Bayes Algorithm for Student Classification Based on Graduation Status

Implementation Naïve Bayes Algorithm for Student Classification Based on Graduation Status Internatonal Journal of Appled Busness and Informaton Systems ISSN: 2597-8993 Vol 1, No 2, September 2017, pp. 6-12 6 Implementaton Naïve Bayes Algorthm for Student Classfcaton Based on Graduaton Status

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

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

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

More information

Air Transport Demand. Ta-Hui Yang Associate Professor Department of Logistics Management National Kaohsiung First Univ. of Sci. & Tech.

Air Transport Demand. Ta-Hui Yang Associate Professor Department of Logistics Management National Kaohsiung First Univ. of Sci. & Tech. Ar Transport Demand Ta-Hu Yang Assocate Professor Department of Logstcs Management Natonal Kaohsung Frst Unv. of Sc. & Tech. 1 Ar Transport Demand Demand for ar transport between two ctes or two regons

More information

A Simple and Efficient Goal Programming Model for Computing of Fuzzy Linear Regression Parameters with Considering Outliers

A Simple and Efficient Goal Programming Model for Computing of Fuzzy Linear Regression Parameters with Considering Outliers 62626262621 Journal of Uncertan Systems Vol.5, No.1, pp.62-71, 211 Onlne at: www.us.org.u A Smple and Effcent Goal Programmng Model for Computng of Fuzzy Lnear Regresson Parameters wth Consderng Outlers

More information

Performance Evaluation of Information Retrieval Systems

Performance Evaluation of Information Retrieval Systems Why System Evaluaton? Performance Evaluaton of Informaton Retreval Systems Many sldes n ths secton are adapted from Prof. Joydeep Ghosh (UT ECE) who n turn adapted them from Prof. Dk Lee (Unv. of Scence

More information

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

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

More information

Announcements. Supervised Learning

Announcements. Supervised Learning Announcements See Chapter 5 of Duda, Hart, and Stork. Tutoral by Burge lnked to on web page. Supervsed Learnng Classfcaton wth labeled eamples. Images vectors n hgh-d space. Supervsed Learnng Labeled eamples

More information

An Image Fusion Approach Based on Segmentation Region

An Image Fusion Approach Based on Segmentation Region Rong Wang, L-Qun Gao, Shu Yang, Yu-Hua Cha, and Yan-Chun Lu An Image Fuson Approach Based On Segmentaton Regon An Image Fuson Approach Based on Segmentaton Regon Rong Wang, L-Qun Gao, Shu Yang 3, Yu-Hua

More information

Hybridization of Expectation-Maximization and K-Means Algorithms for Better Clustering Performance

Hybridization of Expectation-Maximization and K-Means Algorithms for Better Clustering Performance BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 16, No 2 Sofa 2016 Prnt ISSN: 1311-9702; Onlne ISSN: 1314-4081 DOI: 10.1515/cat-2016-0017 Hybrdzaton of Expectaton-Maxmzaton

More information

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

Classifying Acoustic Transient Signals Using Artificial Intelligence

Classifying Acoustic Transient Signals Using Artificial Intelligence Classfyng Acoustc Transent Sgnals Usng Artfcal Intellgence Steve Sutton, Unversty of North Carolna At Wlmngton (suttons@charter.net) Greg Huff, Unversty of North Carolna At Wlmngton (jgh7476@uncwl.edu)

More information

APPLICATION OF A COMPUTATIONALLY EFFICIENT GEOSTATISTICAL APPROACH TO CHARACTERIZING VARIABLY SPACED WATER-TABLE DATA

APPLICATION OF A COMPUTATIONALLY EFFICIENT GEOSTATISTICAL APPROACH TO CHARACTERIZING VARIABLY SPACED WATER-TABLE DATA RFr"W/FZD JAN 2 4 1995 OST control # 1385 John J Q U ~ M Argonne Natonal Laboratory Argonne, L 60439 Tel: 708-252-5357, Fax: 708-252-3 611 APPLCATON OF A COMPUTATONALLY EFFCENT GEOSTATSTCAL APPROACH TO

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

EXTENDED BIC CRITERION FOR MODEL SELECTION

EXTENDED BIC CRITERION FOR MODEL SELECTION IDIAP RESEARCH REPORT EXTEDED BIC CRITERIO FOR ODEL SELECTIO Itshak Lapdot Andrew orrs IDIAP-RR-0-4 Dalle olle Insttute for Perceptual Artfcal Intellgence P.O.Box 59 artgny Valas Swtzerland phone +4 7

More information

Machine Learning. K-means Algorithm

Machine Learning. K-means Algorithm Macne Learnng CS 6375 --- Sprng 2015 Gaussan Mture Model GMM pectaton Mamzaton M Acknowledgement: some sldes adopted from Crstoper Bsop Vncent Ng. 1 K-means Algortm Specal case of M Goal: represent a data

More information

Hybrid Non-Blind Color Image Watermarking

Hybrid Non-Blind Color Image Watermarking Hybrd Non-Blnd Color Image Watermarkng Ms C.N.Sujatha 1, Dr. P. Satyanarayana 2 1 Assocate Professor, Dept. of ECE, SNIST, Yamnampet, Ghatkesar Hyderabad-501301, Telangana 2 Professor, Dept. of ECE, AITS,

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

Complex Numbers. Now we also saw that if a and b were both positive then ab = a b. For a second let s forget that restriction and do the following.

Complex Numbers. Now we also saw that if a and b were both positive then ab = a b. For a second let s forget that restriction and do the following. Complex Numbers The last topc n ths secton s not really related to most of what we ve done n ths chapter, although t s somewhat related to the radcals secton as we wll see. We also won t need the materal

More information

Data Mining: Model Evaluation

Data Mining: Model Evaluation Data Mnng: Model Evaluaton Aprl 16, 2013 1 Issues: Evaluatng Classfcaton Methods Accurac classfer accurac: predctng class label predctor accurac: guessng value of predcted attrbutes Speed tme to construct

More information

Lecture 5: Multilayer Perceptrons

Lecture 5: Multilayer Perceptrons Lecture 5: Multlayer Perceptrons Roger Grosse 1 Introducton So far, we ve only talked about lnear models: lnear regresson and lnear bnary classfers. We noted that there are functons that can t be represented

More information

Parallel matrix-vector multiplication

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

More information

Skew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach

Skew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach Angle Estmaton and Correcton of Hand Wrtten, Textual and Large areas of Non-Textual Document Images: A Novel Approach D.R.Ramesh Babu Pyush M Kumat Mahesh D Dhannawat PES Insttute of Technology Research

More information

Related-Mode Attacks on CTR Encryption Mode

Related-Mode Attacks on CTR Encryption Mode Internatonal Journal of Network Securty, Vol.4, No.3, PP.282 287, May 2007 282 Related-Mode Attacks on CTR Encrypton Mode Dayn Wang, Dongda Ln, and Wenlng Wu (Correspondng author: Dayn Wang) Key Laboratory

More information

TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS. Muradaliyev A.Z.

TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS. Muradaliyev A.Z. TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS Muradalyev AZ Azerbajan Scentfc-Research and Desgn-Prospectng Insttute of Energetc AZ1012, Ave HZardab-94 E-mal:aydn_murad@yahoocom Importance of

More information

On Some Entertaining Applications of the Concept of Set in Computer Science Course

On Some Entertaining Applications of the Concept of Set in Computer Science Course On Some Entertanng Applcatons of the Concept of Set n Computer Scence Course Krasmr Yordzhev *, Hrstna Kostadnova ** * Assocate Professor Krasmr Yordzhev, Ph.D., Faculty of Mathematcs and Natural Scences,

More information

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

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

More information

An Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices

An Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices Internatonal Mathematcal Forum, Vol 7, 2012, no 52, 2549-2554 An Applcaton of the Dulmage-Mendelsohn Decomposton to Sparse Null Space Bases of Full Row Rank Matrces Mostafa Khorramzadeh Department of Mathematcal

More information

A Topology-aware Random Walk

A Topology-aware Random Walk A Topology-aware Random Walk Inkwan Yu, Rchard Newman Dept. of CISE, Unversty of Florda, Ganesvlle, Florda, USA Abstract When a graph can be decomposed nto clusters of well connected subgraphs, t s possble

More information

Feature-Based Matrix Factorization

Feature-Based Matrix Factorization Feature-Based Matrx Factorzaton arxv:1109.2271v3 [cs.ai] 29 Dec 2011 Tanq Chen, Zhao Zheng, Quxa Lu, Wenan Zhang, Yong Yu {tqchen,zhengzhao,luquxa,wnzhang,yyu}@apex.stu.edu.cn Apex Data & Knowledge Management

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

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

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

More information

Optimal Workload-based Weighted Wavelet Synopses

Optimal Workload-based Weighted Wavelet Synopses Optmal Workload-based Weghted Wavelet Synopses Yoss Matas School of Computer Scence Tel Avv Unversty Tel Avv 69978, Israel matas@tau.ac.l Danel Urel School of Computer Scence Tel Avv Unversty Tel Avv 69978,

More information

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 An Iteratve Soluton Approach to Process Plant Layout usng Mxed

More information

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

Load-Balanced Anycast Routing

Load-Balanced Anycast Routing Load-Balanced Anycast Routng Chng-Yu Ln, Jung-Hua Lo, and Sy-Yen Kuo Department of Electrcal Engneerng atonal Tawan Unversty, Tape, Tawan sykuo@cc.ee.ntu.edu.tw Abstract For fault-tolerance and load-balance

More information

A Comprehensive Theoretical Framework for Privacy Preserving Distributed OLAP

A Comprehensive Theoretical Framework for Privacy Preserving Distributed OLAP Purdue Unversty Purdue e-pubs Cyber Center Publcatons Cyber Center 4 A Comprehensve Theoretcal ramework for Prvacy Preservng Dstrbuted OLAP Alfredo Cuzzocrea Unversty of Calabra, Cosenza, Italy, cuzzocrea@s.des.uncal.t

More information

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

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

More information

Load Balancing for Hex-Cell Interconnection Network

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

More information

Hermite Splines in Lie Groups as Products of Geodesics

Hermite Splines in Lie Groups as Products of Geodesics Hermte Splnes n Le Groups as Products of Geodescs Ethan Eade Updated May 28, 2017 1 Introducton 1.1 Goal Ths document defnes a curve n the Le group G parametrzed by tme and by structural parameters n the

More information

Analysis of Collaborative Distributed Admission Control in x Networks

Analysis of Collaborative Distributed Admission Control in x Networks 1 Analyss of Collaboratve Dstrbuted Admsson Control n 82.11x Networks Thnh Nguyen, Member, IEEE, Ken Nguyen, Member, IEEE, Lnha He, Member, IEEE, Abstract Wth the recent surge of wreless home networks,

More information

Array transposition in CUDA shared memory

Array transposition in CUDA shared memory Array transposton n CUDA shared memory Mke Gles February 19, 2014 Abstract Ths short note s nspred by some code wrtten by Jeremy Appleyard for the transposton of data through shared memory. I had some

More information

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,

More information