Ontology Mapping: As a Binary Classification Problem
|
|
- Phebe Blankenship
- 5 years ago
- Views:
Transcription
1 Fourth Internatonal Conference on Semantcs, Knowledge and Grd Ontology Mappng: As a Bnary Classfcaton Problem Mng Mao SAP Research mng.mao@sap.com Yefe Peng Yahoo! ypeng@yahoo-nc.com Mchael Sprng U. of Pttsburgh sprng@ptt.edu Abstract Ontology mappng seeks to fnd semantc correspondences between smlar elements of dfferent ontologes. Ontology mappng s crtcal to acheve semantc nteroperablty n the WWW. To solve the ontology mappng problem, ths paper proposes a non-nstance learnng-based approach that transforms the ontology mappng problem to a bnary classfcaton problem and utlzes machne learnng technques as a soluton. Same as other machne learnng based approaches, a number of features (.e., lngustc, structural and web features) are generated for each mappng canddate. However, n contrast to other learnng-based mappng approaches, the features proposed n our approach are generc and do not rely on the exstence and suffcency of nstances. Therefore our approach can be generalzed to dfferent domans wthout extra tranng efforts. To evaluate our approach, two experments (.e., wthn-task vs. cross-task) are mplemented and the SVM algorthm s appled. Expermental results show that our non-nstance learnng-based ontology mappng approach performs well on most of OAEI benchmark tests when tranng and testng on the same mappng task; and the results of approach vary accordng to the lkelhood of tranng data and testng data when tranng and testng on dfferent mappng tasks.. Introducton The World Wde Web (WWW) s wdely used as a unversal medum for nformaton exchange. However, semantc nteroperablty n the WWW s stll lmted due to the heterogenety of nformaton. Ontology, a formal, explct specfcaton of a shared conceptualzaton [5], has been suggested as a way to solve the problem. Wth the popularty of ontologes, ontology mappng that ams to fnd semantc correspondences between smlar elements of dfferent ontologes has attracted many research attentons from varous domans. Dfferent technques have been examned n ontology mappng, e.g., usng lngustc technques to measure the lexcal smlarty of concepts n ontologes [3], treatng ontologes as structural graphs [0], takng the advantage of nformaton retreval technques [9], applyng heurstc rules to look for specfc mappng patterns [6], and learnng to map ontologes through machne learnng technques [2][3]. Comprehensve surveys of ontology mappng approaches can be found n [4][8]. Prevous learnng-based approaches have acheved hgh accuracy n predcton of correct mappngs n the cases reported n [2][3]. However the approaches ether have a lmtaton that t heavly reles on the avalablty of nstance data when measurng the smlarty of classes/attrbutes, or requre new tranng data to rebuld ther model when doman changes and thus restrct the unversalty of the model. To overcome the lmtatons, we treat the ontology mappng problem as a bnary classfcaton problem. We learn a generc mappng model, whch does not requre the exstence of nstances and doman constrants. To learn a model, a varety of features that can reflect the characterstcs of mappng pars are generated, and then the SVM algorthm s appled. Expermental results show that our nonnstance learnng-based ontology mappng approach performs well n most of OAEI benchmark tests when tranng and testng on the same mappng task; and the results of approach vary accordng to the lkelhood of tranng data and testng data when tranng and testng on dfferent mappng tasks. 2. Problem Statement Ontology s a formal, explct specfcaton of a shared conceptualzaton n terms of classes, attrbutes and relatons [5]. Ontologes are typcally represented as taxonomc trees that nclude classes, propertes, and relatons, and assocated wth nstances. Two sample bblographc ontologes are /08 $ IEEE DOI 0.09/SKG
2 shown n Fg., n whch the ellpses ndcate classes (e.g., "Reference"), the dashed rectangles ndcate propertes (e.g., "publsher"), the lnes wth arrowhead ndcate "subclassof" relaton between two classes, and the sold rectangles ndcate nstances of class (e.g., "Obect-orented data modelng"). Each class and property can also have descrptve nformaton (e.g., ID, label, comment) and restrctons (e.g., ttle, publsher) as ndcated n the brace next to "Book". Proceedng Monograph Collecton -press -event -edtor -organzaton Reference ID: Book Label: Book Comments: A book subclassof that may be a monograph or a collecton of wrtten Book texts. Restrctons: ttle, publsher, author, edton, date etc. subclassof subclassof Fg.. Two sample bblographc ontologes The process of ontology mappng s to fnd semantc correspondences between smlar elements n two homogeneous ontologes, and many ways can be used to udge the qualty of a mappng result. In ths paper, we refer to the "correspondence" as an "=" relatonshp, the "elements" as "classes" and "propertes" of an ontology, and we udge the mappng result by ts correctness,.e., ether correct or ncorrect, whch can be depcted as a bnary set {+,-}. Therefore the ontology mappng problem can be easly transformed as a bnary classfcaton problem represented as followng statement: m( e, e2, r) {, where e s element e from ontology O, e 2 s element e from ontology O 2, and r s the mappng relaton (.e. correspondence) between e and e 2. Accordng to the statement, canddate mappngs n Fg. can be evaluated as followngs: m(book rght, Book left, =) {+}, m(proceedngs, Proc, =) {+}, m(monograph, Monography, =) {+}, m(proceedngs, Talks, =) {-}, m(proceedngs, Monography, =) {-}, etc. 3. Our Approach 3. Overvew -Obect-Orented Data Modelng -Publshed by MIT Press -Year 2000 Composte subclassof Monography Collecton The nsght of our approach s to treat ontology mappng problem as a bnary classfcaton problem, Proc. -publsher -edtor -organzaton subclassof Book } ID: Book Label: Book Comments: A book that may be a monography or a collecton of wrtten texts. Restrctons: humancreator, edton and thus we can take advantage of machne learnng technques. Generally speakng, our approach has 5 steps, see detaled descrpton n 4.4. and Generate varous doman ndependent features (.e., lngustc, structural and web features) to descrbe the characterstcs of ontologes. 2. Randomly generate tranng and testng set for OAEI benchmark tests. 3. Tran a SVM model on tranng set. 4. Classfy testng data on the traned SVM model. 5. Extract mappng results of testng data usng naïve descendant extracton algorthm []. 6. Evaluate testng data aganst ground truth. 7. Fnally, repeat step tmes and get the average evaluaton result to elmnate bas. 3.2 Feature Generaton Applyng machne learnng technques to ontology mappng context rases the queston of what types of nformaton should be used n the learnng process. Many dfferent types of nformaton can contrbute toward decdng the correspondence of a mappng par. Two prncples are followed to select features: The feature should not be lmted to nstances. It could be generated from classes, propertes and/or nstances n ontologes. The feature should be general enough and doman ndependent so that the model could be generalzed to other applcatons regardless of the varety of doman. In the approach, 3 categores,.e., lngustc features, structural features and web features, and total 26 features are generated for each mappng par Lngustc Features Lngustc features are selected accordng to the prncple descrbed n [7]. Totally 6 lngustc features are generated, whch can be dvded nto two types (We do not lst all lngustc features due to the space lmt): Isolated characterstcs of elements n mappng par, e.g. length of elements, number of tokens, etc. 2 Syntactc characterstcs of mappng par, e.g. (normalzed) length dfference between elements, Levensthten edt dstance between two elements, the proporton of word change between elements, number of common tokens n the par, the cosne smlarty of the profle [9][0] of elements, etc. 2
3 3.2.2 Web Features Bollegala, Matsuo et al. [] proposed a page count based co-occurrence measure.e., WebDce, to compute semantc smlarty, whch s defned as followng, where the notaton H(X) and H(Y) denote the page counts for query X and Y respectvely n a search engne, H(XY) denotes the page counts for the conuncton query X AND Y, c s a predefned threshold (e.g. c=5) to reduce the adverse effects caused by random co-occurrences. 0 WebDce( X, Y) 2H ( X Y ) H ( X ) H ( Y) f H(X Y) c otherwse Structural Features Structural nformaton s mportant n estmatng the smlarty of ontologes. Table lsts the structural features of a mappng canddate. 4. Evaluatons 4. Test Ontologes Our test ontologes are OAEI benchmark tests ontologes, orgnatng from the bblography doman. The OAEI benchmark tests nclude one reference ontology O R dedcated to the very narrow doman of bblography, multple test ontologes O T manually dscardng varous nformaton from the reference ontology n order to evaluate how algorthms behave when nformaton s lackng, and 4 real world bblographc ontologes that are generated by MIT 2, UMBC 3, Unversty of Karlsruhe 4 and INRIA 5 respectvely. The OAEI benchmark tests are open tests, whch mean the expected results are provded for all partcpants. 4.2 Evaluaton Crtera We follow the evaluaton crtera used by the OAEI ontology matchng campagn That s, standard nformaton retreval evaluaton measures,.e., precson, recall and f-measure, are computed aganst the reference algnment. The precson, recall and f- measure are defned as follows lpes.exmo.rdf.bb.owl Precson Recall F-measure # correct _ found _ mappngs p # all _ found _ mappngs # correct _ found _ mappngs r # all _ possble_ mappngs p r f 2 p r 4.3 Expermental Desgn Motvaton Two experments were desgned. The motvaton of them s: The st experment nvestgates how the approach performs n the stuaton where people have manually marked some mappng results for a specfc mappng task, but they need help from automatc mappng tools to fnd the rest of mappngs. The 2 nd experment nvestgates whether a model traned on one mappng task can work on another mappng task(s). Moreover, we are nterested n whch benchmark test(s) s more sutable as a tranng model. The motvaton for the 2 nd experment s: n most ontology mappng cases, no ground truth s avalable for a specfc mappng task, but a general model has been learned that can be used to fnd mappngs. Thus, to save users tme and effort, we want to fnd out mappng results usng the exstng model. 4.4 Expermental Methodology and Results 4.4. st Experment Wthn-task The methodology of the st experment s:. For each OAEI benchmark test, we generate canddate mappng pars by smply combne all elements from two ontologes. 2. For each mappng canddate, we mark down ther correctness accordng to the reference algnment (.e. the ground truth). Smultaneously we generate varous features (.e., lngustc, structural and web features) to descrbe the characterstcs of the mappng par. 3. We splt all mappng pars nto two groups (.e., one s for tranng purpose and the other s used as testng set) by randomly choosng (e.g. 50% vs. 50%). We tran two SVM models (.e., SVM- Class and SVM-Property) on tranng set usng SVM-Lght package
4 Table. Structural features Elements Features Descrpton DrPropNumDff The normalzed dfference between the numbers of the classes drect propertes The edt dstance based smlarty between the classes drect propertes,.e., DrPropSm DrPropSm Avg (max( EdtDstSm( p, p2 ))), where p and p 2 are drect propertes of class C and C 2. chnumdff The normalzed dfference between the numbers of the classes subclasses. The edt dstance based smlarty between the classes subclasses,.e., Classes chsm chsm Avg (max( EdtDstSm( subc, subc2 ))), where subc and subc 2 are subclasses of class C and C 2. The edt dstance based smlarty between the classes super classes,.e., pasm pasm Avg (max( EdtDstSm( pac, pac2 ))), where pac and pac 2 are super classes of class C and C 2. depdff The normalzed dfference between the depth to root of the classes domansm The edt dstance based smlarty between the propertes doman Propertes rangesm The edt dstance based smlarty between the propertes range mothersm The edt dstance based smlarty between the propertes mother class 4. We classfy testng data on two models.. On Test #0-#04 and #22-#247, both SVM- 5. We extract mappng results of testng data usng naïve descendant extracton algorthm [] and evaluate the results aganst reference algnment. Class model and SVM-Property model perform as well as PRIOR+. Ths s because the lngustc nformaton of these test ontologes s hghly 6. Fnally to elmnate the bas caused by randomly choosng mappng pars to generate tranng and testng data n step 3, we repeat step tmes and report the average result as our fnal result. In the experment, two SVM models (.e., SVMsmlar wth that of the reference ontology and there s much less nterference such as randomly generated name of classes/propertes. Thus t s easy for both SVM-Class and SVM-Property model to catch useful features lke edt dstance Class model for classes and SVM-Property model for that can contrbute to learnng models. propertes) are traned separately due to the dfference between the structure of classes and propertes. As a result, the mappng pars of classes are tested on SVM-Class model and the mappng pars of propertes are tested on SVM-Property model. Moreover, snce the number of negatve examples s much larger than the number of postve examples n tranng data, we use a fxed cost factor 2. On Test #20-#20, both SVM-Class and SVM- Property model perform relatvely worse than the PRIOR+ (especally on #20, #202, #208, #209). Ths s because the lngustc nformaton changes too much on these tests so that t s hard to catch ts lngustc and web characterstcs n the tranng model. Meanwhle the structural feature s relatvely weak. (.e. 0) n SVM-Lght to equalze the dstrbuton and ensure tranng errors on postve examples outwegh those on negatve examples. Fg. 2 shows the average f-measure of classes of each OAEI benchmark task tested on SVM-Class model. Fg. 3 shows the f-measure of propertes of each OAEI benchmark task tested on SVM-Property model, n whch the f-measures of benchmark tests 3. On Test #248-#266, both SVM-Class and SVM- Property model perform much worse than the PRIOR+. Ths s because there s no name and no comments n the test ontologes at all,.e., both lngustc features and web features are totally unavalable. The only feature avalable for SVM models s structural, whch s relatvely weak. Meanwhle, the PRIOR+ benefts from the #226, #233-#237, #240-#247, #250, #254-#257, profle enrchment process that ntegrates #260-#266 are 0 s because there s no property exstng for those tests. For comparson purpose both nstance nformaton, whch keeps all descrptve nformaton, to both classes and propertes. Fg 2 and 3 nclude the f-measure of 4. On real world cases #30-304, the SVM-Class classes/propertes runnng by PRIOR+ approach [0], a non learnng based ontology mappng approach. The observatons from Fg. 2 and 3 are: model performs much better than the PRIOR+ and the SVM-Property model performs smlarly as the PRIOR+ (.e., slghtly better on #30 and #302 but slghtly worse on #303 and #304). The reason s our learnng based approach utlzes 23
5 SVM-Class Pror+ Class SVM-Property Pror+ Property F-Measure F-Measure Fg. 2. Results of classes on SVM-Class model on all benchmark tests (Wthn-task) Web feature to explore synonymous relatons between concepts n ontologes. By contrast the PRIOR+ approach does not ntegrate any auxlary thesaurus for such a purpose. Our concluson s: For learnng-based approach (wthn-task), the performance s good when mappng task s relatvely easy (.e., #xx and #22-247). When mappng task s more dffcult, ts performance s not as good as the PRIOR+ approach (.e., #20- #20 and #248-#266). But the performance of ths approach s better than the PRIOR+ on real world cases, whch shows the features used n ths approach make more sense on real world cases than on artfcally constructed cases nd Experment Cross-task The methodology of the 2 nd experment s:. Same as step n st experment. 2. Same as step 2 n st experment. 3. We tran two SVM models (.e. SVM-Class and SVM-Property for each benchmark mappng task, except #228, #233, #236, #239-#247, #250, #254, #257, and #260-#266, usng SVM-Lght package. Ths s because no propertes exst n these test ontologes, and thus no SVM-Property model can be traned on them. And thus t does not make sense to test mappng tasks wth both classes and propertes on the model traned wthout property. 4. We classfy testng data of all the other benchmark tests (excludng the one that has been used n tranng model) usng the SVM models. 5. We extract mappng results of usng naïve descendant extracton algorthm and evaluate the results aganst the reference algnment. 6. Fnally we repeat step tmes and report the average f-measure of a group of testng data (e.g., #xx, #2xx, #3xx etc.) on each tranng model as our fnal result. Fg. 4 shows the average f-measure tested on dfferent data sets (.e., all tests, #xx, #2xx, #3xx, Fg. 3. Results of propertes on SVM-Property model on all benchmark tests (Wthn-task) and more specfc #20-#20, #22-#238, #248- #259). Our concluson s: For learnng-based crosstask approach, the performance s good when tranng data and testng data share smlar characterstcs. If the testng mappng task s very smple, t's easy to catch characterstcs n the tranng model and thus get good performance wth more dffcult tranng task. Meanwhle f both tranng and testng tasks are dffcult but wth dfferent characterstcs, the performance s not as good as other approaches. 5. Related Work Dfferent approaches have been explored to solve ontology mappng problem, among whch machne learnng based method s effcent when the concepts n ontologes are assocated wth many nstances, and t works better f many value of nstances are text rather than references to other nstances. In GLUE [2], a well-known machne learnng based ontology mappng system, to measure the smlarty of concepts the author needs to calculate the ont probablty dstrbuton of the concepts that heavly rely on the avalablty of nstance. However, n most cases nstances are ust unavalable or nsuffcent, and t s more common to have references between nstances than text descrpton. Furthermore, the target of the GLUE s every element n the target ontology, whch makes the model unable to be generalzed to any applcaton where doman has changed. Therefore they need new tranng data to rebuld the model for each doman, whch s usually unavalable. Another approach usng machne learnng technques for ontology mappng s QOM [3]. In QOM, the authors frst calculate varous smlartes based on expert encoded rules, and then they use neural network to ntegrate all these smlarty measures. In the contrast, the features we use are not lmted to the varety of smlartes
6 .2 xx xx 3xx all 0.8 F-Measure Conclusons and Future Work In ths paper, we examned a non-nstance learnngbased ontology mappng approach, whch overcomes the lmtatons of prevous learnng-based ontology mappng approaches that ether rely on the avalablty of suffcent nstances or are domandependent. In the approach we treated the ontology mappng problem as a bnary classfcaton problem; generated a number of generc features; utlzed these features to buld tranng model; and conducted two experments to nvestgate the performance of machne learnng technques n dfferent stuatons. The experment results show that our approach performs well on most of OAEI benchmark tests when tranng and testng on the same mappng task; and the results of approach vary accordng to the lkelhood of tranng data and testng data when tranng and testng on dfferent mappng tasks. Future work may nclude: Leverage dfferent features so as to acheve a robust semantc smlarty measure, do feature selecton procedure by maxmzng the f-measure, and perform an actve learnng wth Support Vector Machne algorthm. 7. References. Bollegala, D., Matsuo, Y. and Ishzuka, M. (2007) Measurng Semantc Smlarty between Words Usng Web Search Engnes. Proceedngs of the Internatonal World Wde Web Conference, Banff, Canada Doan, A., J. Madhaven, et al. (2003). "Learnng to Match Ontologes on the Semantc Web." VLDB Journal 2(4): Ehrg, M. and S. Staab. QOM: Quck Ontology Mappng. In the Proceedngs of the 3rd Internatonal Semantc Web Conference (ISWC) pror+ Tranng Set Fg. 4. Testng results of benchmark tests (Cross-task) 4. Euzenat, J., Bach, T., et al. State of the art on ontology algnment, Knowledge web NoE Gruber, T. "A Translaton Approach to Portable Ontology Specfcatons." Knowledge Acquston 5(2): Hovy, E. Combnng and standardzng large-scale, practcal ontologes for machne translaton and other uses. In Proceedngs of the st Internatonal Conference on Language Resources and Evaluaton (LREC), Granada, Span R. Jones, B. Rey et. al. Generatng Query Substtutons. In WWW '06: Proceedngs of the 5th nternatonal conference on World Wde Web. New York, NY, USA, Kalfoglou, Y. and M. Schorlemmer "Ontology mappng: the state of the art." Knowledge Engneerng Revew 8():-3 9. Mao, M. Ontology Mappng: An Informaton Retreval and Interactve Actvaton Network Based Approach. In Proceedng of ISWC 2007, LNCS 4825, pp , Mao, M. and Peng, Y. PRIOR+ System: Results for OAEI In Proceedngs of ISWC 2007 Ontology Matchng Workshop Melcke, C. and Stuckenschmdt, H. Analyzng Mappng Extracton Approaches. In Proceedngs of ISWC 2007 Ontology Matchng Workshop. Busan, Korea. 2. Melnk, S., H. Garca-Molna, et al. Smlarty floodng: a versatle graph matchng algorthm and ts applcaton to schema matchng. Proc. 8th Internatonal Conference on Data Engneerng (ICDE) Qu, Y., Hu, W., and Cheng, G. Constructng vrtual documents for ontology matchng. In Proceedngs of the 5th Internatonal Conference on World Wde Web
Alignment Results of SOBOM for OAEI 2010
Algnment Results of SOBOM for OAEI 2010 Pegang Xu, Yadong Wang, Lang Cheng, Tany Zang School of Computer Scence and Technology Harbn Insttute of Technology, Harbn, Chna pegang.xu@gmal.com, ydwang@ht.edu.cn,
More informationTerm 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 informationAn 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 informationContent 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 informationTsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance
Tsnghua Unversty at TAC 2009: Summarzng Mult-documents by Informaton Dstance Chong Long, Mnle Huang, Xaoyan Zhu State Key Laboratory of Intellgent Technology and Systems, Tsnghua Natonal Laboratory for
More informationBOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET
1 BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET TZU-CHENG CHUANG School of Electrcal and Computer Engneerng, Purdue Unversty, West Lafayette, Indana 47907 SAUL B. GELFAND School
More informationLearning 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 informationCluster Analysis of Electrical Behavior
Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School
More informationA Binarization Algorithm specialized on Document Images and Photos
A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a
More informationUser Authentication Based On Behavioral Mouse Dynamics Biometrics
User Authentcaton Based On Behavoral Mouse Dynamcs Bometrcs Chee-Hyung Yoon Danel Donghyun Km Department of Computer Scence Department of Computer Scence Stanford Unversty Stanford Unversty Stanford, CA
More informationFEATURE 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 informationQuery Clustering Using a Hybrid Query Similarity Measure
Query clusterng usng a hybrd query smlarty measure Fu. L., Goh, D.H., & Foo, S. (2004). WSEAS Transacton on Computers, 3(3), 700-705. Query Clusterng Usng a Hybrd Query Smlarty Measure Ln Fu, Don Hoe-Lan
More informationSubspace 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 informationImprovement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration
Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,
More informationA Background Subtraction for a Vision-based User Interface *
A Background Subtracton for a Vson-based User Interface * Dongpyo Hong and Woontack Woo KJIST U-VR Lab. {dhon wwoo}@kjst.ac.kr Abstract In ths paper, we propose a robust and effcent background subtracton
More informationOutline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1
4/14/011 Outlne Dscrmnatve classfers for mage recognton Wednesday, Aprl 13 Krsten Grauman UT-Austn Last tme: wndow-based generc obect detecton basc ppelne face detecton wth boostng as case study Today:
More informationSupport 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 informationSupport Vector Machines
Support Vector Machnes Decson surface s a hyperplane (lne n 2D) n feature space (smlar to the Perceptron) Arguably, the most mportant recent dscovery n machne learnng In a nutshell: map the data to a predetermned
More informationFuzzy C-Means Initialized by Fixed Threshold Clustering for Improving Image Retrieval
Fuzzy -Means Intalzed by Fxed Threshold lusterng for Improvng Image Retreval NAWARA HANSIRI, SIRIPORN SUPRATID,HOM KIMPAN 3 Faculty of Informaton Technology Rangst Unversty Muang-Ake, Paholyotn Road, Patumtan,
More informationClassifier 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 informationAn 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 informationPerformance Evaluation of Information Retrieval Systems
Why System Evaluaton? Performance Evaluaton of Informaton Retreval Systems Many sldes n ths secton are adapted from Prof. Joydeep Ghosh (UT ECE) who n turn adapted them from Prof. Dk Lee (Unv. of Scence
More informationCSCI 5417 Information Retrieval Systems Jim Martin!
CSCI 5417 Informaton Retreval Systems Jm Martn! Lecture 11 9/29/2011 Today 9/29 Classfcaton Naïve Bayes classfcaton Ungram LM 1 Where we are... Bascs of ad hoc retreval Indexng Term weghtng/scorng Cosne
More informationA 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 informationParallelism for Nested Loops with Non-uniform and Flow Dependences
Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr
More informationUB at GeoCLEF Department of Geography Abstract
UB at GeoCLEF 2006 Mguel E. Ruz (1), Stuart Shapro (2), June Abbas (1), Slva B. Southwck (1) and Davd Mark (3) State Unversty of New York at Buffalo (1) Department of Lbrary and Informaton Studes (2) Department
More informationCORE: A Tool for Collaborative Ontology Reuse and Evaluation
CRE: A Tool for Collaboratve ntology Reuse and Evaluaton Mram Fernández, Iván Cantador, Pablo Castells Escuela Poltécnca Superor Unversdad Autónoma de Madrd Campus de Cantoblanco, 28049 Madrd, Span {mram.fernandez,
More informationPruning Training Corpus to Speedup Text Classification 1
Prunng Tranng Corpus to Speedup Text Classfcaton Jhong Guan and Shugeng Zhou School of Computer Scence, Wuhan Unversty, Wuhan, 430079, Chna hguan@wtusm.edu.cn State Key Lab of Software Engneerng, Wuhan
More informationCollaboratively Regularized Nearest Points for Set Based Recognition
Academc Center for Computng and Meda Studes, Kyoto Unversty Collaboratvely Regularzed Nearest Ponts for Set Based Recognton Yang Wu, Mchhko Mnoh, Masayuk Mukunok Kyoto Unversty 9/1/013 BMVC 013 @ Brstol,
More informationVirtual Machine Migration based on Trust Measurement of Computer Node
Appled Mechancs and Materals Onlne: 2014-04-04 ISSN: 1662-7482, Vols. 536-537, pp 678-682 do:10.4028/www.scentfc.net/amm.536-537.678 2014 Trans Tech Publcatons, Swtzerland Vrtual Machne Mgraton based on
More informationMachine Learning 9. week
Machne Learnng 9. week Mappng Concept Radal Bass Functons (RBF) RBF Networks 1 Mappng It s probably the best scenaro for the classfcaton of two dataset s to separate them lnearly. As you see n the below
More informationTECHNIQUE 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 informationFeature Selection as an Improving Step for Decision Tree Construction
2009 Internatonal Conference on Machne Learnng and Computng IPCSIT vol.3 (2011) (2011) IACSIT Press, Sngapore Feature Selecton as an Improvng Step for Decson Tree Constructon Mahd Esmael 1, Fazekas Gabor
More informationA 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 informationA Unified Framework for Semantics and Feature Based Relevance Feedback in Image Retrieval Systems
A Unfed Framework for Semantcs and Feature Based Relevance Feedback n Image Retreval Systems Ye Lu *, Chunhu Hu 2, Xngquan Zhu 3*, HongJang Zhang 2, Qang Yang * School of Computng Scence Smon Fraser Unversty
More informationAssociative Based Classification Algorithm For Diabetes Disease Prediction
Internatonal Journal of Engneerng Trends and Technology (IJETT) Volume-41 Number-3 - November 016 Assocatve Based Classfcaton Algorthm For Dabetes Dsease Predcton 1 N. Gnana Deepka, Y.surekha, 3 G.Laltha
More informationType-2 Fuzzy Non-uniform Rational B-spline Model with Type-2 Fuzzy Data
Malaysan Journal of Mathematcal Scences 11(S) Aprl : 35 46 (2017) Specal Issue: The 2nd Internatonal Conference and Workshop on Mathematcal Analyss (ICWOMA 2016) MALAYSIAN JOURNAL OF MATHEMATICAL SCIENCES
More informationQuick error verification of portable coordinate measuring arm
Quck error verfcaton of portable coordnate measurng arm J.F. Ouang, W.L. Lu, X.H. Qu State Ke Laborator of Precson Measurng Technolog and Instruments, Tanjn Unverst, Tanjn 7, Chna Tel.: + 86 [] 7-8-99
More informationAvailable online at Available online at Advanced in Control Engineering and Information Science
Avalable onlne at wwwscencedrectcom Avalable onlne at wwwscencedrectcom Proceda Proceda Engneerng Engneerng 00 (2011) 15000 000 (2011) 1642 1646 Proceda Engneerng wwwelsevercom/locate/proceda Advanced
More informationMachine Learning: Algorithms and Applications
14/05/1 Machne Learnng: Algorthms and Applcatons Florano Zn Free Unversty of Bozen-Bolzano Faculty of Computer Scence Academc Year 011-01 Lecture 10: 14 May 01 Unsupervsed Learnng cont Sldes courtesy of
More informationBAYESIAN MULTI-SOURCE DOMAIN ADAPTATION
BAYESIAN MULTI-SOURCE DOMAIN ADAPTATION SHI-LIANG SUN, HONG-LEI SHI Department of Computer Scence and Technology, East Chna Normal Unversty 500 Dongchuan Road, Shangha 200241, P. R. Chna E-MAIL: slsun@cs.ecnu.edu.cn,
More informationDeep Classification in Large-scale Text Hierarchies
Deep Classfcaton n Large-scale Text Herarches Gu-Rong Xue Dkan Xng Qang Yang 2 Yong Yu Dept. of Computer Scence and Engneerng Shangha Jao-Tong Unversty {grxue, dkxng, yyu}@apex.sjtu.edu.cn 2 Hong Kong
More informationSolving two-person zero-sum game by Matlab
Appled Mechancs and Materals Onlne: 2011-02-02 ISSN: 1662-7482, Vols. 50-51, pp 262-265 do:10.4028/www.scentfc.net/amm.50-51.262 2011 Trans Tech Publcatons, Swtzerland Solvng two-person zero-sum game by
More informationOptimizing Document Scoring for Query Retrieval
Optmzng Document Scorng for Query Retreval Brent Ellwen baellwe@cs.stanford.edu Abstract The goal of ths project was to automate the process of tunng a document query engne. Specfcally, I used machne learnng
More informationEnhancement of Infrequent Purchased Product Recommendation Using Data Mining Techniques
Enhancement of Infrequent Purchased Product Recommendaton Usng Data Mnng Technques Noraswalza Abdullah, Yue Xu, Shlomo Geva, and Mark Loo Dscplne of Computer Scence Faculty of Scence and Technology Queensland
More informationA 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 informationThree supervised learning methods on pen digits character recognition dataset
Three supervsed learnng methods on pen dgts character recognton dataset Chrs Flezach Department of Computer Scence and Engneerng Unversty of Calforna, San Dego San Dego, CA 92093 cflezac@cs.ucsd.edu Satoru
More informationLoad Balancing for Hex-Cell Interconnection Network
Int. J. Communcatons, Network and System Scences,,, - Publshed Onlne Aprl n ScRes. http://www.scrp.org/journal/jcns http://dx.do.org/./jcns.. Load Balancng for Hex-Cell Interconnecton Network Saher Manaseer,
More informationModule Management Tool in Software Development Organizations
Journal of Computer Scence (5): 8-, 7 ISSN 59-66 7 Scence Publcatons Management Tool n Software Development Organzatons Ahmad A. Al-Rababah and Mohammad A. Al-Rababah Faculty of IT, Al-Ahlyyah Amman Unversty,
More informationQuerying by sketch geographical databases. Yu Han 1, a *
4th Internatonal Conference on Sensors, Measurement and Intellgent Materals (ICSMIM 2015) Queryng by sketch geographcal databases Yu Han 1, a * 1 Department of Basc Courses, Shenyang Insttute of Artllery,
More informationCompiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz
Compler Desgn Sprng 2014 Regster Allocaton Sample Exercses and Solutons Prof. Pedro C. Dnz USC / Informaton Scences Insttute 4676 Admralty Way, Sute 1001 Marna del Rey, Calforna 90292 pedro@s.edu Regster
More informationLecture 5: Multilayer Perceptrons
Lecture 5: Multlayer Perceptrons Roger Grosse 1 Introducton So far, we ve only talked about lnear models: lnear regresson and lnear bnary classfers. We noted that there are functons that can t be represented
More informationMathematics 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 informationMULTISPECTRAL 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 informationLoad-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 informationHelsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr)
Helsnk Unversty Of Technology, Systems Analyss Laboratory Mat-2.08 Independent research projects n appled mathematcs (3 cr) "! #$&% Antt Laukkanen 506 R ajlaukka@cc.hut.f 2 Introducton...3 2 Multattrbute
More informationA New Approach For the Ranking of Fuzzy Sets With Different Heights
New pproach For the ankng of Fuzzy Sets Wth Dfferent Heghts Pushpnder Sngh School of Mathematcs Computer pplcatons Thapar Unversty, Patala-7 00 Inda pushpndersnl@gmalcom STCT ankng of fuzzy sets plays
More informationCS 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 informationLearning-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 informationNetwork Intrusion Detection Based on PSO-SVM
TELKOMNIKA Indonesan Journal of Electrcal Engneerng Vol.1, No., February 014, pp. 150 ~ 1508 DOI: http://dx.do.org/10.11591/telkomnka.v1.386 150 Network Intruson Detecton Based on PSO-SVM Changsheng Xang*
More informationImproving Web Image Search using Meta Re-rankers
VOLUME-1, ISSUE-V (Aug-Sep 2013) IS NOW AVAILABLE AT: www.dcst.com Improvng Web Image Search usng Meta Re-rankers B.Kavtha 1, N. Suata 2 1 Department of Computer Scence and Engneerng, Chtanya Bharath Insttute
More informationSimulation: Solving Dynamic Models ABE 5646 Week 11 Chapter 2, Spring 2010
Smulaton: Solvng Dynamc Models ABE 5646 Week Chapter 2, Sprng 200 Week Descrpton Readng Materal Mar 5- Mar 9 Evaluatng [Crop] Models Comparng a model wth data - Graphcal, errors - Measures of agreement
More information11. HARMS How To: CSV Import
and Rsk System 11. How To: CSV Import Preparng the spreadsheet for CSV Import Refer to the spreadsheet template to ad algnng spreadsheet columns wth Data Felds. The spreadsheet s shown n the Appendx, an
More informationOntology Generator from Relational Database Based on Jena
Computer and Informaton Scence Vol. 3, No. 2; May 2010 Ontology Generator from Relatonal Database Based on Jena Shufeng Zhou (Correspondng author) College of Mathematcs Scence, Laocheng Unversty No.34
More informationDescription of NTU Approach to NTCIR3 Multilingual Information Retrieval
Proceedngs of the Thrd NTCIR Workshop Descrpton of NTU Approach to NTCIR3 Multlngual Informaton Retreval Wen-Cheng Ln and Hsn-Hs Chen Department of Computer Scence and Informaton Engneerng Natonal Tawan
More informationFace Recognition University at Buffalo CSE666 Lecture Slides Resources:
Face Recognton Unversty at Buffalo CSE666 Lecture Sldes Resources: http://www.face-rec.org/algorthms/ Overvew of face recognton algorthms Correlaton - Pxel based correspondence between two face mages Structural
More informationWishing 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 informationSkew 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 informationSyntactic Tree-based Relation Extraction Using a Generalization of Collins and Duffy Convolution Tree Kernel
Syntactc Tree-based Relaton Extracton Usng a Generalzaton of Collns and Duffy Convoluton Tree Kernel Mahdy Khayyaman Seyed Abolghasem Hassan Abolhassan Mrroshandel Sharf Unversty of Technology Sharf Unversty
More informationComplex 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 informationMeta-heuristics for Multidimensional Knapsack Problems
2012 4th Internatonal Conference on Computer Research and Development IPCSIT vol.39 (2012) (2012) IACSIT Press, Sngapore Meta-heurstcs for Multdmensonal Knapsack Problems Zhbao Man + Computer Scence Department,
More informationA Novel Adaptive Descriptor Algorithm for Ternary Pattern Textures
A Novel Adaptve Descrptor Algorthm for Ternary Pattern Textures Fahuan Hu 1,2, Guopng Lu 1 *, Zengwen Dong 1 1.School of Mechancal & Electrcal Engneerng, Nanchang Unversty, Nanchang, 330031, Chna; 2. School
More informationIntrinsic Plagiarism Detection Using Character n-gram Profiles
Intrnsc Plagarsm Detecton Usng Character n-gram Profles Efstathos Stamatatos Unversty of the Aegean 83200 - Karlovass, Samos, Greece stamatatos@aegean.gr Abstract: The task of ntrnsc plagarsm detecton
More informationInvestigating the Performance of Naïve- Bayes Classifiers and K- Nearest Neighbor Classifiers
Journal of Convergence Informaton Technology Volume 5, Number 2, Aprl 2010 Investgatng the Performance of Naïve- Bayes Classfers and K- Nearest Neghbor Classfers Mohammed J. Islam *, Q. M. Jonathan Wu,
More informationThe 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 informationCLASSIFICATION OF ULTRASONIC SIGNALS
The 8 th Internatonal Conference of the Slovenan Socety for Non-Destructve Testng»Applcaton of Contemporary Non-Destructve Testng n Engneerng«September -3, 5, Portorož, Slovena, pp. 7-33 CLASSIFICATION
More informationDetermining 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 informationExtraction of Fuzzy Rules from Trained Neural Network Using Evolutionary Algorithm *
Extracton of Fuzzy Rules from Traned Neural Network Usng Evolutonary Algorthm * Urszula Markowska-Kaczmar, Wojcech Trelak Wrocław Unversty of Technology, Poland kaczmar@c.pwr.wroc.pl, trelak@c.pwr.wroc.pl
More informationKIDS Lab at ImageCLEF 2012 Personal Photo Retrieval
KD Lab at mageclef 2012 Personal Photo Retreval Cha-We Ku, Been-Chan Chen, Guan-Bn Chen, L-J Gaou, Rong-ng Huang, and ao-en Wang Knowledge, nformaton, and Database ystem Laboratory Department of Computer
More informationHermite Splines in Lie Groups as Products of Geodesics
Hermte Splnes n Le Groups as Products of Geodescs Ethan Eade Updated May 28, 2017 1 Introducton 1.1 Goal Ths document defnes a curve n the Le group G parametrzed by tme and by structural parameters n the
More informationAnnouncements. 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 informationKeywords - Wep page classification; bag of words model; topic model; hierarchical classification; Support Vector Machines
(IJCSIS) Internatonal Journal of Computer Scence and Informaton Securty, Herarchcal Web Page Classfcaton Based on a Topc Model and Neghborng Pages Integraton Wongkot Srura Phayung Meesad Choochart Haruechayasak
More informationShape Representation Robust to the Sketching Order Using Distance Map and Direction Histogram
Shape Representaton Robust to the Sketchng Order Usng Dstance Map and Drecton Hstogram Department of Computer Scence Yonse Unversty Kwon Yun CONTENTS Revew Topc Proposed Method System Overvew Sketch Normalzaton
More informationImage Alignment CSC 767
Image Algnment CSC 767 Image algnment Image from http://graphcs.cs.cmu.edu/courses/15-463/2010_fall/ Image algnment: Applcatons Panorama sttchng Image algnment: Applcatons Recognton of object nstances
More informationPetri Net Based Software Dependability Engineering
Proc. RELECTRONIC 95, Budapest, pp. 181-186; October 1995 Petr Net Based Software Dependablty Engneerng Monka Hener Brandenburg Unversty of Technology Cottbus Computer Scence Insttute Postbox 101344 D-03013
More informationProblem 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 informationVisual Thesaurus for Color Image Retrieval using Self-Organizing Maps
Vsual Thesaurus for Color Image Retreval usng Self-Organzng Maps Chrstopher C. Yang and Mlo K. Yp Department of System Engneerng and Engneerng Management The Chnese Unversty of Hong Kong, Hong Kong ABSTRACT
More informationAudio Content Classification Method Research Based on Two-step Strategy
(IJACSA) Internatonal Journal of Advanced Computer Scence and Applcatons, Audo Content Classfcaton Method Research Based on Two-step Strategy Sume Lang Department of Computer Scence and Technology Chongqng
More informationSLAM 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 informationOutline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like:
Self-Organzng Maps (SOM) Turgay İBRİKÇİ, PhD. Outlne Introducton Structures of SOM SOM Archtecture Neghborhoods SOM Algorthm Examples Summary 1 2 Unsupervsed Hebban Learnng US Hebban Learnng, Cntd 3 A
More informationFrom Comparing Clusterings to Combining Clusterings
Proceedngs of the Twenty-Thrd AAAI Conference on Artfcal Intellgence (008 From Comparng Clusterngs to Combnng Clusterngs Zhwu Lu and Yuxn Peng and Janguo Xao Insttute of Computer Scence and Technology,
More informationSemantic Image Retrieval Using Region Based Inverted File
Semantc Image Retreval Usng Regon Based Inverted Fle Dengsheng Zhang, Md Monrul Islam, Guoun Lu and Jn Hou 2 Gppsland School of Informaton Technology, Monash Unversty Churchll, VIC 3842, Australa E-mal:
More informationDetection of hand grasping an object from complex background based on machine learning co-occurrence of local image feature
Detecton of hand graspng an object from complex background based on machne learnng co-occurrence of local mage feature Shnya Moroka, Yasuhro Hramoto, Nobutaka Shmada, Tadash Matsuo, Yoshak Shra Rtsumekan
More informationResolving Surface Forms to Wikipedia Topics
Resolvng Surface Forms to Wkpeda Topcs Ypng Zhou Lan Ne Omd Rouhan-Kalleh Flavan Vasle Scott Gaffney Yahoo! Labs at Sunnyvale {zhouy,lanne,omd,flavan,gaffney}@yahoo-nc.com Abstract Ambguty of entty mentons
More informationFederated Search of Text-Based Digital Libraries in Hierarchical Peer-to-Peer Networks
Federated Search of Text-Based Dgtal Lbrares n Herarchcal Peer-to-Peer Networks Je Lu School of Computer Scence Carnege Mellon Unversty Pttsburgh, PA 15213 jelu@cs.cmu.edu Jame Callan School of Computer
More informationSteps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and. Accessibility (Gravity with Competition) Indices
Steps for Computng the Dssmlarty, Entropy, Herfndahl-Hrschman and Accessblty (Gravty wth Competton) Indces I. Dssmlarty Index Measurement: The followng formula can be used to measure the evenness between
More informationA Generation Model to Unify Topic Relevance and Lexicon-based Sentiment for Opinion Retrieval
A Generaton Model to Unfy Topc Relevance and Lexcon-based Sentment for Opnon Retreval Mn Zhang State key lab of Intellgent Tech.& Sys, Dept. of Computer Scence, Tsnghua Unversty, Bejng, 00084, Chna 86-0-6279-2595
More information12/2/2009. Announcements. Parametric / Non-parametric. Case-Based Reasoning. Nearest-Neighbor on Images. Nearest-Neighbor Classification
Introducton to Artfcal Intellgence V22.0472-001 Fall 2009 Lecture 24: Nearest-Neghbors & Support Vector Machnes Rob Fergus Dept of Computer Scence, Courant Insttute, NYU Sldes from Danel Yeung, John DeNero
More informationReliable Negative Extracting Based on knn for Learning from Positive and Unlabeled Examples
94 JOURNAL OF COMPUTERS, VOL. 4, NO. 1, JANUARY 2009 Relable Negatve Extractng Based on knn for Learnng from Postve and Unlabeled Examples Bangzuo Zhang College of Computer Scence and Technology, Jln Unversty,
More informationAPPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT
3. - 5. 5., Brno, Czech Republc, EU APPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT Abstract Josef TOŠENOVSKÝ ) Lenka MONSPORTOVÁ ) Flp TOŠENOVSKÝ
More information