Query classification using topic models and support vector machine

Size: px
Start display at page:

Download "Query classification using topic models and support vector machine"

Transcription

1 Query classfcaton usng topc models and support vector machne Deu-Thu Le Unversty of Trento, Italy Raffaella Bernard Unversty of Trento, Italy Abstract Ths paper descrbes a query classfcaton system for a specalzed doman. We take as a case study queres asked to a search engne of an art, cultural and hstory lbrary and classfy them aganst the lbrary catalogung categores. We show how clck-through lnks,.e., the lnks that a user clcks after submttng a query, can be exploted for extractng nformaton useful to enrch the query as well as for creatng the tranng set for a machne learnng based classfer. Moreover, we show how Topc Model can be exploted to further enrch the query wth hdden topcs nduced from the lbrary meta-data. The expermental evaluatons show that ths system consderably outperforms a matchng and rankng classfcaton approach, where queres (and categores) were also enrched wth smlar nformaton. 1 Introducton Query classfcaton (QC) s the task of automatcally labelng user queres nto a gven target taxonomy. Provdng query classfcaton can help the nformaton provders understand users needs based on the categores that the users are searchng for. The man challenges of ths task come from the nature of user queres, whch are usually very short and ambguous. Snce queres contan only several to a dozen words, a QC system often requres ether a rather large tranng set or an enrchment of queres wth other nformaton (Shen et al., 2006a), (Broder et al., 2007). Ths study wll focus on QC n art, culture and hstory doman, usng the Brdgeman art lbrary 1, although our framework s general enough to be used n dfferent domans. Manually creatng a tranng 1 set of queres to buld a classfer n a specfc doman s very tme-consumng. In ths study, we wll descrbe our method of automatcally creatng a tranng set based on the clck-through lnks and how we buld an SVM (Support Vector Machne) classfer wth the ntegraton of enrched nformaton. In (Le et al., 2011), t has been shown that clck-through nformaton and topc models are useful for query enrchment when the ultmate goal s query classfcaton. We wll follow ths enrchment step, but ntegrate ths nformaton nto a SVM classfer nstead of usng matchng and rankng between queres and categores as n (Le et al., 2011). The purpose of ths paper s to determne (1) whether the query enrchment wth clck-though nformaton and hdden topcs s useful for a machne learnng query classfcaton system usng SVM; and (2) whether ntegratng ths enrched nformaton nto a machne learnng classfer can perform better than the matchng and rankng system. In the next secton, we wll brefly revew the man streams of related work n QC. In secton 3, we wll descrbe the Brdgeman art lbrary. Secton 4 accounts for our proposed query classfcaton framework. In secton 5, we wll present our experment and evaluaton. Secton 6 concludes by dscussng our man achevements and proposng future work. 2 Related work Intal studes n QC classfy queres nto several dfferent types based on the nformaton needed by the user. (Broder, 2002) consdered three dfferent types of queres: nformatonal queres, navgatonal queres and transactonal queres. Ths stream of study focuses on the type of the queres, rather than topcal classfcaton of the queres. Another stream of work deals wth the problem 19 Proceedngs of the 2012 Student Research Workshop, pages 19 24, Jeju, Republc of Korea, 8-14 July c 2012 Assocaton for Computatonal Lngustcs

2 of classfyng queres nto a more complex taxonomy contanng dfferent topcs. Our study falls nto ths second stream. To classfy queres consderng ther meanng, some work consdered only nformaton avalable n queres (e.g., (Betzel et al., 2005) only used terms n queres). Some other work has attempted to enrch queres wth nformaton from external onlne dataset, e.g., web pages (Shen et al., 2006a; Broder et al., 2007) and web drectores (Shen et al., 2006b). Our work s smlar to ther n the dea of explotng addtonal dataset. However, nstead of usng search engnes as a way of collectng relevant documents, we use the metadata of the lbrary tself as a reference set. Furthermore, we employ topc models to analyze topcs for queres, rather than enrchng queres wth words selected from those webpages drectly as n (Shen et al., 2006a; Broder et al., 2007). The context of a gven query can provde useful nformaton to determne ts categores. Prevous studes have confrmed the mportance of search context n QC. (Cao et al., 2009) consdered the context to be both prevous queres wthn the same sesson and pages of the clcked urls. In our approach, we wll also consder clck through nformaton to enrch the queres and analyze topcs. In (Le et al., 2011), queres and categores are enrched wth both nformaton mned from the clckthrough lnks as well as topcs derved from a topc model estmated from the lbrary metadata. Subsequently, the queres are mapped to the categores based on ther cosne smlarty. Our proposed approach dffers from (Le et al., 2011) n three respects: (1) we enrch the queres, but not the categores (2) we employ a machne learnng system and ntegrate ths enrched nformaton as features to learn an SVM classfer (3) we assume that the category of a query s closely related to the category of the correspondng clck-through lnk, hence we automatcally create a tranng data for the SVM classfer by analyzng the query log. 3 Brdgeman Art Lbrary Brdgeman Art Lbrary (BAL) 2 s one of the world s top mage lbrares for art, culture and hstory. It contans mages from over 8,000 collectons and 2 more than 29,000 artsts, provdng a central source of fne art for mage users. Works of art n the lbrary have been annotated wth ttles and keywords. Some of them are categorzed nto a two-level taxonomy, a more fne-graned classfcaton of the Brdgeman browse menu. In our study, we do not use the mage tself but only the nformaton assocated wth t,.e., the ttle, keywords and categores. We wll take the 55 top-level categores from ths taxonomy, whch have been organzed by a doman expert, as our target taxonomy. 4 Buldng QC usng topc models and SVM Followng (Le et al., 2011), we enrch queres both wth the nformaton mned from the lbrary va clck-through lnks and the nformaton collected from the lbrary metadata va topc modelng. To perform the query enrchment wth topcs derved from the lbrary metadata, there are several mportant steps: Collectng and organzng the lbrary metadata as a reference set: the lbrary metadata contans the nformaton about artworks that have been annotated by experts. To take advantage of ths nformaton automatcally, we collected all annotated artworks and organzed them by ther gven categores. Estmatng a topc model for ths reference set: Ths step s performed usng hdden topc analyss models. In ths framework, we choose to use latent drchlet allocaton, LDA (Ble et al., 2003b). Analyzng topcs for queres and ntegratng topcs nto data for both the tranng set and new queres: After the reference set has been analyzed usng topc models, t wll be used to nfer topcs for queres. The topc model wll then be ntegrated nto the data to buld a classfer. 4.1 Query enrchment va clck-through lnks We automatcally extracted clck-through lnks from the query log (whch provdes us wth the ttle of the mage that the user clcks) to enrch the query, represented as a vector q, wth the ttle of one randomly-chosen clck-through assocated wth t. To further explot the clck-through lnk, we fnd the correspondng artwork and extract ts keywords: q t kw, where t, kw are the vectors of words 20

3 n the ttle and keywords respectvely. 4.2 Hdden Topc Models The underlyng dea s based upon a probablstc procedure of generatng a new set of artworks, where each set refers to ttles and keywords of all artworks n a category: Frst, each set w m = (w m,n ) Nm n=1 s generated by samplng a dstrbuton over topcs ϑ m from a Drchlet dstrbuton (Dr( α )), where N m s the number of words n that set m. After that, the topc assgnment for each observed word w m,n s performed by samplng a word place holder z m,n from a multnomal dstrbuton (Mult( ϑ m )). Then a word w m,n s pcked by samplng from the multnomal dstrbuton (Mult( ϕ zm,n )). Ths process s repeated untl all K topcs have been generated for the whole collecton. Table 1: Generaton process for LDA M: the total number of artwork sets K: the number of (hdden/latent) topcs V : vocabulary sze α, β : Drchlet parameters ϑ m: topc dstrbuton for document m ϕ k : word dstrbuton for topc k N m: the length of document m z m,n: topc ndex of nth word n document m w m,n: a partcular word for word placeholder [m, n] Θ = { ϑ m} M m=1: a M K matrx Φ = { ϕ k } K k=1: a K V matrx In order to estmate parameters for LDA (.e., the set of topcs and ther word probabltes Φ and the partcular topc mxture of each document Θ), dfferent nference technques can be used, such as varatonal Bayes (Ble et al., 2003b), or Gbbs samplng (Henrch, 2004). In ths work, we wll use Gbbs samplng followng the descrpton gven n (Henrch, 2004). Generally, the topc assgnment of a partcular word t s computed as: p(z =k z, w)= [ V n (t) k, + β t v=1 n(v) k +β v] 1 n (k) m, + α k [ K j=1 n(j) m +α j ] 1 (1) where n (t) k, s the number of tmes the word t s assgned to topc k except the current assgnment; V v=1 n(v) k 1 s the total number of words assgned to topc k except the current assgnment; n (k) m, s the number of words n set m assgned to topc k except the current assgnment; and K j=1 n(j) m 1 s the total number of words n set m except the current word t. In normal cases, Drchlet parameters α, and β are symmetrc, that s, all α k (k = 1..K) are the same, and smlarly for β v (v = 1..V ). 4.3 Hdden topc analyss of the Brdgeman metadata The Brdgeman metadata contans nformaton about artworks n the lbrary that have been annotated by the lbrarans. We extracted ttles and keywords of each artwork, those for whch we had a query wth a clck-through lnk correspondng to t, and grouped them together by ther sub-categores. Each group s consdered as a document w m = (w m,n ) Nm n=1, wth the number of total documents M = 732 and the vocabulary sze V = 136K words. In ths experment, we fx the number of topcs K = 100. We used the GbbsLDA++ mplementaton 3 to estmate ths topc model. 4.4 Buldng query classfer wth hdden topcs Let Q = { q } =1 =N be the set of all queres enrched va the clck-through lnks, where each enrched query s q = q t kw. We also performed Gbbs samplng for all q n order to estmate ts topc dstrbuton ϑ = {ϑ,1,..., ϑ,k } where the probablty ϑ,k of topc k n q s computed as: + α k ϑ,k = (2) + α j where n (k) s the number of words n query assgned to topc k and n (j) s the total number of words appearng n the enrched query. In order to ntegrate the topc dstrbuton ϑ = {ϑ,1,..., ϑ,k } nto the vector of words q = {w,1, w,2,..., w,n }, followng (Phan et al., 2010), we only keep topcs whose ϑ,k s larger than a threshold cut-off and use a scale parameter to do the dscretzaton for topcs: the number of tmes topc k ntegrated to q s round(ϑ scale). After that, we buld a Support Vector Machne classfer usng SVM lght V n (k) K j=1 n(j) 21

4 5 Evaluaton In ths secton, we wll descrbe our tranng set, gold standard and the performance of our system n comparson wth the one n (Le et al., 2011). 5.1 Tranng set Manually annotatng queres to create a tranng set n ths doman s a dffcult task (e.g., t requres the expert to search the query and look at the pcture correspondng to the query, etc.). Therefore, we have automatcally generated a tranng set by explotng a 6-month query log as follow. Frst, each query has been mapped to ts clckthrough nformaton to extract the sub-category assocated to the correspondng mage. Then, from ths sub-category, we obtaned ts correspondng top-cateogry (among the 55 we consder) as defned n BAL taxonomy. The dstrbuton of queres n dfferent categores vares qute a lot among the 55 target categores reflectng the artwork dstrbuton (e.g., there are many more artworks n the lbrary belongng to the category Relgon and Belef than to the category Costume and Fashon ). We have preserved such dstrbuton over the target categores when selectng randomly the 15,490 queres to buld our tranng set. After removng all punctuatons and stop words, we obtaned a tranng set contanng 50,337 words n total. Each word n ths set serves as a feature for the SVM classfer. 5.2 Test set We used the test set of 1,049 queres used n (Le et al., 2011), whch s separate from the tranng set. These queres have been manually annotated by a BAL expert (up to 3 categores per query). Note that these queres have also been selected automatcally whle preservng the dstrbuton over the target categores observed n the 6-month query log. We call ths the manual gold standard. In addton, we also made use of another gold standard obtaned by mappng the clck-through nformaton of these queres wth ther categores, smlar to the way n whch we obtan the tranng set. We call ths the va-ct gold standard. 5.3 Expermental settngs To evaluate the mpact of clck-though nformaton and topcs n the classfer, we desgned the followng experments, where QR s the method wthout any enrchment and QR-CT -HT s wth the enrchment va both clck-through and hdden topcs. Settng Query enrchment QR q QR-HT q HT QR-CT q = q + t + kw QR-CT -HT q HT q : query q : query enrched wth clck-through nformaton t : clck-through mage s ttle kw: clck-through mage s keywords HT : hdden topcs from Brdgeman metadata Table 2: Expermental Settng Hts Settng Manual GS va-ct # 1 # 2 # 3 T op 3 GS QR QR-HT QR - CT QR - CT - HT Table 3: Results of query classfcaton: number of correct categores found (for 1,049 queres) Fgure 1: The mpact of clck-through nformaton wth matchng-rankng (mr) and our approach (svm) To answer our frst research queston, namely whether clck-through nformaton and hdden topcs are useful for ths query classfer, we examne the number of correct categores found by the classfer bult both wth and wthout the enrchment. The results of the experment are reported n Table 3. As can be seen from the table, we notce that the clckthrough nformaton plays an mportant role. In par- 22

5 tcular, t ncreases the number of correct categores found from 311 to 388 (compared wth the manual GS) and from 231 to 266 (usng the va-ct GS). To answer our second research queston, namely whether ntegratng the enrched nformaton nto a machne learnng classfer can perform better than the matchng and rankng method, we also compare the results of our approach wth the one n (Le et al., 2011). Fgure 1 shows the mpact of the clckthrough nformaton for the SVM classfer (svm) n comparson wth the matchng and rankng approach (mr). Fgure 2 shows the mpact of the hdden topcs n both cases. We can see that n both cases our classfer outperforms the matchng-rankng one consderably (e.g., from 183 to 388 correct categores found n the QR-CT-HT method). Fgure 2: The mpact of hdden topcs wth matchngrankng (mr) and our approach (svm) However, n the case where we use only queres wthout clck-through nformaton, we can see that hdden topcs do not brng a very strong mpact (the number of correct categores found only slghtly ncreases by 7 - usng the manual gold standard). The result mght come from the fact that ths topc model was bult from the metadata, usng only clckthrough nformaton, but has not been learned wth queres. 6 Concluson In ths study, we have presented a machne learnng classfer for query classfcaton n an art mage archve. Snce queres are usually very short, thus dffcult to classfy, we frst extend them wth ther clck-through nformaton. Then, these queres are further enrched wth topcs learned from the BAL metadata followng (Le et al., 2011). The result from ths study has confrmed agan the effect of clck-through nformaton and hdden topcs n the query classfcaton task usng SVM. We have also descrbed our method of automatcally creatng a tranng set based on the selecton of queres mapped to the clck-through lnks and ther correspondng avalable categores usng a 6-month query log. The result of ths study has shown a consderable ncrease n the performance of ths approach over the matchng-rankng system reported n (Le et al., 2011). 7 Future work For future work, we are n the process of enhancng our expermentaton n several drectons: Consderng more than one clck-through mage per query: In ths work, we have consdered only one category per query to create the tranng set, whle t mght be more reasonable to take nto account all clck-through mages of a gven query. In future work, we plan to enrch the queres wth ether all clck-through mages or wth the most relevant one nstead of randomly pckng one clck-through mage. In many cases, a clck-through lnk s not necessarly related to the meanng of a query (e.g., when users just randomly clck on an mage that they fnd nterestng). Thus, t mght be useful to flter out those clck-through mages that are not relevant. Enrchng queres wth top hts returned by the BAL search engne: In the query logs, there are many queres that do not have an assocated clckthrough lnk. Hence, we plan to explot other enrchment method that do not rely on those lnks, n partcular we wll try to explot the nformaton comng from the top returned hts gven by the lbrary search engne. Analyzng queres n the same sesson: It has been shown n some studes (Cao et al., 2009) that analyzng queres n the same sesson can help determne ther categores. Our next step s to enrch a new query wth the nformaton comng from the other prevous queres n the same sesson. Optmzng LDA hyperparameters and topc number selecton: Currently, we fxed the number of topcs K = 100, the Drchlet hyperparameters α = 50/K = 0.5 and β = 0.1 as n (Grffths and 23

6 Steyvers, 2004). In the future, we wll explore ways to optmze these nput values to see the effect of dfferent topc models n our query classfcaton task. Explotng vsual features from the BAL mages: The BAL dataset provdes an nterestng case study n whch we plan to further analyze mages to enrch queres wth ther vsual features. Combnng text and vsual features has drawn a lot of attenton n the IR research communty. We beleve that explotng vsual features from ths art archve could lead to nterestng results n ths specfc doman. A possble approach would be extractng vsual features from the clck-through mages and representng them together wth textual features n a jont topc dstrbuton (e.g., (Ble et al., 2003a; L et al., 2010)). Comparng system wth other approaches: In the future, we plan to compare our system wth other query classfcaton systems and smlar technques for query expanson n general. Furthermore, the evaluaton phase has not been carred out thoroughly snce t was dffcult to compare the one-class output wth the gold-standard, where the number of correct categores per query s not fxed. In the future, we plan to explot the output of our mult-class classfer to assgn up to three categores for each query and compute the precson at n. Acknowledgments Ths work has been partally supported by the GALATEAS project ( CIP-ICT PSP ) funded by the European Unon under the ICT PSP program. References Steven M. Betzel, Erc C. Jensen, Ophr Freder, and Davd Grossman Automatc web query classfcaton usng labeled and unlabeled tranng data. In In Proceedngs of the 28th Annual Internatonal ACM SIGIR Conference on Research and Development n Informaton Retreval, pages ACM Press. Davd M. Ble, Mchael I, Davd M. Ble, and Mchael I. 2003a. Modelng annotated data. In In Proc. of the 26th Intl. ACM SIGIR Conference. Davd M. Ble, Andrew Y. Ng, and Mchael I. Jordan. 2003b. Latent drchlet allocaton. J. Mach. Learn. Res., 3: , March. Andre Z. Broder, Marcus Fontoura, Evgeny Gabrlovch, Amruta Josh, Vanja Josfovsk, and Tong Zhang Robust classfcaton of rare queres usng web knowledge. In Proceedngs of the 30th annual nternatonal ACM SIGIR conference on Research and development n nformaton retreval, SIGIR 07, pages , New York, NY, USA. ACM. Andre Broder A taxonomy of web search. SIGIR Forum, 36:3 10, September. Huanhuan Cao, Derek Hao Hu, Dou Shen, Dax Jang, Jan-Tao Sun, Enhong Chen, and Qang Yang Context-aware query classfcaton. In SIGIR 09, The 32nd Annual ACM SIGIR Conference. Thomas L Grffths and Mark Steyvers Fndng scentfc topcs. Proceedngs of the Natonal Academy of Scences of the Unted States of Amerca, 101 Suppl 1(Suppl 1): Gregor Henrch Parameter estmaton for text analyss. Techncal report. Deu-Thu Le, Raffaella Bernard, and Edwn Vald Query classfcaton va topc models for an art mage archve. In Recent Advances n Natural Language Processng, RANLP, Bulgara. L-Ja L, Chong Wang, Yongwhan Lm, Davd Ble, and L Fe-Fe Buldng and usng a semantvsual mage herarchy. In The Twenty-Thrd IEEE Conference on Computer Vson and Pattern Recognton, San Francsco, CA, June. Xuan-Heu Phan, Cam-Tu Nguyen, Deu-Thu Le, Le- Mnh Nguyen, Susumu Horguch, and Quang-Thuy Ha A hdden topc-based framework towards buldng applcatons wth short web documents. IEEE Transactons on Knowledge and Data Engneerng, 99(PrePrnts). Dou Shen, Rong Pan, Jan-Tao Sun, Jeffrey Junfeng Pan, Kangheng Wu, Je Yn, and Gang Yang. 2006a. Query enrchment for web-query classfcaton. ACM Transactons on Informaton Systems, 24(3): Dou Shen, Jan-Tao Sun, Qang Yang, and Zheng Chen. 2006b. Buldng brdges for web query classfcaton. In SIGIR

Query Clustering Using a Hybrid Query Similarity Measure

Query 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 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

UB at GeoCLEF Department of Geography Abstract

UB 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 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

A Unified Framework for Semantics and Feature Based Relevance Feedback in Image Retrieval Systems

A 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 information

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

Keywords - Wep page classification; bag of words model; topic model; hierarchical classification; Support Vector Machines

Keywords - 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 information

Deep Classification in Large-scale Text Hierarchies

Deep 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 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

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

Combining Multiple Resources, Evidence and Criteria for Genomic Information Retrieval

Combining Multiple Resources, Evidence and Criteria for Genomic Information Retrieval Combnng Multple Resources, Evdence and Crtera for Genomc Informaton Retreval Luo S 1, Je Lu 2 and Jame Callan 2 1 Department of Computer Scence, Purdue Unversty, West Lafayette, IN 47907, USA ls@cs.purdue.edu

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

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

CAN COMPUTERS LEARN FASTER? Seyda Ertekin Computer Science & Engineering The Pennsylvania State University

CAN COMPUTERS LEARN FASTER? Seyda Ertekin Computer Science & Engineering The Pennsylvania State University CAN COMPUTERS LEARN FASTER? Seyda Ertekn Computer Scence & Engneerng The Pennsylvana State Unversty sertekn@cse.psu.edu ABSTRACT Ever snce computers were nvented, manknd wondered whether they mght be made

More information

Improving Web Image Search using Meta Re-rankers

Improving 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 information

A Novel Term_Class Relevance Measure for Text Categorization

A Novel Term_Class Relevance Measure for Text Categorization A Novel Term_Class Relevance Measure for Text Categorzaton D S Guru, Mahamad Suhl Department of Studes n Computer Scence, Unversty of Mysore, Mysore, Inda Abstract: In ths paper, we ntroduce a new measure

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

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

User Authentication Based On Behavioral Mouse Dynamics Biometrics

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

More information

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

Web-supported Matching and Classification of Business Opportunities

Web-supported Matching and Classification of Business Opportunities Web-supported Matchng and Classfcaton of Busness Opportuntes. DIRO Unversté de Montréal C.P. 628, succursale Centre-vlle Montréal, Québec, H3C 3J7, Canada Jng Ba, Franços Parads,2, Jan-Yun Ne {bajng, paradfr,

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

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

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

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

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

Personalized Concept-Based Clustering of Search Engine Queries

Personalized Concept-Based Clustering of Search Engine Queries IEEE TRANSACTIONS ON JOURNAL NAME, MANUSCRIPT ID 1 Personalzed Concept-Based Clusterng of Search Engne Queres Kenneth Wa-Tng Leung, Wlfred Ng, and Dk Lun Lee Abstract The exponental growth of nformaton

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

Cross-lingual Pseudo Relevance Feedback Based on Weak Relevant Topic Alignment

Cross-lingual Pseudo Relevance Feedback Based on Weak Relevant Topic Alignment Cross-lngual Pseudo Relevance Feedback Based on Weak Relevant opc Algnment WANG Xu-wen Insttute of Medcal Informaton & Lbrary, Chnese Academy of Medcal Scences, Beng 100020 wang.xuwen@mcams.ac.cn ZHANG

More information

FINDING IMPORTANT NODES IN SOCIAL NETWORKS BASED ON MODIFIED PAGERANK

FINDING IMPORTANT NODES IN SOCIAL NETWORKS BASED ON MODIFIED PAGERANK FINDING IMPORTANT NODES IN SOCIAL NETWORKS BASED ON MODIFIED PAGERANK L-qng Qu, Yong-quan Lang 2, Jng-Chen 3, 2 College of Informaton Scence and Technology, Shandong Unversty of Scence and Technology,

More information

Intelligent Information Acquisition for Improved Clustering

Intelligent Information Acquisition for Improved Clustering Intellgent Informaton Acquston for Improved Clusterng Duy Vu Unversty of Texas at Austn duyvu@cs.utexas.edu Mkhal Blenko Mcrosoft Research mblenko@mcrosoft.com Prem Melvlle IBM T.J. Watson Research Center

More information

Backpropagation: In Search of Performance Parameters

Backpropagation: In Search of Performance Parameters Bacpropagaton: In Search of Performance Parameters ANIL KUMAR ENUMULAPALLY, LINGGUO BU, and KHOSROW KAIKHAH, Ph.D. Computer Scence Department Texas State Unversty-San Marcos San Marcos, TX-78666 USA ae049@txstate.edu,

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

Three supervised learning methods on pen digits character recognition dataset

Three 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 information

Discriminative Dictionary Learning with Pairwise Constraints

Discriminative Dictionary Learning with Pairwise Constraints Dscrmnatve Dctonary Learnng wth Parwse Constrants Humn Guo Zhuoln Jang LARRY S. DAVIS UNIVERSITY OF MARYLAND Nov. 6 th, Outlne Introducton/motvaton Dctonary Learnng Dscrmnatve Dctonary Learnng wth Parwse

More information

A Statistical Model Selection Strategy Applied to Neural Networks

A Statistical Model Selection Strategy Applied to Neural Networks A Statstcal Model Selecton Strategy Appled to Neural Networks Joaquín Pzarro Elsa Guerrero Pedro L. Galndo joaqun.pzarro@uca.es elsa.guerrero@uca.es pedro.galndo@uca.es Dpto Lenguajes y Sstemas Informátcos

More information

CUM: An Efficient Framework for Mining Concept Units

CUM: An Efficient Framework for Mining Concept Units CUM: An Effcent Framework for Mnng Concept Unts P.Santh Thlagam Ananthanarayana V.S Department of Informaton Technology Natonal Insttute of Technology Karnataka - Surathkal Inda 575025 santh_soc@yahoo.co.n,

More information

Deep Classifier: Automatically Categorizing Search Results into Large-Scale Hierarchies

Deep Classifier: Automatically Categorizing Search Results into Large-Scale Hierarchies Deep Classfer: Automatcally Categorzng Search Results nto Large-Scale Herarches Dkan Xng 1, Gu-Rong Xue 1, Qang Yang 2, Yong Yu 1 1 Shangha Jao Tong Unversty, Shangha, Chna {xaobao,grxue,yyu}@apex.sjtu.edu.cn

More information

CSCI 5417 Information Retrieval Systems Jim Martin!

CSCI 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 information

Semantic Image Retrieval Using Region Based Inverted File

Semantic 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 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

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

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

Relevance Feedback for Image Retrieval

Relevance Feedback for Image Retrieval Vashal D Dhale et al, / (IJCSIT Internatonal Journal of Computer Scence and Informaton Technologes, Vol 4 (2, 203, 39-323 Relevance Feedback for Image Retreval Vashal D Dhale, Dr A R Mahaan, Prof Uma Thakur

More information

Classification of Face Images Based on Gender using Dimensionality Reduction Techniques and SVM

Classification of Face Images Based on Gender using Dimensionality Reduction Techniques and SVM Classfcaton of Face Images Based on Gender usng Dmensonalty Reducton Technques and SVM Fahm Mannan 260 266 294 School of Computer Scence McGll Unversty Abstract Ths report presents gender classfcaton based

More information

IN recent years, we have been witnessing the explosive

IN recent years, we have been witnessing the explosive IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL. 15, NO. 4, JULY/AUGUST 2003 1 Query Expanson by Mnng User Logs Hang Cu, J-Rong Wen, Jan-Yun Ne, and We-Yng Ma, Member, IEEE Abstract Queres to

More information

Text Similarity Computing Based on LDA Topic Model and Word Co-occurrence

Text Similarity Computing Based on LDA Topic Model and Word Co-occurrence 2nd Internatonal Conference on Software Engneerng, Knowledge Engneerng and Informaton Engneerng (SEKEIE 204) Text Smlarty Computng Based on LDA Topc Model and Word Co-occurrence Mngla Shao School of Computer,

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

A Web Site Classification Approach Based On Its Topological Structure

A Web Site Classification Approach Based On Its Topological Structure Internatonal Journal on Asan Language Processng 20 (2):75-86 75 A Web Ste Classfcaton Approach Based On Its Topologcal Structure J-bn Zhang,Zh-mng Xu,Kun-l Xu,Q-shu Pan School of Computer scence and Technology,Harbn

More information

Web Document Classification Based on Fuzzy Association

Web Document Classification Based on Fuzzy Association Web Document Classfcaton Based on Fuzzy Assocaton Choochart Haruechayasa, Me-Lng Shyu Department of Electrcal and Computer Engneerng Unversty of Mam Coral Gables, FL 33124, USA charuech@mam.edu, shyu@mam.edu

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

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

BAYESIAN MULTI-SOURCE DOMAIN ADAPTATION

BAYESIAN 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 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

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

A Knowledge Management System for Organizing MEDLINE Database

A Knowledge Management System for Organizing MEDLINE Database A Knowledge Management System for Organzng MEDLINE Database Hyunk Km, Su-Shng Chen Computer and Informaton Scence Engneerng Department, Unversty of Florda, Ganesvlle, Florda 32611, USA Wth the exploson

More information

Face Recognition Based on SVM and 2DPCA

Face Recognition Based on SVM and 2DPCA Vol. 4, o. 3, September, 2011 Face Recognton Based on SVM and 2DPCA Tha Hoang Le, Len Bu Faculty of Informaton Technology, HCMC Unversty of Scence Faculty of Informaton Scences and Engneerng, 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

Fuzzy C-Means Initialized by Fixed Threshold Clustering for Improving Image Retrieval

Fuzzy 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 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

Learning Topic Structure in Text Documents using Generative Topic Models

Learning Topic Structure in Text Documents using Generative Topic Models Learnng Topc Structure n Text Documents usng Generatve Topc Models Ntsh Srvastava CS 397 Report Advsor: Dr Hrsh Karnck Abstract We present a method for estmatng the topc structure for a document corpus

More information

The Effect of Similarity Measures on The Quality of Query Clusters

The Effect of Similarity Measures on The Quality of Query Clusters The effect of smlarty measures on the qualty of query clusters. Fu. L., Goh, D.H., Foo, S., & Na, J.C. (2004). Journal of Informaton Scence, 30(5) 396-407 The Effect of Smlarty Measures on The Qualty of

More information

Collaboratively Regularized Nearest Points for Set Based Recognition

Collaboratively 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 information

Gender Classification using Interlaced Derivative Patterns

Gender Classification using Interlaced Derivative Patterns Gender Classfcaton usng Interlaced Dervatve Patterns Author Shobernejad, Ameneh, Gao, Yongsheng Publshed 2 Conference Ttle Proceedngs of the 2th Internatonal Conference on Pattern Recognton (ICPR 2) DOI

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

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

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

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

Description of NTU Approach to NTCIR3 Multilingual Information Retrieval

Description 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 information

BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET

BOOSTING 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 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

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

Fuzzy Modeling of the Complexity vs. Accuracy Trade-off in a Sequential Two-Stage Multi-Classifier System

Fuzzy Modeling of the Complexity vs. Accuracy Trade-off in a Sequential Two-Stage Multi-Classifier System Fuzzy Modelng of the Complexty vs. Accuracy Trade-off n a Sequental Two-Stage Mult-Classfer System MARK LAST 1 Department of Informaton Systems Engneerng Ben-Guron Unversty of the Negev Beer-Sheva 84105

More information

Decision Strategies for Rating Objects in Knowledge-Shared Research Networks

Decision Strategies for Rating Objects in Knowledge-Shared Research Networks Decson Strateges for Ratng Objects n Knowledge-Shared Research etwors ALEXADRA GRACHAROVA *, HAS-JOACHM ER **, HASSA OUR ELD ** OM SUUROE ***, HARR ARAKSE *** * nsttute of Control and System Research,

More information

A Hidden Markov Model Variant for Sequence Classification

A Hidden Markov Model Variant for Sequence Classification Proceedngs of the Twenty-Second Internatonal Jont Conference on Artfcal Intellgence A Hdden Markov Model Varant for Sequence Classfcaton Sam Blasak and Huzefa Rangwala Computer Scence, George Mason Unversty

More information

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms Course Introducton Course Topcs Exams, abs, Proects A quc loo at a few algorthms 1 Advanced Data Structures and Algorthms Descrpton: We are gong to dscuss algorthm complexty analyss, algorthm desgn technques

More information

Outline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1

Outline. 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 information

Alignment Results of SOBOM for OAEI 2010

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 information

PRÉSENTATIONS DE PROJETS

PRÉSENTATIONS DE PROJETS PRÉSENTATIONS DE PROJETS Rex Onlne (V. Atanasu) What s Rex? Rex s an onlne browser for collectons of wrtten documents [1]. Asde ths core functon t has however many other applcatons that make t nterestng

More information

Experiments in Text Categorization Using Term Selection by Distance to Transition Point

Experiments in Text Categorization Using Term Selection by Distance to Transition Point Experments n Text Categorzaton Usng Term Selecton by Dstance to Transton Pont Edgar Moyotl-Hernández, Héctor Jménez-Salazar Facultad de Cencas de la Computacón, B. Unversdad Autónoma de Puebla, 14 Sur

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

Feature Kernel Functions: Improving SVMs Using High-level Knowledge

Feature Kernel Functions: Improving SVMs Using High-level Knowledge Feature Kernel Functons: Improvng SVMs Usng Hgh-level Knowledge Qang Sun, Gerald DeJong Department of Computer Scence, Unversty of Illnos at Urbana-Champagn qangsun@uuc.edu, dejong@cs.uuc.edu Abstract

More information

Optimizing Document Scoring for Query Retrieval

Optimizing 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 information

Pruning Training Corpus to Speedup Text Classification 1

Pruning 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 information

Resolving Surface Forms to Wikipedia Topics

Resolving 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 information

Vehicle Fault Diagnostics Using Text Mining, Vehicle Engineering Structure and Machine Learning

Vehicle Fault Diagnostics Using Text Mining, Vehicle Engineering Structure and Machine Learning Internatonal Journal of Intellgent Informaton Systems 205; 4(3): 58-70 Publshed onlne July 8, 205 (http://www.scencepublshnggroup.com//s) do: 0.648/.s.2050403.2 ISSN: 2328-7675 (Prnt); ISSN: 2328-7683

More information

Brushlet Features for Texture Image Retrieval

Brushlet Features for Texture Image Retrieval DICTA00: Dgtal Image Computng Technques and Applcatons, 1 January 00, Melbourne, Australa 1 Brushlet Features for Texture Image Retreval Chbao Chen and Kap Luk Chan Informaton System Research Lab, School

More information

Learning to Classify Documents with Only a Small Positive Training Set

Learning to Classify Documents with Only a Small Positive Training Set Learnng to Classfy Documents wth Only a Small Postve Tranng Set Xao-L L 1, Bng Lu 2, and See-Kong Ng 1 1 Insttute for Infocomm Research, Heng Mu Keng Terrace, 119613, Sngapore 2 Department of Computer

More information

A New Feature of Uniformity of Image Texture Directions Coinciding with the Human Eyes Perception 1

A New Feature of Uniformity of Image Texture Directions Coinciding with the Human Eyes Perception 1 A New Feature of Unformty of Image Texture Drectons Concdng wth the Human Eyes Percepton Xng-Jan He, De-Shuang Huang, Yue Zhang, Tat-Mng Lo 2, and Mchael R. Lyu 3 Intellgent Computng Lab, Insttute of Intellgent

More information

CHAPTER 2 DECOMPOSITION OF GRAPHS

CHAPTER 2 DECOMPOSITION OF GRAPHS CHAPTER DECOMPOSITION OF GRAPHS. INTRODUCTION A graph H s called a Supersubdvson of a graph G f H s obtaned from G by replacng every edge uv of G by a bpartte graph,m (m may vary for each edge by dentfyng

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

Remote Sensing Image Retrieval Algorithm based on MapReduce and Characteristic Information

Remote Sensing Image Retrieval Algorithm based on MapReduce and Characteristic Information Remote Sensng Image Retreval Algorthm based on MapReduce and Characterstc Informaton Zhang Meng 1, 1 Computer School, Wuhan Unversty Hube, Wuhan430097 Informaton Center, Wuhan Unversty Hube, Wuhan430097

More information

Semi Supervised Learning using Higher Order Cooccurrence Paths to Overcome the Complexity of Data Representation

Semi Supervised Learning using Higher Order Cooccurrence Paths to Overcome the Complexity of Data Representation Sem Supervsed Learnng usng Hgher Order Cooccurrence Paths to Overcome the Complexty of Data Representaton Murat Can Ganz Computer Engneerng Department, Faculty of Engneerng Marmara Unversty, İstanbul,

More information

Selecting Query Term Alterations for Web Search by Exploiting Query Contexts

Selecting Query Term Alterations for Web Search by Exploiting Query Contexts Selectng Query Term Alteratons for Web Search by Explotng Query Contexts Guhong Cao Stephen Robertson Jan-Yun Ne Dept. of Computer Scence and Operatons Research Mcrosoft Research at Cambrdge Dept. of Computer

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

Enhancement of Infrequent Purchased Product Recommendation Using Data Mining Techniques

Enhancement 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 information

Application of k-nn Classifier to Categorizing French Financial News

Application of k-nn Classifier to Categorizing French Financial News Applcaton of k-nn Classfer to Categorzng French Fnancal News Huazhong KOU, Georges GARDARIN 2, Alan D'heygère 2, Karne Zetoun PRSM Laboratory, Unversty of Versalles Sant-Quentn 45 Etats-Uns Road, 78035

More information

Background Removal in Image indexing and Retrieval

Background Removal in Image indexing and Retrieval Background Removal n Image ndexng and Retreval Y Lu and Hong Guo Department of Electrcal and Computer Engneerng The Unversty of Mchgan-Dearborn Dearborn Mchgan 4818-1491, U.S.A. Voce: 313-593-508, Fax:

More information

Federated Search of Text-Based Digital Libraries in Hierarchical Peer-to-Peer Networks

Federated 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 information

Incremental Learning with Support Vector Machines and Fuzzy Set Theory

Incremental Learning with Support Vector Machines and Fuzzy Set Theory The 25th Workshop on Combnatoral Mathematcs and Computaton Theory Incremental Learnng wth Support Vector Machnes and Fuzzy Set Theory Yu-Mng Chuang 1 and Cha-Hwa Ln 2* 1 Department of Computer Scence and

More information