Feature Selection for Web Page Classification Using Swarm Optimization

Size: px
Start display at page:

Download "Feature Selection for Web Page Classification Using Swarm Optimization"

Transcription

1 Feaure Selecion for Web Page Classificaion Using Swarm Opimizaion B. Leela Devi, A. Sankar Absrac The web s increased populariy has included a huge amoun of informaion, due o which auomaed web page classificaion sysems are essenial o improve search engines performance. Web pages have many feaures like HTML or XML ags, hyperlinks, URLs and ex conens which can be considered during an auomaed classificaion process. I is known ha Webpage classificaion is enhanced by hyperlinks as i reflecs Web page linkages. The aim of his sudy is o reduce he number of feaures o be used o improve he accuracy of he classificaion of web pages. In his paper, a novel feaure selecion mehod using an improved Paricle Swarm Opimizaion (PSO) using principle of evoluion is proposed. The exraced feaures were esed on he WebKB daase using a parallel Neural Nework o reduce he compuaional cos. Keywords Web page classificaion, WebKB Daase, Term Frequency-Inverse Documen Frequency (TF-IDF), Paricle Swarm Opimizaion (PSO). I. INTRODUCTION HE increase in usage of Web and is growh are well Tknown. Texual daa on he Web is esimaed a one erra bye, in addiion o audio and video images which imposes new challenges o Web direcories. Web direcories enable users o search he Web, by classifying Web documens ino subjecs. Web pages manual classificaion suffers as Web documens increase [].Tex classificaion aims o caegorize documens ino a specific number of predefined classes using documen feaures. Tex classificaion has a crucial role in rerieval and managemen asks like informaion exracion, informaion rerieval, documen filering, and building hierarchical direcories [2]. When ex classificaion focuses on web pages, i is called web classificaion or web page classificaion. Classificaion assigns predefined class labels o unseen or es daa. For his, a se of labelled daa rains a classifier which hen labels unseen daa. Classificaion is supervised learning [3]. The process is no differen in web page classificaion so ha here are one or more predefined class labels. Classificaion model assigns class labels o web pages which are hyperex wih many feaures like exual okens, markup ags, URLs and hos names in URLs. As web pages have addiional properies, his classificaion differs from radiional ex classificaion. Web page classificaion has subfields like subjec classificaion and funcional classificaion. In he former, he Leela Devi B is wih Professional Group of Insiuions, Palladam, Tamilnadu, India ( leeladevi_2008@rediffmail.com). Sankar A is wih PSG College of Technology, Coimbaore, Tamilnadu, India. classifier is concerned wih web page conen and deermines he web page subjec. For example, online newspapers caegories like finance, spor, and echnology are examples of subjec classificaion. Funcional classificaion deals wih funcion or ype of a web page. For example, deermining wheher a web page is a personal homepage or a course page is funcional classificaion. Subjec and funcional classificaion are popular classificaion ypes [2]. A HTML documen s individual componen is an HTML elemen made up of a ree of HTML elemens and nodes like ex nodes wih all elemens having specified aribues. Elemens have conen and ex. HTML represens semanics or meaning [4]. HTML markup has key componens, including characer references, characer-based daa ypes, elemens (and aribues), and eniy references. Anoher componen is documen ype declaraion, riggering sandards mode rendering. Semanic HTML is wriing HTML emphasizing encoded informaion s meaning over presenaion (looks). HTML includes semanic markup from incepion, and presenaional markup like <fon>, <i> and <cener> ags. There are also semanically neural ags. HTML and associaed proocols from incepion were acceped quickly. Bu, here were no sandards in he early years of language. Though HTML was conceived as a semanic language wihou presenaion deails, pracical use pushed presenaional elemens and aribues o i, driven by varied browser vendors. Laes HTML sandards are effors o overcome chaoic language developmen and o creae a raional foundaion o build meaningful and well-presened documens. Feaure selecion is an imporan classificaion sep. Web pages are in HTML forma meaning ha web pages are semisrucured daa, wih HTML ags and hyperlinks in addiion o pure ex. Due o his web pages propery, feaure selecion in classificaion is differen from radiional classificaion. Feaure selecion reduces daa dimension wih ens or hundreds or housands of feaures which canno be processed furher. A major problem of web page classificaion is he high dimensionaliy of he feaure space. Bes feaure subses have leas feaures ha mos conribue o classificaion accuracy. To improve web page classificaion performance, many approaches impored from feaure selecion or ex classificaion were applied. Informaion gain [3], muual informaion [5], documen frequency [6], and erm srengh [7] are popular feaure selecion echniques. Informaion gain (IG) measures informaion in bis abou he class predicion, when he only informaion available is a feaure and corresponding 340

2 class disribuion. This sudy proposes a new feaure selecion echnique using PSO algorihm for web page classificaion. Secion II reviews relaed work in lieraure. Secion III describes mehods used and Secion IV discusses he resuls of experimens. Secion V concludes he paper. II. RELATED WORKS Auomaic web page classificaion was emphasized by [8] hrough minimum feaures. They also proposed a procedure o generae opimum feaures for web pages. The opimum feaures model, machine learning classifiers. Experimen wih a bench marking daa se wih such machine learning classifiers improved classificaion accuracy. To improve classificaion resuls wih resources use, a full web page is no needed. As web pages conain high dimensions daa, hey are preprocessed o idenify bes represenaive feaures reflecing caegories. A mulilevel feaure selecion wihou bias o frequen erms was proposed. Resuls reveal he new feaure selecion process idenifies lesser feaures wih high informaion gain ensuring classificaion accuracy. Previous works reveal a web page being pariioned ino many segmens or blocks. Blocks imporance o a page is no equal. I was proved ha differeniaing noisy and unimporan blocks from pages helped web mining, search, and accessibiliy. No uniform approach was presened, in he same works, o measure imporan web page porions. A user sudy by [9] found people having consisen views on web page blocks imporance. The new work invesigaes how o locae a model o auomaically assign imporance values o web page blocks. Block imporance esimaion as a learning problem is defined. Firs, Vision-based Page Segmenaion (VIPS) algorihm pariions a web page ino semanic blocks wih hierarchy srucure. Then spaial feaures like posiion, size, and conen feaures including many images and links were exraced for feaure vecor consrucion in each block. Learning algorihms like SVM and NN rain block imporance models based on his. In he proposed experimens, bes models achieve 79% performances wih Micro-F and 85.9%, wih Micro-Accuracy. Web classificaion was ried via differen echnologies. Xhemali e al. [0] compared NN, NB, and DT classifiers for auomaic analysis and classificaion of raining course web pages aribue daa. The sudy inroduced an enhanced NB classifier and ran he same daa sample hrough DT and NN classifiers o deermine classifiers success rae in raining courses domain. Research revealed ha enhanced NB classifier ouperformed radiional NB classifier. They performed well if no beer han popular, rival echniques. The new sudy revealed ha NB classifier is he bes choice for raining courses domain, achieving a F-Measure value of over 97%, despie being rained wih fewer samples han classificaion sysems encounered. A graph-based semi-supervised learning algorihm applied o he Web page classificaion was proposed by []. The algorihm used a similariy measure beween Web pages o consruc a k-neares neighbor graph. Preliminary experimens on a WebKB daase showed ha he new algorihm exploied unlabeled daa as also labeled ones o ge higher Web page classificaion accuracy. The effec of considering named eniies as web page classificaion feaures was invesigaed by [2]. The ess were in five differen domains baseball, fooball, healh, poliics and science "wih web pages from online news providers. Resuls showed ha incorporaing named eniies leads o sligh gains in classifier performance for narrow domains, bu is no rue for all domains. Resuls showed ha classificaion based on named eniies can be good for some domains (baseball) bu is lower han lexical erms based represenaion. Saraç and Ozel [3] aimed o apply a recen opimizaion echnique called Firefly Algorihm (FA), o choose bes feaures for Web page classificaion. FA seleced a feaures subse and o evaluae seleced feaures finess, he J48 classifier of Weka daa mining ool was used. WebKB and Conference daases evaluaed he proposed feaure selecion sysem s effeciveness. The resul showed ha when a feaures subse was seleced by using FA, WebKB and Conference daases were classified wihou accuracy loss, also as feaures decreased he ime needed o classify he new Web pages, reduced. A GA o selec bes feaures for Web page classificaion problem o improve accuracy and classifiers run ime performance was proposed by [4]. The increased informaion on he Web has raised a need for accurae auomaed classifiers for Web pages o mainain Web direcories and increase search engines' performance. To decrease feaure space, a GA ha deermines bes feaures for a se of Web pages was developed. I was found ha when GA proposed feaures were used, and a knn classifier was employed, accuracy wen up o 96%. A new cenroid-based approach o classify web pages by genre using characer n-grams from differen informaion sources like ile, URL, headings and anchors was proposed by [5]. To deal wih web pages complexiy and web genres rapid evoluion, he new approach implemened a muli-label and adapive classificaion scheme where web pages were classified singly, and each affeced more han one genre. According o similariy beween a new page and every genre cenroid, he new approach eiher adaped he genre cenroid under consideraion or considered he new page as a noise page and discarded i. The resuls showed beer resuls han curren muli-label classifiers. An approach of Web page classificaion using NB classifier based on ICA was proposed by [6]. To perform classificaion, a Web page was firs represened by a feaures vecor wih differen weighs, and a weigh calculaed mehod was improved. As feaures were big, PCA seleced relevan feaures from a preprocessing secion as inpu for improved ICA algorihm (MFICA). Finally, MFICA oupu was sen o a NB classifier for classificaion o boos classifier performance. Evaluaion proved ha he ICA model based NB classifier ensured accepable classificaion accuracy. 34

3 III. METHODOLOGY WebKB daase is used here for evaluaion. Four differen feaure selecion echniques were described and a new feaure selecion mehod using PSO algorihm was proposed. Daase WebKB Sop words Porer Semming TF-IDF Compuaion (TFxIDF) Feaure Selecion using PSO algorihm Classificaion performed using ANN Fig. Flowchar for he proposed mehod WebKB Daa Se WebKB daase [7] is a se of webpages collaed by he World Wide Knowledge Base (WebKB) projec of he CMU ex learning group downloaded from The 4 Universiies Daase Homepage [8]. The pages are colleced from various universiies compuer science deparmens in 997 and manually classified o 7 differen classes, including suden, faculy, saff, deparmen, course, projec, and ohers. For every class, he collecion has web pages from four universiies; i.e. Cornell, Wisconsin Texas, and Washingon universiies, and miscellaneous pages from oher universiies. All 8,282 web pages are classified manually ino he seven caegories so ha he suden caegory has 64 pages, faculy 24, saff 37, deparmen 82, course 930, projec 504, and ohers have 3764 pages. The class oher is a collecion of pages no deemed he main page and do no represen any insance of he earlier six classes. WebKB daase includes 867 web pages from Cornell Universiy, 827 pages from Texas Universiy, 205 from Washingon Universiy, 263 from Wisconsin Universiy, and finally 420 miscellaneous pages from oher universiies. This sudy uses Projec, Faculy, Suden, and Course classes from he WebKB daase. As Saff and Deparmen classes have limied posiive examples, hey are no considered. Sopwords Sop words are filered prior o, or afer, naural language daa processing. Sop words are commonly used words frequenly filered from ex in informaion rerieval asks. When removing sop words, noise are ge ridded of, and space is saved o sore documens. For example, consider an insance "I am a suden of compuer science a Wisconsin Universiy." The sopwords "I", "am", a, of, and "a" are lef ou of he full-ex index. Thus on removal of he sopwords he insance is represened by suden compuer science Wisconsin Universiy. Semming Semming is a ool used in vocabulary mismach problem, where query words do no mach documen words. Semmers conflae cerain varian forms of same word like (paper, papers) and (hold, holds, holding ) [9]. Afer removing high frequency words, indexing conflaes word varians ino same sem or roo using a semming algorihm. For example, words "engineering", "engineers" or "engineered" are reduced o he sem "engineer". Grouping words in informaion rerieval, wih he same roo under same sem (or indexing erm) increases success rae when maching documens o a query [20]. Porer s algorihm is based on seps ha each sep removes a ype of suffix by subsiuion rules. These rules only apply when specific condiions hold; he resuling sem mus have a minimal lengh. Mos rules have a condiion based on a socalled measure. A measure is a number of vowel-consonan sequences (where consecuive vowels/consonans are couned as one) presen in a resuling sem. This condiion mus preven leers which resemble a suffix, bu only a par of he sem is removed [2]. For example, suden, sudens on Porer semming is suden. Similarly, for sudied, sudies, sudy, sudying is sudi. Term Frequency-Inverse Documen Frequency (TF-IDF) Inverse Documen Frequency (IDF) represens scaling facor. If erm occurs frequenly in documens, is IDF value is less as erm has lower discriminaive power [22]. IDF() is defined as (): d IDF log d d is a se of documens wih erm. Similar documens have similar relaive erm frequencies. Similariy is measured among documen ses or beween a documen and query. Cosine measure locaes documens [23]; he cosine measure is go by (2) 2, sim v v 2 2 () vv. v v (2) where v and v 2 are wo documen vecors, v. v 2 defined as vv i i 2i and v v. v. TF-IDF funcion weighs each vecor componen (each relaing o a word of a vocabulary) of every documen as follows. Firs, i incorporaes word frequency in a documen. Thus, he more a word appears in a documen (is TF, erm frequency is high) he more i is hough o be significan in he documen. Also, IDF measures how infrequen a word is in a collecion. This is esimaed using an enire raining ex collecion a hand.tf IDF combines weighs of TF and IDF by muliplying hem. TF gives more weigh o a frequen erm 342

4 in an essay while IDF downscales he weigh, if a erm occurs in many essays [24]. Feaure Selecion Using PSO PSO originaed from he simulaion of birds social behavior in a flock. In PSO, a paricle flies in search space wih a velociy adjused by is flying memory and companion's flying experience. A paricle has is objecive funcion value decided by a finess funcion (3): v wv c r p x c r p x (3) id id id id 2 2 gd id where i represens ih paricle and d is a soluion space dimension, c denoes a cogniion learning facor, and c2 indicaes a social learning facor, r and r2 are random numbers uniformly disribued in (0,), p id and p gd sand for posiion wih bes finess found so far for he ih and he bes posiion in he neighborhood, v id and v id are velociies a ime and ime, and x id is posiion of ih paricle a ime. Every paricle moves o a new poenial soluion based on (4): x x v, d, 2,..., D, id id id A binary PSO where a paricle moves in a sae space resriced o 0 and on each dimension, regarding changes in probabiliies ha a bi will be in one sae or he oher was proposed by [25] as in (5), (6): id, x id, (4), rand() S( v ) (5) 0 S. v e v (6) Funcion S(v) is a sigmoid limiing ransformaion and rand ( ) is a random number from a uniform disribuion in [0.0,.0]. This sudy uses a binary PSO algorihm version for PSO. Each paricle s posiion is given in a binary sring form represening a feaure selecion siuaion. Feaure Selecion Using Proposed PSO The process for proposed PSO is given by. Sar PSO 2. Find gbes and Pbes 3. Each paricle is updaed according o (7) v wv c r p x c r p x (7) id id id id 2 2 gd id 4. Sar random muaion hill climbing for gbes as follows: Random muaion hill climbing is a local search mehod ha has a sochasic componen. i. Choose a binary sring a random. Call his sring bes_evaluaed. ii. Muae a bi chosen a random in bes_evaluaed. iii. Compue he finess of he muaed sring. If he finess is greaer han he finess of bes_evaluaed, hen se bes_evaluaed o he muaed sring. iv. If maximum number of ieraions has been performed reurn bes_evaluaed_ oherwise, go o Sep ii. 5. Sar parallel hill climbing on he new pbes values. Arificial Neural Nework (ANN) Realisic cogniive simulaion and Compuer Assised Learning (CAL) inspired search for opimal learning and eaching mehodology and classical eaching performance. ANN learning model use led o fair assessmens performance of he suggesed learning and eaching opics. So, opimal uoring mehod is reached afer analysis and evaluaion of simulaion resuls. Fig. 2 depics an ANN learning and eaching model s block diagram which presens 2 diverse learning paradigms simulaion. Boh are relaed o ineracive uoring and learning process and self-organized learning. The firs is relaed o classical (uor supervised) learning seen in classrooms (face o face uoring). This moves ineracively hrough a bidirecional communicaion process beween uor and learner(s). The second paradigm underakes self-organized (unsupervised) uoring process. Fig. 2 Generalized ANN block diagram Error vecor referring o Fig. 2, a a ime insan (n) observed during learning process is (8) en ( ) = y( n) dn ( ) (8) where e(n) is error correcing signal conrolling learning process, adapively, x(n) is inpu simulus, y(n) is oupu response vecor, and d(n) desired numeric value(s). Equaion (9) is deduced: T V ( n)= X ( n) W ( n) k j kj Y n V n e - Vk( n) - Vk( n) k( )= ( k( ))=(-e )/( ) e ( n) = d ( n) - y ( n) k k k W ( n)= W ( n) W ( n) kj kj kj where X is inpu vecor, W weigh vecor, an acivaion (odd sigmoid) funcion characerized by λ as gain facor and Y as oupu. e k is error value and dk he desired oupu. Noing ha ΔW kj (n) is a dynamic change of weigh vecor value connecing k h and ih neurons. The weigh vecor value dynamic changes for supervised phase are (0): (9) 343

5 W = e ( n) X ( n) (0) kj k j where, η is learning rae value during learning process. Bu for unsupervised paradigm, dynamic weigh vecor value change is given by (): W = Y ( n) X ( n) () kj k j Noe ha e k (n) in is subsiued by y k (n) a arbirary ime insan (n) during learning. The proposed parallel NN has 2 sub-neworks. Every nework has 2 hidden layers, wih differing ransfer funcions. This sudy uses ransfer funcion sigmoid and anh. Differen funcions advanage is ha muual inerference is reduced during complex asks simulaneous processing and execuion. Table I gives he proposed NN parameers. TABLE I PARAMETERS FOR THE PROPOSED MLP NN Parameer Value Inpu Neuron 50 Oupu Neuron 4 Number of Hidden Layer 2 Number of processing elemens upper 4 Number of processing elemens lower 4 Transfer funcion of hidden layer upper Tanh Transfer funcion of hidden layer lower Sigmoid Learning Rule of hidden layer Momenum Sep size 0. Momenum 0.7 Transfer funcion of oupu layer Tanh Learning Rule of oupu layer Momenum Sep size 0. Momenum 0.7 TABLE II THE TOP 5 WORDS SELECTED BY VARIOUS FEATURE SELECTION TECHNIQUES annual professor annual annual associae science aspecs assignmen compue hu assignmen associae course cornell associae componen define universal componen define develop reurn define direc direc compue direc people geomeric annual noe professor hall deparmenal people research hour research professor reurn noe public research phd overview lecure resume universal page people reurn public people professional phd lecure phd phd universal noe annual professor annual annual associae science aspecs assignmen compue hu assignmen associae course cornell associae componen IV. EXPERIMENTAL RESULTS The proposed PSO based feaure selecion for web page classificaion using HTML ags is calculaed using he 4 Universiies Daase and compared wih Correlaion Feaure Selecion (CFS), Muual Informaion (MI) and PSO feaure exracion mehod. Five classes are classified (Suden, Course, Faculy, Projec, and Ohers). The accuracy, precision, recall and f measure are compued: The op 5 words seleced by various feaure selecion echnique is abulaed in Table II. The Neural Nework classifies he web pages based on he keywords. The following ables and figures show he experimenal resuls in deail. Tables III-V show he average precision, recall and F measure obained for various feaure exracion. The resuls are shown graphically from Figs TABLE III PRECISION Suden Faculy Course Projec Oher Precision suden faculy course projec oher Class label Fig. 3 Precision From Table III and Fig. 3, i is observed ha Precision is improved for Proposed PSO when compared o CFS, MI, and PSO. On an average, Precision increases for Proposed PSO by 2.3% when compared o PSO, by 6.23% when compared o MI and by 5.47% when compared o CFS. For class label course, precision of Proposed PSO increases by 3.94% when compared o PSO. TABLE IV RECALL Suden Faculy Course Projec Oher CFS MI PSO Proposed PSO 344

6 Recall Fig. 4 Recall From Table IV and Fig. 4 i is observed ha Recall is improved for Proposed PSO when compared o CFS, MI, and PSO. On an average, Recall increases for Proposed PSO by 2.94% when compared o PSO, by 6.79% when compared o MI and by 4.7% when compared o CFS. For class label faculy, recall of Proposed PSO increases by 3.82% when compared o CFS. TABLE V F MEASURE Suden Faculy Course Projec Oher f measure suden faculy course projec oher Class label suden faculy course projec oher Class label Fig. 5 f measure From Table V and Fig. 5 i is observed ha f measure is improved for Proposed PSO when compared o CFS, MI, and PSO. On an average, f measure increases for Proposed PSO by 2.65% when compared o PSO, by 6.72% when compared o MI and by 5.30% when compared o CFS. For class label suden, f measure of Proposed PSO increases by 7.54% when compared o MI. V. CONCLUSION Auomaic Web-page classificaion by using hyperex is a big approach o caegorize large Webpage quaniies. Two major approaches were sudied for Web-page classificaion: CFS MI PSO Proposed PSO CFS MI PSO Proposed PSO conen-based and conex-based approaches. Conen-based classificaion mehods use words or phrases of a arge documen o build he classifier and achieve limied accuracy. This sudy proposed a new feaure selecion mehod using PSO algorihm. Resuls from experimens showed ha he new mehod ouperformed oher feaure selecion mehods and ensured good classificaion accuracy. REFERENCES [] Mangai, J. A., & Kumar, V. S. (20). A Novel Approach for Web Page Classificaion using Opimum. IJCSNS, (5), 252. [2] X. Qi and B. D. Davison, Web page classificaion: feaures and algorihms, ACM Compuing Surveys, vol. 4, no. 2, aricle 2, [3] T. M. Michell, Machine Learning, McGraw-Hill, NewYork, NY, USA, s ediion, 997. [4] Golub, K. and A. Ardo (2005, Sepember). Imporance of HTML srucural elemens and meadaa in auomaed subjec classificaion. In Proceedings of he 9h European Conference on Research and Advanced Technology for Digial Libraries (ECDL), Volume 3652 of LNCS, Berlin, pp Springer. [5] C. E. Shannon, A mahemaical heory of communicaion, The Bell Sysem Technical Journal, vol. 27, pp , 948. [6] Y. Yang and J. O. Pedersen, A comparaive sudy on feaure selecion in ex caegorizaion, in Proceedings of he 4 h Inernaional Conference on Machine Learning (ICML 97), pp , Nashville, Tenn, USA, July 997. [7] W. J. Wilbur and K. Sirokin, The auomaic idenificaion of sop words, Journal of Informaion Science, vol. 8,no., pp , [8] Mangai, J. A., & Kumar, V. S. (20). A Novel Approach for Web Page Classificaion using Opimum. IJCSNS, (5), 252. [9] Song, R., Liu, H., Wen, J. R., & Ma, W. Y. (2004, May). Learning block imporance models for web pages. In Proceedings of he 3h inernaional conference on World Wide Web (pp ). ACM. [0] Xhemali, D., Hinde, C. J., & Sone, R. G. (2009). Naive bayes vs. decision rees vs. neural neworks in he classificaion of raining web pages. [] Liu, R., Zhou, J., & Liu, M. (2006, Ocober). Graph-based semisupervised learning algorihm for web page classificaion. In Inelligen Sysems Design and Applicaions, ISDA'06. Sixh Inernaional Conference on (Vol. 2, pp ). IEEE. [2] Samarawickrama, S., & Jayarane, L. (202, Sepember). Effec of Named Eniies in Web Page Classificaion. In Compuaional Inelligence, Modelling and Simulaion (CIMSiM), 202 Fourh Inernaional Conference on (pp ). IEEE. [3] Saraç, E., & Ozel, S. A. (203, June). Web page classificaion using firefly opimizaion. In Innovaions in Inelligen Sysems and Applicaions (INISTA), 203 IEEE Inernaional Symposium on (pp. - 5). IEEE. [4] Ozel, S. A. (20, June). A geneic algorihm based opimal feaure selecion for web page classificaion. In Innovaions in Inelligen Sysems and Applicaions (INISTA), 20 Inernaional Symposium on (pp ). IEEE. [5] Jebari, C., & Wani, M. A. (202, December). A Muli-label and Adapive Genre Classificaion of Web Pages. In Machine Learning and Applicaions (ICMLA), 202 h Inernaional Conference on (Vol., pp ). IEEE. [6] He, Z., & Liu, Z. (2008, Ocober). A Novel Approach o Naïve Bayes Web Page Auomaic Classificaion. In Fuzzy Sysems and Knowledge Discovery, FSKD'08. Fifh Inernaional Conference on (Vol. 2, pp ). IEEE. [7] Sun, A., Lim, E. P., & Ng, W. K. (2002, November). Web classificaion using suppor vecor machine. In Proceedings of he 4h inernaional workshop on Web informaion and daa managemen (pp ). ACM. [8] Kan, M. Y., &Thi, H. O. N. (2005, Ocober). Fas webpage classificaion using URL feaures. In Proceedings of he 4h ACM inernaional conference on Informaion and knowledge managemen (pp ). ACM. [9] Larkey, L. S., Balleseros, L., & Connell, M. E. (2002, Augus). Improving semming for Arabic informaion rerieval: ligh semming and co-occurrence analysis. In Proceedings of he 25h annual 345

7 inernaional ACM SIGIR conference on Research and developmen in informaion rerieval (pp ). ACM. [20] Savoy, J. (999). A semming procedure and sopword lis for general French corpora. JASIS, 50(0), [2] Kraaij, W., & Pohlmann, R. (994). Porer s semming algorihm for Duch. Informaieweenschap, [22] Papineni, K. (200, June). Why inverse documen frequency?. In Proceedings of he second meeing of he Norh American Chaper of he Associaion for Compuaional Linguisics on Language echnologies (pp. -8). Associaion for Compuaional Linguisics. [23] Nigam, K., McCallum, A. K., Thrun, S., & Michell, T. (2000). Tex classificaion from labeled and unlabeled documens using EM. Machine learning, 39(2), [24] Soucy, P., & Mineau, G. W. (2005, July). Beyond TFIDF weighing for ex caegorizaion in he vecor space model. In IJCAI (Vol. 5, pp ). [25] Kennedy, J.; Eberhar, R.C., A discree binary version of he paricle swarm algorihm, Sysems, Man, and Cyberneics, 997. 'Compuaional Cyberneics and Simulaion'., 997 IEEE Inernaional Conference on Volume 5, 2-5 Oc. 997 Page(s): vol.5. Leela Devi B is wih he School of Compuer Applicaions a Professional Group of Insiuions, Palladam. She is currenly pursuing his docorae in India. Sankar A is currenly working in PSG College of Technology, Coimbaore, India. 346

Handling uncertainty in semantic information retrieval process

Handling uncertainty in semantic information retrieval process Handling uncerainy in semanic informaion rerieval process Chkiwa Mounira 1, Jedidi Anis 1 and Faiez Gargouri 1 1 Mulimedia, InfoRmaion sysems and Advanced Compuing Laboraory Sfax Universiy, Tunisia m.chkiwa@gmail.com,

More information

Learning in Games via Opponent Strategy Estimation and Policy Search

Learning in Games via Opponent Strategy Estimation and Policy Search Learning in Games via Opponen Sraegy Esimaion and Policy Search Yavar Naddaf Deparmen of Compuer Science Universiy of Briish Columbia Vancouver, BC yavar@naddaf.name Nando de Freias (Supervisor) Deparmen

More information

A Matching Algorithm for Content-Based Image Retrieval

A Matching Algorithm for Content-Based Image Retrieval A Maching Algorihm for Conen-Based Image Rerieval Sue J. Cho Deparmen of Compuer Science Seoul Naional Universiy Seoul, Korea Absrac Conen-based image rerieval sysem rerieves an image from a daabase using

More information

IntentSearch:Capturing User Intention for One-Click Internet Image Search

IntentSearch:Capturing User Intention for One-Click Internet Image Search JOURNAL OF L A T E X CLASS FILES, VOL. 6, NO. 1, JANUARY 2010 1 InenSearch:Capuring User Inenion for One-Click Inerne Image Search Xiaoou Tang, Fellow, IEEE, Ke Liu, Jingyu Cui, Suden Member, IEEE, Fang

More information

FIELD PROGRAMMABLE GATE ARRAY (FPGA) AS A NEW APPROACH TO IMPLEMENT THE CHAOTIC GENERATORS

FIELD PROGRAMMABLE GATE ARRAY (FPGA) AS A NEW APPROACH TO IMPLEMENT THE CHAOTIC GENERATORS FIELD PROGRAMMABLE GATE ARRAY (FPGA) AS A NEW APPROACH TO IMPLEMENT THE CHAOTIC GENERATORS Mohammed A. Aseeri and M. I. Sobhy Deparmen of Elecronics, The Universiy of Ken a Canerbury Canerbury, Ken, CT2

More information

Analysis of Various Types of Bugs in the Object Oriented Java Script Language Coding

Analysis of Various Types of Bugs in the Object Oriented Java Script Language Coding Indian Journal of Science and Technology, Vol 8(21), DOI: 10.17485/ijs/2015/v8i21/69958, Sepember 2015 ISSN (Prin) : 0974-6846 ISSN (Online) : 0974-5645 Analysis of Various Types of Bugs in he Objec Oriened

More information

Image segmentation. Motivation. Objective. Definitions. A classification of segmentation techniques. Assumptions for thresholding

Image segmentation. Motivation. Objective. Definitions. A classification of segmentation techniques. Assumptions for thresholding Moivaion Image segmenaion Which pixels belong o he same objec in an image/video sequence? (spaial segmenaion) Which frames belong o he same video sho? (emporal segmenaion) Which frames belong o he same

More information

Improving Ranking of Search Engines Results Based on Power Links

Improving Ranking of Search Engines Results Based on Power Links IPASJ Inernaional Journal of Informaion Technology (IIJIT) Web Sie: hp://www.ipasj.org/iijit/iijit.hm A Publisher for Research Moivaion... Email: edioriiji@ipasj.org Volume 2, Issue 9, Sepember 2014 ISSN

More information

Implementing Ray Casting in Tetrahedral Meshes with Programmable Graphics Hardware (Technical Report)

Implementing Ray Casting in Tetrahedral Meshes with Programmable Graphics Hardware (Technical Report) Implemening Ray Casing in Terahedral Meshes wih Programmable Graphics Hardware (Technical Repor) Marin Kraus, Thomas Erl March 28, 2002 1 Inroducion Alhough cell-projecion, e.g., [3, 2], and resampling,

More information

CHANGE DETECTION - CELLULAR AUTOMATA METHOD FOR URBAN GROWTH MODELING

CHANGE DETECTION - CELLULAR AUTOMATA METHOD FOR URBAN GROWTH MODELING CHANGE DETECTION - CELLULAR AUTOMATA METHOD FOR URBAN GROWTH MODELING Sharaf Alkheder, Jun Wang and Jie Shan Geomaics Engineering, School of Civil Engineering, Purdue Universiy 550 Sadium Mall Drive, Wes

More information

Simple Network Management Based on PHP and SNMP

Simple Network Management Based on PHP and SNMP Simple Nework Managemen Based on PHP and SNMP Krasimir Trichkov, Elisavea Trichkova bsrac: This paper aims o presen simple mehod for nework managemen based on SNMP - managemen of Cisco rouer. The paper

More information

Design Alternatives for a Thin Lens Spatial Integrator Array

Design Alternatives for a Thin Lens Spatial Integrator Array Egyp. J. Solids, Vol. (7), No. (), (004) 75 Design Alernaives for a Thin Lens Spaial Inegraor Array Hala Kamal *, Daniel V azquez and Javier Alda and E. Bernabeu Opics Deparmen. Universiy Compluense of

More information

STEREO PLANE MATCHING TECHNIQUE

STEREO PLANE MATCHING TECHNIQUE STEREO PLANE MATCHING TECHNIQUE Commission III KEY WORDS: Sereo Maching, Surface Modeling, Projecive Transformaion, Homography ABSTRACT: This paper presens a new ype of sereo maching algorihm called Sereo

More information

Open Access Research on an Improved Medical Image Enhancement Algorithm Based on P-M Model. Luo Aijing 1 and Yin Jin 2,* u = div( c u ) u

Open Access Research on an Improved Medical Image Enhancement Algorithm Based on P-M Model. Luo Aijing 1 and Yin Jin 2,* u = div( c u ) u Send Orders for Reprins o reprins@benhamscience.ae The Open Biomedical Engineering Journal, 5, 9, 9-3 9 Open Access Research on an Improved Medical Image Enhancemen Algorihm Based on P-M Model Luo Aijing

More information

Web System for the Remote Control and Execution of an IEC Application

Web System for the Remote Control and Execution of an IEC Application Web Sysem for he Remoe Conrol and Execuion of an IEC 61499 Applicaion Oana ROHAT, Dan POPESCU Faculy of Auomaion and Compuer Science, Poliehnica Universiy, Splaiul Independenței 313, Bucureși, 060042,

More information

Difficulty-aware Hybrid Search in Peer-to-Peer Networks

Difficulty-aware Hybrid Search in Peer-to-Peer Networks Difficuly-aware Hybrid Search in Peer-o-Peer Neworks Hanhua Chen, Hai Jin, Yunhao Liu, Lionel M. Ni School of Compuer Science and Technology Huazhong Univ. of Science and Technology {chenhanhua, hjin}@hus.edu.cn

More information

CAMERA CALIBRATION BY REGISTRATION STEREO RECONSTRUCTION TO 3D MODEL

CAMERA CALIBRATION BY REGISTRATION STEREO RECONSTRUCTION TO 3D MODEL CAMERA CALIBRATION BY REGISTRATION STEREO RECONSTRUCTION TO 3D MODEL Klečka Jan Docoral Degree Programme (1), FEEC BUT E-mail: xkleck01@sud.feec.vubr.cz Supervised by: Horák Karel E-mail: horak@feec.vubr.cz

More information

Real Time Integral-Based Structural Health Monitoring

Real Time Integral-Based Structural Health Monitoring Real Time Inegral-Based Srucural Healh Monioring The nd Inernaional Conference on Sensing Technology ICST 7 J. G. Chase, I. Singh-Leve, C. E. Hann, X. Chen Deparmen of Mechanical Engineering, Universiy

More information

MORPHOLOGICAL SEGMENTATION OF IMAGE SEQUENCES

MORPHOLOGICAL SEGMENTATION OF IMAGE SEQUENCES MORPHOLOGICAL SEGMENTATION OF IMAGE SEQUENCES B. MARCOTEGUI and F. MEYER Ecole des Mines de Paris, Cenre de Morphologie Mahémaique, 35, rue Sain-Honoré, F 77305 Fonainebleau Cedex, France Absrac. In image

More information

COSC 3213: Computer Networks I Chapter 6 Handout # 7

COSC 3213: Computer Networks I Chapter 6 Handout # 7 COSC 3213: Compuer Neworks I Chaper 6 Handou # 7 Insrucor: Dr. Marvin Mandelbaum Deparmen of Compuer Science York Universiy F05 Secion A Medium Access Conrol (MAC) Topics: 1. Muliple Access Communicaions:

More information

Relevance Ranking using Kernels

Relevance Ranking using Kernels Relevance Ranking using Kernels Jun Xu 1, Hang Li 1, and Chaoliang Zhong 2 1 Microsof Research Asia, 4F Sigma Cener, No. 49 Zhichun Road, Beijing, China 100190 2 Beijing Universiy of Poss and Telecommunicaions,

More information

Video Content Description Using Fuzzy Spatio-Temporal Relations

Video Content Description Using Fuzzy Spatio-Temporal Relations Proceedings of he 4s Hawaii Inernaional Conference on Sysem Sciences - 008 Video Conen Descripion Using Fuzzy Spaio-Temporal Relaions rchana M. Rajurkar *, R.C. Joshi and Sananu Chaudhary 3 Dep of Compuer

More information

Shortest Path Algorithms. Lecture I: Shortest Path Algorithms. Example. Graphs and Matrices. Setting: Dr Kieran T. Herley.

Shortest Path Algorithms. Lecture I: Shortest Path Algorithms. Example. Graphs and Matrices. Setting: Dr Kieran T. Herley. Shores Pah Algorihms Background Seing: Lecure I: Shores Pah Algorihms Dr Kieran T. Herle Deparmen of Compuer Science Universi College Cork Ocober 201 direced graph, real edge weighs Le he lengh of a pah

More information

EECS 487: Interactive Computer Graphics

EECS 487: Interactive Computer Graphics EECS 487: Ineracive Compuer Graphics Lecure 7: B-splines curves Raional Bézier and NURBS Cubic Splines A represenaion of cubic spline consiss of: four conrol poins (why four?) hese are compleely user specified

More information

An efficient approach to improve throughput for TCP vegas in ad hoc network

An efficient approach to improve throughput for TCP vegas in ad hoc network Inernaional Research Journal of Engineering and Technology (IRJET) e-issn: 395-0056 Volume: 0 Issue: 03 June-05 www.irje.ne p-issn: 395-007 An efficien approach o improve hroughpu for TCP vegas in ad hoc

More information

Win, Yuzana; Masada, Tomonari. Citation. Issue Date Right. this work in other works.

Win, Yuzana; Masada, Tomonari. Citation. Issue Date Right. this work in other works. NAOSITE: Nagasaki Universiy's Ac Tile Exploring Technical Phrase Frames f Auhor(s) Ciaion Win, Yuzana; Masada, Tomonari 05 IEEE 9h Inernaional Confer Neworking and Applicaions Worksho Issue Dae 05-04-7

More information

Time Expression Recognition Using a Constituent-based Tagging Scheme

Time Expression Recognition Using a Constituent-based Tagging Scheme Track: Web Conen Analysis, Semanics and Knowledge Time Expression Recogniion Using a Consiuen-based Tagging Scheme Xiaoshi Zhong and Erik Cambria School of Compuer Science and Engineering Nanyang Technological

More information

A time-space consistency solution for hardware-in-the-loop simulation system

A time-space consistency solution for hardware-in-the-loop simulation system Inernaional Conference on Advanced Elecronic Science and Technology (AEST 206) A ime-space consisency soluion for hardware-in-he-loop simulaion sysem Zexin Jiang a Elecric Power Research Insiue of Guangdong

More information

Scheduling. Scheduling. EDA421/DIT171 - Parallel and Distributed Real-Time Systems, Chalmers/GU, 2011/2012 Lecture #4 Updated March 16, 2012

Scheduling. Scheduling. EDA421/DIT171 - Parallel and Distributed Real-Time Systems, Chalmers/GU, 2011/2012 Lecture #4 Updated March 16, 2012 EDA421/DIT171 - Parallel and Disribued Real-Time Sysems, Chalmers/GU, 2011/2012 Lecure #4 Updaed March 16, 2012 Aemps o mee applicaion consrains should be done in a proacive way hrough scheduling. Schedule

More information

Assignment 2. Due Monday Feb. 12, 10:00pm.

Assignment 2. Due Monday Feb. 12, 10:00pm. Faculy of rs and Science Universiy of Torono CSC 358 - Inroducion o Compuer Neworks, Winer 218, LEC11 ssignmen 2 Due Monday Feb. 12, 1:pm. 1 Quesion 1 (2 Poins): Go-ack n RQ In his quesion, we review how

More information

Coded Caching with Multiple File Requests

Coded Caching with Multiple File Requests Coded Caching wih Muliple File Requess Yi-Peng Wei Sennur Ulukus Deparmen of Elecrical and Compuer Engineering Universiy of Maryland College Park, MD 20742 ypwei@umd.edu ulukus@umd.edu Absrac We sudy a

More information

Hidden Markov Model and Chapman Kolmogrov for Protein Structures Prediction from Images

Hidden Markov Model and Chapman Kolmogrov for Protein Structures Prediction from Images Hidden Markov Model and Chapman Kolmogrov for Proein Srucures Predicion from Images Md.Sarwar Kamal 1, Linkon Chowdhury 2, Mohammad Ibrahim Khan 2, Amira S. Ashour 3, João Manuel R.S. Tavares 4, Nilanjan

More information

Design and Application of Computer-aided English Online Examination System NONG DeChang 1, a

Design and Application of Computer-aided English Online Examination System NONG DeChang 1, a 3rd Inernaional Conference on Maerials Engineering, Manufacuring Technology and Conrol (ICMEMTC 2016) Design and Applicaion of Compuer-aided English Online Examinaion Sysem NONG DeChang 1, a 1,2 Guangxi

More information

The University of Sheffield Department of Computer Science. Indexing XML Databases: Classifications, Problems Identification and a New Approach

The University of Sheffield Department of Computer Science. Indexing XML Databases: Classifications, Problems Identification and a New Approach The Universiy of Sheffield Deparmen of Compuer Science Indexing XML Daabases: Classificaions, Problems Idenificaion and a New Approach Research Memorandum CS-7-5 Mohammed Al-Badawi Compuer Science Dep

More information

LOW-VELOCITY IMPACT LOCALIZATION OF THE COMPOSITE TUBE USING A NORMALIZED CROSS-CORRELATION METHOD

LOW-VELOCITY IMPACT LOCALIZATION OF THE COMPOSITE TUBE USING A NORMALIZED CROSS-CORRELATION METHOD 21 s Inernaional Conference on Composie Maerials Xi an, 20-25 h Augus 2017 LOW-VELOCITY IMPACT LOCALIZATION OF THE COMPOSITE TUBE USING A NORMALIZED CROSS-CORRELATION METHOD Hyunseok Kwon 1, Yurim Park

More information

A Hierarchical Object Recognition System Based on Multi-scale Principal Curvature Regions

A Hierarchical Object Recognition System Based on Multi-scale Principal Curvature Regions A Hierarchical Objec Recogniion Sysem Based on Muli-scale Principal Curvaure Regions Wei Zhang, Hongli Deng, Thomas G Dieerich and Eric N Morensen School of Elecrical Engineering and Compuer Science Oregon

More information

Audio Engineering Society. Convention Paper. Presented at the 119th Convention 2005 October 7 10 New York, New York USA

Audio Engineering Society. Convention Paper. Presented at the 119th Convention 2005 October 7 10 New York, New York USA Audio Engineering Sociey Convenion Paper Presened a he 119h Convenion 2005 Ocober 7 10 New Yor, New Yor USA This convenion paper has been reproduced from he auhor's advance manuscrip, wihou ediing, correcions,

More information

Motion Level-of-Detail: A Simplification Method on Crowd Scene

Motion Level-of-Detail: A Simplification Method on Crowd Scene Moion Level-of-Deail: A Simplificaion Mehod on Crowd Scene Absrac Junghyun Ahn VR lab, EECS, KAIST ChocChoggi@vr.kais.ac.kr hp://vr.kais.ac.kr/~zhaoyue Recen echnological improvemen in characer animaion

More information

Dynamic Route Planning and Obstacle Avoidance Model for Unmanned Aerial Vehicles

Dynamic Route Planning and Obstacle Avoidance Model for Unmanned Aerial Vehicles Volume 116 No. 24 2017, 315-329 ISSN: 1311-8080 (prined version); ISSN: 1314-3395 (on-line version) url: hp://www.ijpam.eu ijpam.eu Dynamic Roue Planning and Obsacle Avoidance Model for Unmanned Aerial

More information

Representing Non-Manifold Shapes in Arbitrary Dimensions

Representing Non-Manifold Shapes in Arbitrary Dimensions Represening Non-Manifold Shapes in Arbirary Dimensions Leila De Floriani,2 and Annie Hui 2 DISI, Universiy of Genova, Via Dodecaneso, 35-646 Genova (Ialy). 2 Deparmen of Compuer Science, Universiy of Maryland,

More information

STRING DESCRIPTIONS OF DATA FOR DISPLAY*

STRING DESCRIPTIONS OF DATA FOR DISPLAY* SLAC-PUB-383 January 1968 STRING DESCRIPTIONS OF DATA FOR DISPLAY* J. E. George and W. F. Miller Compuer Science Deparmen and Sanford Linear Acceleraor Cener Sanford Universiy Sanford, California Absrac

More information

Moving Object Detection Using MRF Model and Entropy based Adaptive Thresholding

Moving Object Detection Using MRF Model and Entropy based Adaptive Thresholding Moving Objec Deecion Using MRF Model and Enropy based Adapive Thresholding Badri Narayan Subudhi, Pradipa Kumar Nanda and Ashish Ghosh Machine Inelligence Uni, Indian Saisical Insiue, Kolkaa, 700108, India,

More information

Evaluation and Improvement of Region-based Motion Segmentation

Evaluation and Improvement of Region-based Motion Segmentation Evaluaion and Improvemen of Region-based Moion Segmenaion Mark Ross Universiy Koblenz-Landau, Insiue of Compuaional Visualisics, Universiässraße 1, 56070 Koblenz, Germany Email: ross@uni-koblenz.de Absrac

More information

Robust Multi-view Face Detection Using Error Correcting Output Codes

Robust Multi-view Face Detection Using Error Correcting Output Codes Robus Muli-view Face Deecion Using Error Correcing Oupu Codes Hongming Zhang,2, Wen GaoP P, Xilin Chen 2, Shiguang Shan 2, and Debin Zhao Deparmen of Compuer Science and Engineering, Harbin Insiue of Technolog

More information

Lecture 18: Mix net Voting Systems

Lecture 18: Mix net Voting Systems 6.897: Advanced Topics in Crypography Apr 9, 2004 Lecure 18: Mix ne Voing Sysems Scribed by: Yael Tauman Kalai 1 Inroducion In he previous lecure, we defined he noion of an elecronic voing sysem, and specified

More information

Weighted Voting in 3D Random Forest Segmentation

Weighted Voting in 3D Random Forest Segmentation Weighed Voing in 3D Random Fores Segmenaion M. Yaqub,, P. Mahon 3, M. K. Javaid, C. Cooper, J. A. Noble NDORMS, Universiy of Oxford, IBME, Deparmen of Engineering Science, Universiy of Oxford, 3 MRC Epidemiology

More information

4 Error Control. 4.1 Issues with Reliable Protocols

4 Error Control. 4.1 Issues with Reliable Protocols 4 Error Conrol Jus abou all communicaion sysems aemp o ensure ha he daa ges o he oher end of he link wihou errors. Since i s impossible o build an error-free physical layer (alhough some shor links can

More information

Visually Summarizing the Web using Internal Images and Keyphrases

Visually Summarizing the Web using Internal Images and Keyphrases Visually Summarizing he Web using Inernal Images and Keyphrases M.V.Gedam, S. A. Taale Deparmen of compuer engineering, PUNE Universiy Vidya Praishhan s College of Engg., India Absrac Visual summarizaion

More information

Visual Indoor Localization with a Floor-Plan Map

Visual Indoor Localization with a Floor-Plan Map Visual Indoor Localizaion wih a Floor-Plan Map Hang Chu Dep. of ECE Cornell Universiy Ihaca, NY 14850 hc772@cornell.edu Absrac In his repor, a indoor localizaion mehod is presened. The mehod akes firsperson

More information

An Improved Square-Root Nyquist Shaping Filter

An Improved Square-Root Nyquist Shaping Filter An Improved Square-Roo Nyquis Shaping Filer fred harris San Diego Sae Universiy fred.harris@sdsu.edu Sridhar Seshagiri San Diego Sae Universiy Seshigar.@engineering.sdsu.edu Chris Dick Xilinx Corp. chris.dick@xilinx.com

More information

An Adaptive Spatial Depth Filter for 3D Rendering IP

An Adaptive Spatial Depth Filter for 3D Rendering IP JOURNAL OF SEMICONDUCTOR TECHNOLOGY AND SCIENCE, VOL.3, NO. 4, DECEMBER, 23 175 An Adapive Spaial Deph Filer for 3D Rendering IP Chang-Hyo Yu and Lee-Sup Kim Absrac In his paper, we presen a new mehod

More information

IAJIT First Online Publication

IAJIT First Online Publication An Improved Feaure Exracion and Combinaion of Muliple Classifiers for Query-by- ming Naha Phiwma and Parinya Sanguansa 2 Deparmen of Compuer Science, Suan Dusi Rajabha Universiy, Thailand 2 Faculy of Engineering

More information

A Face Detection Method Based on Skin Color Model

A Face Detection Method Based on Skin Color Model A Face Deecion Mehod Based on Skin Color Model Dazhi Zhang Boying Wu Jiebao Sun Qinglei Liao Deparmen of Mahemaics Harbin Insiue of Technology Harbin China 150000 Zhang_dz@163.com mahwby@hi.edu.cn sunjiebao@om.com

More information

Partition-based document identifier assignment (PBDIA) algorithm. (long queries)

Partition-based document identifier assignment (PBDIA) algorithm. (long queries) ( ) Pariion-based documen idenifier assignmen (PBDIA) algorihm PBDIA (long queries) (parallel IR) :,,,, d-gap Compressing an invered file can grealy improve query performance of an informaion rerieval

More information

Voltair Version 2.5 Release Notes (January, 2018)

Voltair Version 2.5 Release Notes (January, 2018) Volair Version 2.5 Release Noes (January, 2018) Inroducion 25-Seven s new Firmware Updae 2.5 for he Volair processor is par of our coninuing effors o improve Volair wih new feaures and capabiliies. For

More information

Nonparametric CUSUM Charts for Process Variability

Nonparametric CUSUM Charts for Process Variability Journal of Academia and Indusrial Research (JAIR) Volume 3, Issue June 4 53 REEARCH ARTICLE IN: 78-53 Nonparameric CUUM Chars for Process Variabiliy D.M. Zombade and V.B. Ghue * Dep. of aisics, Walchand

More information

Mobile Robots Mapping

Mobile Robots Mapping Mobile Robos Mapping 1 Roboics is Easy conrol behavior percepion modelling domain model environmen model informaion exracion raw daa planning ask cogniion reasoning pah planning navigaion pah execuion

More information

Sam knows that his MP3 player has 40% of its battery life left and that the battery charges by an additional 12 percentage points every 15 minutes.

Sam knows that his MP3 player has 40% of its battery life left and that the battery charges by an additional 12 percentage points every 15 minutes. 8.F Baery Charging Task Sam wans o ake his MP3 player and his video game player on a car rip. An hour before hey plan o leave, he realized ha he forgo o charge he baeries las nigh. A ha poin, he plugged

More information

AML710 CAD LECTURE 11 SPACE CURVES. Space Curves Intrinsic properties Synthetic curves

AML710 CAD LECTURE 11 SPACE CURVES. Space Curves Intrinsic properties Synthetic curves AML7 CAD LECTURE Space Curves Inrinsic properies Synheic curves A curve which may pass hrough any region of hreedimensional space, as conrased o a plane curve which mus lie on a single plane. Space curves

More information

Network management and QoS provisioning - QoS in Frame Relay. . packet switching with virtual circuit service (virtual circuits are bidirectional);

Network management and QoS provisioning - QoS in Frame Relay. . packet switching with virtual circuit service (virtual circuits are bidirectional); QoS in Frame Relay Frame relay characerisics are:. packe swiching wih virual circui service (virual circuis are bidirecional);. labels are called DLCI (Daa Link Connecion Idenifier);. for connecion is

More information

Reinforcement Learning by Policy Improvement. Making Use of Experiences of The Other Tasks. Hajime Kimura and Shigenobu Kobayashi

Reinforcement Learning by Policy Improvement. Making Use of Experiences of The Other Tasks. Hajime Kimura and Shigenobu Kobayashi Reinforcemen Learning by Policy Improvemen Making Use of Experiences of The Oher Tasks Hajime Kimura and Shigenobu Kobayashi Tokyo Insiue of Technology, JAPAN genfe.dis.iech.ac.jp, kobayasidis.iech.ac.jp

More information

Quantitative macro models feature an infinite number of periods A more realistic (?) view of time

Quantitative macro models feature an infinite number of periods A more realistic (?) view of time INFINIE-HORIZON CONSUMPION-SAVINGS MODEL SEPEMBER, Inroducion BASICS Quaniaive macro models feaure an infinie number of periods A more realisic (?) view of ime Infinie number of periods A meaphor for many

More information

Managing XML Versions and Replicas in a P2P Context

Managing XML Versions and Replicas in a P2P Context Managing XML Versions and Replicas in a P2P Conex Deise de Brum Saccol1 1,2, Nina Edelweiss 2, Renaa de Maos Galane 2,4, Carlo Zaniolo 3 2 Insiuo de Informáica - Universidade Federal do Rio Grande do Sul

More information

A GRAPHICS PROCESSING UNIT IMPLEMENTATION OF THE PARTICLE FILTER

A GRAPHICS PROCESSING UNIT IMPLEMENTATION OF THE PARTICLE FILTER A GRAPHICS PROCESSING UNIT IMPLEMENTATION OF THE PARTICLE FILTER ABSTRACT Modern graphics cards for compuers, and especially heir graphics processing unis (GPUs), are designed for fas rendering of graphics.

More information

Outline. EECS Components and Design Techniques for Digital Systems. Lec 06 Using FSMs Review: Typical Controller: state

Outline. EECS Components and Design Techniques for Digital Systems. Lec 06 Using FSMs Review: Typical Controller: state Ouline EECS 5 - Componens and Design Techniques for Digial Sysems Lec 6 Using FSMs 9-3-7 Review FSMs Mapping o FPGAs Typical uses of FSMs Synchronous Seq. Circuis safe composiion Timing FSMs in verilog

More information

Gender Classification of Faces Using Adaboost*

Gender Classification of Faces Using Adaboost* Gender Classificaion of Faces Using Adaboos* Rodrigo Verschae 1,2,3, Javier Ruiz-del-Solar 1,2, and Mauricio Correa 1,2 1 Deparmen of Elecrical Engineering, Universidad de Chile 2 Cener for Web Research,

More information

CENG 477 Introduction to Computer Graphics. Modeling Transformations

CENG 477 Introduction to Computer Graphics. Modeling Transformations CENG 477 Inroducion o Compuer Graphics Modeling Transformaions Modeling Transformaions Model coordinaes o World coordinaes: Model coordinaes: All shapes wih heir local coordinaes and sies. world World

More information

A Web Browsing Traffic Model for Simulation: Measurement and Analysis

A Web Browsing Traffic Model for Simulation: Measurement and Analysis A Web Browsing Traffic Model for Simulaion: Measuremen and Analysis Lourens O. Walers Daa Neworks Archiecure Group Universiy of Cape Town Privae Bag, Rondebosch, 7701 Tel: (021) 650 2663, Fax: (021) 689

More information

CONTEXT MODELS FOR CRF-BASED CLASSIFICATION OF MULTITEMPORAL REMOTE SENSING DATA

CONTEXT MODELS FOR CRF-BASED CLASSIFICATION OF MULTITEMPORAL REMOTE SENSING DATA ISPRS Annals of he Phoogrammery, Remoe Sensing and Spaial Informaion Sciences, Volume I-7, 2012 XXII ISPRS Congress, 25 Augus 01 Sepember 2012, Melbourne, Ausralia CONTEXT MODELS FOR CRF-BASED CLASSIFICATION

More information

The Impact of Product Development on the Lifecycle of Defects

The Impact of Product Development on the Lifecycle of Defects The Impac of Produc Developmen on he Lifecycle of Rudolf Ramler Sofware Compeence Cener Hagenberg Sofware Park 21 A-4232 Hagenberg, Ausria +43 7236 3343 872 rudolf.ramler@scch.a ABSTRACT This paper invesigaes

More information

Improved TLD Algorithm for Face Tracking

Improved TLD Algorithm for Face Tracking Absrac Improved TLD Algorihm for Face Tracking Huimin Li a, Chaojing Yu b and Jing Chen c Chongqing Universiy of Poss and Telecommunicaions, Chongqing 400065, China a li.huimin666@163.com, b 15023299065@163.com,

More information

Multi-task Emoji Learning

Multi-task Emoji Learning Muli-ask Emoji Learning Francesco Barbieri Luís Marujo Pradeep Karuuri William Brendel Large Scale Tex Undersanding Sysems Lab, TALN, UPF, Barcelona, Spain Snap Inc. Research, Venice, California, USA {name.surname}@upf.edu,

More information

J. Vis. Commun. Image R.

J. Vis. Commun. Image R. J. Vis. Commun. Image R. 20 (2009) 9 27 Conens liss available a ScienceDirec J. Vis. Commun. Image R. journal homepage: www.elsevier.com/locae/jvci Face deecion and racking using a Boosed Adapive Paricle

More information

Parallel and Distributed Systems for Constructive Neural Network Learning*

Parallel and Distributed Systems for Constructive Neural Network Learning* Parallel and Disribued Sysems for Consrucive Neural Nework Learning* J. Flecher Z. Obradovi School of Elecrical Engineering and Compuer Science Washingon Sae Universiy Pullman WA 99164-2752 Absrac A consrucive

More information

An accurate Alzheimer disease diagnosis approach based on samples balanced genetic algorithm and extreme learning machine using MRI.

An accurate Alzheimer disease diagnosis approach based on samples balanced genetic algorithm and extreme learning machine using MRI. Curr Neurobiol 2016; 7 (3): 90-99 ISSN 0975-9042 An accurae Alzheimer disease diagnosis approach based on samples balanced geneic algorihm and exreme learning machine using MRI. Vasily Sachnev* School

More information

IROS 2015 Workshop on On-line decision-making in multi-robot coordination (DEMUR 15)

IROS 2015 Workshop on On-line decision-making in multi-robot coordination (DEMUR 15) IROS 2015 Workshop on On-line decision-making in muli-robo coordinaion () OPTIMIZATION-BASED COOPERATIVE MULTI-ROBOT TARGET TRACKING WITH REASONING ABOUT OCCLUSIONS KAROL HAUSMAN a,, GREGORY KAHN b, SACHIN

More information

COMP26120: Algorithms and Imperative Programming

COMP26120: Algorithms and Imperative Programming COMP26120 ecure C3 1/48 COMP26120: Algorihms and Imperaive Programming ecure C3: C - Recursive Daa Srucures Pee Jinks School of Compuer Science, Universiy of Mancheser Auumn 2011 COMP26120 ecure C3 2/48

More information

Joint Feature Learning With Robust Local Ternary Pattern for Face Recognition

Joint Feature Learning With Robust Local Ternary Pattern for Face Recognition Join Feaure Learning Wih Robus Local Ternary Paern for Face Recogniion Yuvaraju.M 1, Shalini.S 1 Assisan Professor, Deparmen of Elecrical and Elecronics Engineering, Anna Universiy Regional Campus, Coimbaore,

More information

Occlusion-Free Hand Motion Tracking by Multiple Cameras and Particle Filtering with Prediction

Occlusion-Free Hand Motion Tracking by Multiple Cameras and Particle Filtering with Prediction 58 IJCSNS Inernaional Journal of Compuer Science and Nework Securiy, VOL.6 No.10, Ocober 006 Occlusion-Free Hand Moion Tracking by Muliple Cameras and Paricle Filering wih Predicion Makoo Kao, and Gang

More information

Image Based Computer-Aided Manufacturing Technology

Image Based Computer-Aided Manufacturing Technology Sensors & Transducers 03 by IFSA hp://www.sensorsporal.com Image Based Compuer-Aided Manufacuring Technology Zhanqi HU Xiaoqin ZHANG Jinze LI Wei LI College of Mechanical Engineering Yanshan Universiy

More information

Video-Based Face Recognition Using Probabilistic Appearance Manifolds

Video-Based Face Recognition Using Probabilistic Appearance Manifolds Video-Based Face Recogniion Using Probabilisic Appearance Manifolds Kuang-Chih Lee Jeffrey Ho Ming-Hsuan Yang David Kriegman klee10@uiuc.edu jho@cs.ucsd.edu myang@honda-ri.com kriegman@cs.ucsd.edu Compuer

More information

I. INTRODUCTION. Keywords -- Web Server, Perceived User Latency, HTTP, Local Measuring. interchangeably.

I. INTRODUCTION. Keywords -- Web Server, Perceived User Latency, HTTP, Local Measuring. interchangeably. Evaluaing Web User Perceived Laency Using Server Side Measuremens Marik Marshak 1 and Hanoch Levy School of Compuer Science Tel Aviv Universiy, Tel-Aviv, Israel mmarshak@emc.com, hanoch@pos.au.ac.il 1

More information

Less Pessimistic Worst-Case Delay Analysis for Packet-Switched Networks

Less Pessimistic Worst-Case Delay Analysis for Packet-Switched Networks Less Pessimisic Wors-Case Delay Analysis for Packe-Swiched Neworks Maias Wecksén Cenre for Research on Embedded Sysems P O Box 823 SE-31 18 Halmsad maias.wecksen@hh.se Magnus Jonsson Cenre for Research

More information

Video streaming over Vajda Tamás

Video streaming over Vajda Tamás Video sreaming over 802.11 Vajda Tamás Video No all bis are creaed equal Group of Picures (GoP) Video Sequence Slice Macroblock Picure (Frame) Inra (I) frames, Prediced (P) Frames or Bidirecional (B) Frames.

More information

A Bayesian Approach to Video Object Segmentation via Merging 3D Watershed Volumes

A Bayesian Approach to Video Object Segmentation via Merging 3D Watershed Volumes A Bayesian Approach o Video Objec Segmenaion via Merging 3D Waershed Volumes Yu-Pao Tsai 1,3, Chih-Chuan Lai 1,2, Yi-Ping Hung 1,2, and Zen-Chung Shih 3 1 Insiue of Informaion Science, Academia Sinica,

More information

User Adjustable Process Scheduling Mechanism for a Multiprocessor Embedded System

User Adjustable Process Scheduling Mechanism for a Multiprocessor Embedded System Proceedings of he 6h WSEAS Inernaional Conference on Applied Compuer Science, Tenerife, Canary Islands, Spain, December 16-18, 2006 346 User Adjusable Process Scheduling Mechanism for a Muliprocessor Embedded

More information

Location. Electrical. Loads. 2-wire mains-rated. 0.5 mm² to 1.5 mm² Max. length 300 m (with 1.5 mm² cable). Example: Belden 8471

Location. Electrical. Loads. 2-wire mains-rated. 0.5 mm² to 1.5 mm² Max. length 300 m (with 1.5 mm² cable). Example: Belden 8471 Produc Descripion Insallaion and User Guide Transiser Dimmer (454) The DIN rail mouned 454 is a 4channel ransisor dimmer. I can operae in one of wo modes; leading edge or railing edge. All 4 channels operae

More information

A NEW APPROACH FOR 3D MODELS TRANSMISSION

A NEW APPROACH FOR 3D MODELS TRANSMISSION A NEW APPROACH FOR 3D MODELS TRANSMISSION A. Guarnieri a, F. Piroi a, M. Ponin a, A. Veore a a CIRGEO, Inerdep. Research Cener of Carography, Phoogrammery, Remoe Sensing and GIS Universiy of Padova, Agripolis

More information

M(t)/M/1 Queueing System with Sinusoidal Arrival Rate

M(t)/M/1 Queueing System with Sinusoidal Arrival Rate 20 TUTA/IOE/PCU Journal of he Insiue of Engineering, 205, (): 20-27 TUTA/IOE/PCU Prined in Nepal M()/M/ Queueing Sysem wih Sinusoidal Arrival Rae A.P. Pan, R.P. Ghimire 2 Deparmen of Mahemaics, Tri-Chandra

More information

(Structural Time Series Models for Describing Trend in All India Sunflower Yield Using SAS

(Structural Time Series Models for Describing Trend in All India Sunflower Yield Using SAS (Srucural Time Series Models for Describing Trend in All India Sunflower Yield Using SAS Himadri Ghosh, Prajneshu and Savia Wadhwa I.A.S.R.I., Library Avenue, New Delhi-110 01 him_adri@iasri.res.in, prajnesh@iasri.res.in,

More information

4.1 3D GEOMETRIC TRANSFORMATIONS

4.1 3D GEOMETRIC TRANSFORMATIONS MODULE IV MCA - 3 COMPUTER GRAPHICS ADMN 29- Dep. of Compuer Science And Applicaions, SJCET, Palai 94 4. 3D GEOMETRIC TRANSFORMATIONS Mehods for geomeric ransformaions and objec modeling in hree dimensions

More information

Autonomic Cognitive-based Data Dissemination in Opportunistic Networks

Autonomic Cognitive-based Data Dissemination in Opportunistic Networks Auonomic Cogniive-based Daa Disseminaion in Opporunisic Neworks Lorenzo Valerio, Marco Coni, Elena Pagani and Andrea Passarella IIT-CNR, Pisa, Ialy Email: {marco.coni,andrea.passarella,lorenzo.valerio}@ii.cnr.i

More information

FLORIDA INTERNATIONAL UNIVERSITY. Miami, Florida DIMUSE: AN INTEGRATED FRAMEWORK FOR DISTRIBUTED MULTIMEDIA

FLORIDA INTERNATIONAL UNIVERSITY. Miami, Florida DIMUSE: AN INTEGRATED FRAMEWORK FOR DISTRIBUTED MULTIMEDIA FLORIDA INTERNATIONAL UNIVERSITY Miami, Florida DIMUSE: AN INTEGRATED FRAMEWORK FOR DISTRIBUTED MULTIMEDIA SYSTEM WITH DATABASE MANAGEMENT AND SECURITY SUPPORT A disseraion submied in parial fulfillmen

More information

CS 152 Computer Architecture and Engineering. Lecture 7 - Memory Hierarchy-II

CS 152 Computer Architecture and Engineering. Lecture 7 - Memory Hierarchy-II CS 152 Compuer Archiecure and Engineering Lecure 7 - Memory Hierarchy-II Krse Asanovic Elecrical Engineering and Compuer Sciences Universiy of California a Berkeley hp://www.eecs.berkeley.edu/~krse hp://ins.eecs.berkeley.edu/~cs152

More information

Optimal Crane Scheduling

Optimal Crane Scheduling Opimal Crane Scheduling Samid Hoda, John Hooker Laife Genc Kaya, Ben Peerson Carnegie Mellon Universiy Iiro Harjunkoski ABB Corporae Research EWO - 13 November 2007 1/16 Problem Track-mouned cranes move

More information

MATH Differential Equations September 15, 2008 Project 1, Fall 2008 Due: September 24, 2008

MATH Differential Equations September 15, 2008 Project 1, Fall 2008 Due: September 24, 2008 MATH 5 - Differenial Equaions Sepember 15, 8 Projec 1, Fall 8 Due: Sepember 4, 8 Lab 1.3 - Logisics Populaion Models wih Harvesing For his projec we consider lab 1.3 of Differenial Equaions pages 146 o

More information

Vulnerability Evaluation of Multimedia Subsystem Based on Complex Network

Vulnerability Evaluation of Multimedia Subsystem Based on Complex Network JOURAL OF MULTIMDIA, VOL. 8, O. 4, AUGUST 23 439 Vulnerabiliy valuaion of Mulimedia Subsysem Based on Complex ewor Xiaoling Tang Insiue of Higher ducaion Research, Jilin Business and Technology College,

More information

Using CANopen Slave Driver

Using CANopen Slave Driver CAN Bus User Manual Using CANopen Slave Driver V1. Table of Conens 1. SDO Communicaion... 1 2. PDO Communicaion... 1 3. TPDO Reading and RPDO Wriing... 2 4. RPDO Reading... 3 5. CANopen Communicaion Parameer

More information

Motor Control. 5. Control. Motor Control. Motor Control

Motor Control. 5. Control. Motor Control. Motor Control 5. Conrol In his chaper we will do: Feedback Conrol On/Off Conroller PID Conroller Moor Conrol Why use conrol a all? Correc or wrong? Supplying a cerain volage / pulsewidh will make he moor spin a a cerain

More information

Answering Relationship Queries on the Web

Answering Relationship Queries on the Web Answering Relaionship Queries on he Web Gang Luo Chunqiang Tang Ying-li Tian IBM T.J. Wason Research Cener {luog, cang, ylian}@us.ibm.com ABSTRACT Finding relaionships beween eniies on he Web, e.g., he

More information