How To Create An Emoton Recognzer
|
|
- Cecil Boyd
- 3 years ago
- Views:
Transcription
1 Recognzng Low/Hgh Anger n Speech for Call Centers FU-MING LEE*, LI-HUA LI, RU-YI HUANG Department of Informaton Management Chaoyang Unversty of Technology 168 Jfong E. Rd., Wufong Townshp Tachung County, 41349, Tawan (R.O.C.) Abstract: - Automatc mult-level anger recognton n speech s an mportant factor to enhance user satsfacton for call centers. In ths research, three emotonal states, namely, neutral, low anger, and hgh anger of acted corpora wth telephone qualty are specfed for emoton recognton. The corpora are collected from amateur actors and, thereafter, verfed by the actors themselves. The emoton recognzer s mplemented by usng Back-propagaton Network (BPN) wth acoustc features of speech. Due to the varaton of expresson methods by dfferent person, the feature values of the tranng used are too complex to make the BPN model convergent. To overcome the problem, a codfed method s developed to smplfy the feature values. Wth the codfed nputs, the BPN model and a comparatve Decson Tree C5.0 gve qute satsfactory test performances for anger recognton. Therefore, they can be used as a part of a decson support system for proper applcatons n call centers. Key-Words: - Mult-level anger recognton, call center, acoustc features, Back-propagaton Network 1 Introducton Recently, there has been an ncreasng nterest n automatc emoton recognton n speech [9][13][15] [17][18], especally the negatve emoton recognton. One motvaton comes from the desre to develop human machne nterfaces that are more adaptve and responsve to a user s behavor [5][9][10]. The other motvaton comes from the need of a varety of applcatons for busness [11][13][18]. In partcular, n a call center envronment, anger s dentfed as the negatve and the most mportant emoton. Three types of spoken materal, namely, acted, nduced, and real-lfe corpora, are generally collected for emoton analyss and recognton [2][7]. To test the performance of the emoton recognzer, we must know the actual emotonal state of each test example of the corpora. Although, nduced and real-lfe corpora are more natural than acted one, the state of each example of those corpora s unknown unless the speaker tells us the true answer rght after the corpora beng collected. In servce practces, t s almost mpossble to get the answer. Even f the state s labeled by several ndependent labelers, the nter-labeler agreement s very low [5]. Furthermore, even f all labelers acheve agreement, the answer may be wrong. On the contrary, the state of each example of acted corpora can be easly verfed by the speaker. Therefore, n ths research acted corpora are collected for buldng and evaluatng the recognton model. Features of the speech sgnal need to be extracted for automatc emoton recognton. There are large numbers of lngustc (lexcal and dalogc) features and paralngustc (acoustc, fluent, etc.) features that can be attrbuted to the emotonal state of the sgnal [7]. Among them acoustc features are the most wdely employed [9][13][19][20]. In ths research lngustc features are excluded snce the sentences spoken n acted corpora are controlled and expressed wth the whole spectrum of consdered emotons. Furthermore, because there s no agreement on the best set of relevant acoustc features for emoton recognton, our strategy s to use as many features as possble. Many types of emoton recognzer of spoken materal have been used, e.g. Artfcal Neural Networks (ANN) [3][11][13][14][15][20], K-nearest Neghbors [6][9][10][11][13][15][16][20], Decson Trees [7][16], Gaussan models [5][12][15], and Support Vector Machnes [7][11][15][16][18][20]. However, the most effcent model s stll not well establshed and from publshed results appears to be data-dependent [18]. Because the major capablty of ANN s ther flexblty n approxmatng operaton between nputs and outputs and one of the most popular models n ANN s Back-propagaton * Correspondng author (E-mal: fmlee@cyut.edu.tw; TEL: ext. 4286) ISSN:
2 Network (BPN), the recognton model s mplemented by usng BPN n ths research. Furthermore, the performance of BPN s compared wth the performance of Decson Tree C5.0, whch s one of the most wdely used recognzer. The goal of ths research s to create an emoton recognzer that can process telephone qualty voce messages n real tme and can be used as a part of a decson support system for proper applcatons n call centers. Three emotonal states, namely, neutral, low anger, and hgh anger are consdered for emoton recognton. Ths dstncton s motvated by the wsh to recognze low/hgh anger, especally low anger, whch wll be handled by the conclaton strateges. In our opnon mult-level anger recognton s an mportant factor to enhance caller satsfacton. However, recognton results are sparse n the lterature. In the work of Burkhardt et al. [5], prosodc features of real-lfe corpora are extracted; an algorthm based on Gaussan denstes are used to recognze emotons; the accuracy of the recognzer for no anger, low anger, and hgh anger emotons are 89%, 49% and 16%, respectvely, whch shows that the organzer does not dstngush between these emotons very well. In ths research, the acted corpora are collected from 48 amateur actors and, thereafter, verfed by the actors themselves. From the corpora, 1401 are created and are parttoned nto tranng set and test set,.e and 294, respectvely. Each data set has even for three emotonal states. From each example, 47 acoustc features are extracted by usng the Praat program [4] and, then, are normalzed for developng and evaluatng the emoton recognzers. A standard F1 measure s used to assess the performance of the emoton recognzers. Due to the varaton of expresson methods by dfferent people, the BPN model can not converge n tranng stage. To overcome the convergent problem, a codfed method s proposed to smplfy the complexty of the normalzed feature values. Thereafter, the BPN model wth codfed nputs s convergent; F1 measures of the model for test wth neutral, low anger, hgh anger, and anger (.e. low anger or hgh anger) emotonal states are 71.5%, 47.0%, 64.4%, and 86.0%, respectvely. F1 measures of Decson Tree C5.0 wth codfed nputs for test wth neutral, low anger, hgh anger, and anger emotonal states are 66.0%, 41.2%, 64.6%, and 82.4%, respectvely. The results show that the performance of anger emoton detecton s satsfactory. However, the performance of low anger emoton detecton s poor snce low anger s often confused wth neutral or hgh anger. 2 Corpora Collecton In ths research, three emotonal states, namely, neutral, low anger, and hgh anger of acted corpora are specfed for emoton analyss and recognton. The followng s the detals about the corpora and the collectng method. Recordng equpments: Each utterance of the corpora s recorded by usng a close-talk mcrophone wth a recorder equpped n Mcrosoft Wndows. To smulate the voce n telephone, the format of the recorded utterance s set at 8000 Hz, 8 bt, and mono. Amateur actors: There are 25 male actors and 23 female actors partcpatng n ths actvty. They are undergraduate or graduate students wth age between 20 and 30. Gven sentences: Thrty short sentences are collected from the FAQs of webstes of two telecommuncaton servce provders n Tawan. Each actor s asked to select ten of them whch he (or she) supposes that t would be easer for hm (or her) to speak at all three emotonal states consdered n ths research. Recordng envronment: The recordng actvty takes place at a normal offce. In a recordng case, an actor speaks successvely a sentence three tmes wth neutral, low anger, and hgh anger emotonal states, respectvely. Qualty verfcaton: Rght after a recordng case the actor s asked to verfy the qualty of the utterance by playng t back. Unless the actor feels that the states of the utterance he portrays on demand are ndeed consstent wth the states he usually acts n the real lfe, the case wll be recorded agan. Emotonal : Three emotonal of an utterance are cut by a research member of us. After examnng all the utterances, we fnd and dscard 13 utterances wth hgh background noses. Thus, the corpora of 1401 are created wth 467 per emotonal state. 3 Features Extracton In ths research, we ntend to extract the acoustc features, namely, ptch, ntensty, formant, pulse, shmmer, jtter, harmoncs to nose rato (HNR), and duraton, by usng the Praat program [4]. Before features extractng, every emotonal example s fltered by a band-pass flter of the Praat program wth band frequency set between 200 Hz and 3200 Hz whch s close to the band frequency of voce n telephone. In features extractng, wndow length s ISSN:
3 set at 0.02 seconds. The formant frequences are estmated by usng the lnear predcted codng (LPC) method. The predcton order s set as 16. Ptch floor and ptch celng are set accordng to the gender of the actor. For male s emotonal example, the 2 parameters are set at 75 Hz and 300 Hz, respectvely. For female s emotonal example, the 2 parameters are set at 100 Hz and 500 Hz, respectvely. The rest parameters n the Praat program are set as default values. The followng s the detals about the 47 features extracted from each emotonal example. Ptch features: A ptch object represents perodcty canddates as a functon of tme,.e. a ptch contour. Related ptch features are statstcal propertes of the ptch contour. Sx ptch features are used n ths research: maxmum, mnmum, mean, standard devaton, mean absolute slope, mean absolute slope wthout octave jumps. Intensty features: An ntensty object represents an ntensty contour at certan lnearly spaced tme ponts. Related ntensty features are statstcal propertes of the ntensty contour. Four ntensty features are used n ths research: maxmum, mnmum, mean, standard devaton. Formant features: A formant object represents spectral structure as a functon of tme,.e. a formant contour. Related formant features are statstcal propertes of the formant contour. Twenty formant features are used n ths research. The followng s the detals about the features. Frst formant (F1): maxmum, mnmum, mean, standard devaton. Second formant (F2): maxmum, mnmum, mean, standard devaton. Thrd formant (F3): maxmum, mnmum, mean, standard devaton. Fourth formant (F4): maxmum, mnmum, mean, standard devaton. Bandwdths: B1, B2, B3, B4. Pulse features: It s generated at every pont n the pont process. Two related pulse features are used n ths research: mean perod, standard devaton perod. Jtter features: Fve jtter features are used n ths research: jtter (local), jtter (local, absolute), jtter (rap), jtter (ppq5), jtter (ddp). The defntons of these features please referred to the Praat program [4]. Shmmer feature: Fve shmmer features are used n ths research: shmmer (local), shmmer (apq3), shmmer (apq5), shmmer (apq11), shmmer (ddp). The defntons of these features please referred to the Praat program [4]. HNR features: It represents the degree of acoustc perodcty. Four related HNR features are used n ths research: maxmum, mnmum, mean, standard devaton. Duraton feature: It represents the rato of voced frame s number and unvoced frame s number. 4 Features Preprocessng To tran the emoton recognzer and test ts performance, the 1401 emotonal are parttoned nto tranng set and test set. The tranng set conssts of 1107 emotonal created by randomly chosen 20 male actors and 18 female actors; the test set conssts of 294 emotonal created by the remanng 5 male actors and 5 female actors. Because the unts of the features are dfferent and the ranges of the features values of the emotonal are dverse, all of the 47 features are normalzed as follows. For all e tranng set test set, let f e f ( e) f,mn ( ) =, f,max f,mn 1, f f ( e) > 1, f ( e) = f ( e), f 0 f ( e) 1, 0, otherwse, (1) where f (e) denotes the th feature value of emotonal example e, f( e ) denotes the normalzed value of f (e), f,max denotes the 2nd maxmum value of {f (e): e tranng set}, f,mn denotes the 2nd mnmum value of {f (e): e tranng set}. To avod to get unusual extreme feature values, both f,max and f,mn are determned as the 2nd extreme values mentoned above. Furthermore, to conform to practces, we suppose that the nformaton about the test set s unknown n advance. Therefore, both f,max and f,mn are determned only based on emotonal of tranng set. 5 Performance Measure In ths research, F1 measure s used to assess the performance of recognzers. It has been used very commonly as a standard performance measure n the recognton problem, and s defned as follows. 2PR nc nc F1 =, R =, P = P + R m m (2) c c ISSN:
4 where n c denotes the number of test whose emotonal state s correctly state c, m c denotes the number of test wth emotonal state c, and m' c denotes the number of test whose emotonal state s emoton state c. 6 BPN wth Normalzed Inputs The BPN model proposed n ths research s depcted on Fg. 1. It conssts of three knds of layers,.e. nput layer, hdden layer, and output layer. The nputs conssts of 47 normalzed features. The number of hdden nodes s set as 25. The number of output nodes s set as the number of emotonal states. In tranng stage, the target values of the tranng example for the outputs are set accordng to the emotonal state of the example descrbed n Table 1. In test stage, the recognzed emotonal state of the test example s determned accordng to the output wth hghest value. For example, f y l (e) has the hghest value, the emotonal state of the test example s low anger. f1( e) f2( e) f ( e) 47 Fg. 1 The BPN model In the experments, all the 1107 tranng are used to tran the BPN model by usng the Matlab s Neural Network Toolbox. However, the BPN model can not converge. To fnd out the reason, the tranng set and the test set are regrouped by dscardng wth certan emotonal state; the 47 features are preprocessed as (1) wth regrouped tranng set and test set; the BPN model s reformed by dscardng the output node correspondng to the certan emotonal state; the target values of the tranng example for the remanng 2 outputs are set as shown n Table 1, and the reformed BPN model s traned by usng the regrouped tranng. The reformed BPN model can not converge ether n the case where hgh anger emotonal are dscarded or n the case where neutral emotonal are dscarded. Nevertheless, n the case where low anger emotonal are dscarded, the reformed BPN model s convergent and ts performance s shown n Table 2. The F1 measures for test wth neutral or hgh anger emotonal states are above 85%. Table 1 Target value of tranng example for outputs output y emotonal state n (e) y l (e) y h (e) neutral low anger hgh anger Table 2 Performance of BPN wthout node y l (e) hgh measures (%) neutral anger P R F1 neutral hgh anger In collectng the acted corpora, we observe that the ways to speak a gven sentence wth neutral, low anger, and hgh anger emoton are qute dfferent for dfferent person. Thus, for each feature, the feature values of the three emotonal and the patterns of the values dstrbuton are qute dfferent for varous persons. Furthermore, for the three emotonal states, some actors expresson methods are hard to dscrmnate, especally the methods between neutral and low anger or between low anger and hgh anger. Ths s the reason why the BPN model and the reformed BPN models wth node y l (e) can not converge, and the reformed BPN model wthout node y l (e) are convergent and has excellent test performance. Because the qualty of the s verfed by the actor of the rght after a recordng case, the low anger are stll vald and should be consdered. To overcome the convergent problem of the BPN model, the complexty of the normalzed feature values of the emotonal s smplfed by codfyng method descrbed n the next secton. 7 Features Codfcaton To codfy th normalzed feature value of any emotonal example, th normalzed feature values of the 1107 tranng are collected. They are ordered by ther values and are parttoned nto fve segments depcted on Fg. 2, n whch four partton ponts are determned as (3). Fg. 2 Partton ponts for the th normalzed feature f, s, lp + f, s+ 1, fp p, s =, for = 1, 2,..., 47, s = 1, 2, 3, 4, and (3) 2 for the th normalzed feature p,s denotes the sth partton pont, f, s, lp denotes the value of the last pont at segment s, f, s+ 1, fp denotes the value of the frst pont at segment s+1. ISSN:
5 Every normalzed feature value of an emotonal example s codfed nto four dgts defned n Table 3. For nstance, when th normalzed feature value of an emotonal example belongs n certan segment depcted on Fg. 2, the value s codfed nto four bnary dgts correspondng to the segment. In Table 3, the bnary codes are desgned such that the hammng dstance of any two codes represents the segment dstance of the correspondng two segments. For example, the hammng dstance of two codes correspondng to segment 1 and segment 5 s 4. The detals are descrbed as n Table 4. Table 3 The bnary code for each segment segment seg. 1 seg. 2 seg. 3 seg. 4 seg. 5 bnary code Table 4 The hammng dstance of any two codes segment seg. 1 seg. 2 seg. 3 seg. 4 seg. 5 seg seg seg seg seg BPN wth Codfed Inputs To accommodate the codfed nputs, the BPN model s redesgned wth 188 bnary nput nodes and 95 hdden nodes. The number of outputs and the target value of the tranng example are desgned by the same way for the BPN model mentoned n Secton 6. The BPN model wth codfed nputs s traned by the 1107 tranng. Fnally, the BPN model wth codfed nputs s converged and the performance of the model s shown n Table 5. The F1 measures for test wth neutral, low anger, and hgh anger emotonal states are 71.5%, 47.0%, and 64.4%, respectvely. Table 5 Performance of BPN wth codfed nputs neutral low anger hgh measures (%) anger P R F1 neutral low anger hgh anger Table 6 Performance of BPN wth codfed nputs for neutral and anger emotons measures (%) neutral anger P R F1 neutral anger As mentoned n Secton 6, low anger emotonal are often confused wth neutral or hgh anger emotonal. Thus, although the convergent problem of the BPN model can be overcome by codfyng the complex normalzed features, the performance of the BPN model wth codfed nputs are stll not good for the three emotonal states, especally low anger. However, wth respect to applcatons n call center, the performance for anger,.e. low anger or hgh anger, emoton recognton s the man concern. Therefore, after combnng the results of both low anger and hgh anger, the performance of the BPN model wth codfed nputs s recalculated as Table 6. The F1 measure for test wth anger emotonal state s 86.0%. To compare the performance of the BPN model wth codfed nputs, Decson Tree C5.0 wth codfed nputs s developed and evaluated wth the same tranng set and the test set by usng SPSS Clementne. The performance of Decson Tree C5.0 s shown n Table 7 and Table 8. The F1 measures for test wth neutral, low anger, hgh anger, and anger emotonal states are 66.0%, 41.2%, 64.6%, and 82.4%, respectvely. The results of the two models show that the performance of anger emoton detecton s satsfactory. However, the performance of low anger emoton detecton s poor snce low anger s often confused wth neutral or hgh anger. From the experments, we notce that the BPN model s superor to the Decson Tree C5.0. Table 7 Performance of Decson Tree C5.0 wth codfed nputs neutral low anger hgh measures (%) anger P R F1 neutral low anger hgh anger Table 8 Performance of Decson Tree C5.0 wth codfed nputs for neutral and anger emotons measures (%) neutral anger P R F1 neutral anger Conclusons Automatc mult-level anger recognton n speech s an mportant factor to enhance user satsfacton for call centers. To recognze the level of anger, one dffculty comes from the lack of clear defnton or descrpton to dstnct the levels of anger and the other dffculty comes from the need of actual emotonal state nformaton of the telephone speech. In our opnon, the caller hmself s the only person who exactly knows the answer and the emotonal state of the caller s the man concern of call centers. In ths research, neutral, low anger, and hgh anger emotons are consdered. To have the exact emotonal state nformaton of speech, the acted corpora are verfed by the actor hmself. Due to the varaton of expresson methods by dfferent people, the feature values of the tranng ISSN:
6 used are too complex to make the BPN model convergent. To overcome the problem, a codfed method s developed to smplfy the feature values. Wth the codfed nputs, the results of the BPN model and a comparatve Decson Tree C5.0 show that the performance of anger (.e. low anger or hgh anger) emoton detecton s satsfactory. Therefore, they can be used as a part of a decson support system for proper applcatons n call centers. References: [1] J. Adell, A. Bonafonte, and D. Escudero, Analyss of Prosodc Features: Towards Modelng of Emotonal and Pragmatc Attrbutes of Speech, XXI Cogreso de la Socedad Española para el Procesamento del Lenguaje Natural, [2] A. Batlner, K. Fscher, R. Huber, J. Splker, and E. Nöth, Desperately Seekng Emotons or: Actors, Wzards, and Human Bengs, Proc. of The Inter. Speech Communcaton Assocaton Workshop on Speech and Emoton, 2000, pp [3] M. W. Bhatt, Y. Wang, and L. Guan, A Neural Network Approach for Human Emoton Recognton n Speech, Proc. of IEEE Inter. Symposum on Crcuts and Systems, Vol. 2, 2004, pp [4] P. Boersma and D. Weennk, Praat: Dong Phonetcs by Computer (Verson ), [5] F. Burkhardt, M. V. Ballegooy, R. Englert, and R. Huber, An Emoton-Aware Voce Portal, Proc. of The 16th Conf. on Electronc Speech Sgnal Processng, 2005, pp [6] F. Dellaert, T. Polzn, and A. Wabel, Recognzng Emoton n Speech, Proc. of The Inter. Conf. on Spoken Language Processng, 1996, pp [7] L. Devllers, L. Vdrascu, and L. Lamel, Challenges n Real-Lfe Emoton Annotaton and Machne Learnng Based Detecton, Neural Networks, Vol. 18, No. 4, 2005, pp [8] O.-W. Kwon, K. Chan, J. Hao, and T.-W. Lee, Emoton Recognton by Speech Sgnals, Proc. of The 8th European Conf. on Speech Communcaton and Technology, 2003, pp [9] C. M. Lee, S. S. Narayanan, and R. Peraccn, Recognton of Negatve Emotons from the Speech Sgnal, Proc. of Automatc Speech Recognton and Understandng, 2001, pp [10] C. M. Lee and S. S. Narayanan, Toward Detectng Emotons n Spoken Dalogs, IEEE Trans. on Speech and Audo Processng, Vol. 13, No. 2, 2005, pp [11] D. Morrson, R. Wang, and L. C. De Slva, Ensemble Methods for Spoken Emoton Recognton n Call-Centers, Speech Communcaton, Vol. 49, 2007, pp [12] D. Neberg, K. Elenus, and K. Laskowsk, Emoton Recognton n Spontaneous Speech Usng GMMs, Proc. of The 19th Inter. Conf. on Spoken Language Processng, 2006, pp [13] V. A. Petrushn, Emoton Recognton n Speech Sgnal: Expermental Study, Development, and Applcaton, Proc. of The 6th Inter. Conf. on Spoken Language Processng, 2000, pp [14] H. Sato, Y. Mtsukura, M. Fukum, and N. Akamatsu, Emotonal Speech Classfcaton wth Prosodc Parameters by Usng Neural Networks, Proc. of The 7th Australan and New Zealand Intellgent Informaton Systems Conf., 2001, pp [15] B. Schuller, G. Rgoll, and M. Lang, Speech Emoton Recognton Combnng Acoustc Features and Lngustc Informaton n a Hybrd Support Vector Machne - Belef Network Archtecture, Proc. of Internat Conf. on Acoustcs, Speech, and Sgnal Processng, Vol. 1, 2004, pp [16] M. Sham and W. Verhelst, An Evaluaton of the Robustness of Exstng Supervsed Machne Learnng Approaches to The Classfcaton of Emotons n Speech, Speech Communcaton, Vol. 49, 2007, pp [17] D. Ververds and C. Kotropoulos, Emotonal Speech Recognton: Resources, Features, and Methods, Speech Communcaton, Vol. 48, 2006, pp [18] L. Vdrascu and L. Devllers, Detecton of Real-Lfe Emotons n Call Centers, Proc. of The 18th Inter. Conf. on Spoken Language Processng, 2005, pp [19] S. Yldrm, M. Bulut, C. M. Lee, A. Kazemzadeh, C. Busso, Z. Deng, S. Lee, and S. Narayanan, An Acoustc Study of Emotons Expressed n Speech, Proc. of Internat Conf. on Spoken Language Processng, Vol. 1, 2004, pp [20] F. Yu, E. Chang, Y.-Q. Xu, and H.-Y. Shum, Emoton Detecton from Speech to Enrch Multmeda Content, Proc. of The 2nd IEEE Pacfc Rm Conf. on Multmeda, 2001, pp ISSN:
The Development of Web Log Mining Based on Improve-K-Means Clustering Analysis
The Development of Web Log Mnng Based on Improve-K-Means Clusterng Analyss TngZhong Wang * College of Informaton Technology, Luoyang Normal Unversty, Luoyang, 471022, Chna wangtngzhong2@sna.cn Abstract.
More informationWhat is Candidate Sampling
What s Canddate Samplng Say we have a multclass or mult label problem where each tranng example ( x, T ) conssts of a context x a small (mult)set of target classes T out of a large unverse L of possble
More informationRecurrence. 1 Definitions and main statements
Recurrence 1 Defntons and man statements Let X n, n = 0, 1, 2,... be a MC wth the state space S = (1, 2,...), transton probabltes p j = P {X n+1 = j X n = }, and the transton matrx P = (p j ),j S def.
More informationVision Mouse. Saurabh Sarkar a* University of Cincinnati, Cincinnati, USA ABSTRACT 1. INTRODUCTION
Vson Mouse Saurabh Sarkar a* a Unversty of Cncnnat, Cncnnat, USA ABSTRACT The report dscusses a vson based approach towards trackng of eyes and fngers. The report descrbes the process of locatng the possble
More informationFeature selection for intrusion detection. Slobodan Petrović NISlab, Gjøvik University College
Feature selecton for ntruson detecton Slobodan Petrovć NISlab, Gjøvk Unversty College Contents The feature selecton problem Intruson detecton Traffc features relevant for IDS The CFS measure The mrmr measure
More informationForecasting the Direction and Strength of Stock Market Movement
Forecastng the Drecton and Strength of Stock Market Movement Jngwe Chen Mng Chen Nan Ye cjngwe@stanford.edu mchen5@stanford.edu nanye@stanford.edu Abstract - Stock market s one of the most complcated systems
More informationA Secure Password-Authenticated Key Agreement Using Smart Cards
A Secure Password-Authentcated Key Agreement Usng Smart Cards Ka Chan 1, Wen-Chung Kuo 2 and Jn-Chou Cheng 3 1 Department of Computer and Informaton Scence, R.O.C. Mltary Academy, Kaohsung 83059, Tawan,
More informationAn Interest-Oriented Network Evolution Mechanism for Online Communities
An Interest-Orented Network Evoluton Mechansm for Onlne Communtes Cahong Sun and Xaopng Yang School of Informaton, Renmn Unversty of Chna, Bejng 100872, P.R. Chna {chsun,yang}@ruc.edu.cn Abstract. Onlne
More informationOn-Line Fault Detection in Wind Turbine Transmission System using Adaptive Filter and Robust Statistical Features
On-Lne Fault Detecton n Wnd Turbne Transmsson System usng Adaptve Flter and Robust Statstcal Features Ruoyu L Remote Dagnostcs Center SKF USA Inc. 3443 N. Sam Houston Pkwy., Houston TX 77086 Emal: ruoyu.l@skf.com
More informationFace Verification Problem. Face Recognition Problem. Application: Access Control. Biometric Authentication. Face Verification (1:1 matching)
Face Recognton Problem Face Verfcaton Problem Face Verfcaton (1:1 matchng) Querymage face query Face Recognton (1:N matchng) database Applcaton: Access Control www.vsage.com www.vsoncs.com Bometrc Authentcaton
More informationForecasting the Demand of Emergency Supplies: Based on the CBR Theory and BP Neural Network
700 Proceedngs of the 8th Internatonal Conference on Innovaton & Management Forecastng the Demand of Emergency Supples: Based on the CBR Theory and BP Neural Network Fu Deqang, Lu Yun, L Changbng School
More informationA Hierarchical Anomaly Network Intrusion Detection System using Neural Network Classification
IDC IDC A Herarchcal Anomaly Network Intruson Detecton System usng Neural Network Classfcaton ZHENG ZHANG, JUN LI, C. N. MANIKOPOULOS, JAY JORGENSON and JOSE UCLES ECE Department, New Jersey Inst. of Tech.,
More informationResearch on Evaluation of Customer Experience of B2C Ecommerce Logistics Enterprises
3rd Internatonal Conference on Educaton, Management, Arts, Economcs and Socal Scence (ICEMAESS 2015) Research on Evaluaton of Customer Experence of B2C Ecommerce Logstcs Enterprses Yle Pe1, a, Wanxn Xue1,
More informationA Novel Methodology of Working Capital Management for Large. Public Constructions by Using Fuzzy S-curve Regression
Novel Methodology of Workng Captal Management for Large Publc Constructons by Usng Fuzzy S-curve Regresson Cheng-Wu Chen, Morrs H. L. Wang and Tng-Ya Hseh Department of Cvl Engneerng, Natonal Central Unversty,
More informationSPEE Recommended Evaluation Practice #6 Definition of Decline Curve Parameters Background:
SPEE Recommended Evaluaton Practce #6 efnton of eclne Curve Parameters Background: The producton hstores of ol and gas wells can be analyzed to estmate reserves and future ol and gas producton rates and
More informationLuby s Alg. for Maximal Independent Sets using Pairwise Independence
Lecture Notes for Randomzed Algorthms Luby s Alg. for Maxmal Independent Sets usng Parwse Independence Last Updated by Erc Vgoda on February, 006 8. Maxmal Independent Sets For a graph G = (V, E), an ndependent
More informationImproved SVM in Cloud Computing Information Mining
Internatonal Journal of Grd Dstrbuton Computng Vol.8, No.1 (015), pp.33-40 http://dx.do.org/10.1457/jgdc.015.8.1.04 Improved n Cloud Computng Informaton Mnng Lvshuhong (ZhengDe polytechnc college JangSu
More informationGender Classification for Real-Time Audience Analysis System
Gender Classfcaton for Real-Tme Audence Analyss System Vladmr Khryashchev, Lev Shmaglt, Andrey Shemyakov, Anton Lebedev Yaroslavl State Unversty Yaroslavl, Russa vhr@yandex.ru, shmaglt_lev@yahoo.com, andrey.shemakov@gmal.com,
More informationSTANDING WAVE TUBE TECHNIQUES FOR MEASURING THE NORMAL INCIDENCE ABSORPTION COEFFICIENT: COMPARISON OF DIFFERENT EXPERIMENTAL SETUPS.
STADIG WAVE TUBE TECHIQUES FOR MEASURIG THE ORMAL ICIDECE ABSORPTIO COEFFICIET: COMPARISO OF DIFFERET EXPERIMETAL SETUPS. Angelo Farna (*), Patrzo Faust (**) (*) Dpart. d Ing. Industrale, Unverstà d Parma,
More informationbenefit is 2, paid if the policyholder dies within the year, and probability of death within the year is ).
REVIEW OF RISK MANAGEMENT CONCEPTS LOSS DISTRIBUTIONS AND INSURANCE Loss and nsurance: When someone s subject to the rsk of ncurrng a fnancal loss, the loss s generally modeled usng a random varable or
More informationHow To Understand The Results Of The German Meris Cloud And Water Vapour Product
Ttel: Project: Doc. No.: MERIS level 3 cloud and water vapour products MAPP MAPP-ATBD-ClWVL3 Issue: 1 Revson: 0 Date: 9.12.1998 Functon Name Organsaton Sgnature Date Author: Bennartz FUB Preusker FUB Schüller
More informationCan Auto Liability Insurance Purchases Signal Risk Attitude?
Internatonal Journal of Busness and Economcs, 2011, Vol. 10, No. 2, 159-164 Can Auto Lablty Insurance Purchases Sgnal Rsk Atttude? Chu-Shu L Department of Internatonal Busness, Asa Unversty, Tawan Sheng-Chang
More informationAPPLICATION OF PROBE DATA COLLECTED VIA INFRARED BEACONS TO TRAFFIC MANEGEMENT
APPLICATION OF PROBE DATA COLLECTED VIA INFRARED BEACONS TO TRAFFIC MANEGEMENT Toshhko Oda (1), Kochro Iwaoka (2) (1), (2) Infrastructure Systems Busness Unt, Panasonc System Networks Co., Ltd. Saedo-cho
More informationTHE APPLICATION OF DATA MINING TECHNIQUES AND MULTIPLE CLASSIFIERS TO MARKETING DECISION
Internatonal Journal of Electronc Busness Management, Vol. 3, No. 4, pp. 30-30 (2005) 30 THE APPLICATION OF DATA MINING TECHNIQUES AND MULTIPLE CLASSIFIERS TO MARKETING DECISION Yu-Mn Chang *, Yu-Cheh
More informationBrigid Mullany, Ph.D University of North Carolina, Charlotte
Evaluaton And Comparson Of The Dfferent Standards Used To Defne The Postonal Accuracy And Repeatablty Of Numercally Controlled Machnng Center Axes Brgd Mullany, Ph.D Unversty of North Carolna, Charlotte
More informationPerformance Analysis and Comparison of QoS Provisioning Mechanisms for CBR Traffic in Noisy IEEE 802.11e WLANs Environments
Tamkang Journal of Scence and Engneerng, Vol. 12, No. 2, pp. 143149 (2008) 143 Performance Analyss and Comparson of QoS Provsonng Mechansms for CBR Traffc n Nosy IEEE 802.11e WLANs Envronments Der-Junn
More informationVoIP Playout Buffer Adjustment using Adaptive Estimation of Network Delays
VoIP Playout Buffer Adjustment usng Adaptve Estmaton of Network Delays Mroslaw Narbutt and Lam Murphy* Department of Computer Scence Unversty College Dubln, Belfeld, Dubln, IRELAND Abstract The poor qualty
More information8 Algorithm for Binary Searching in Trees
8 Algorthm for Bnary Searchng n Trees In ths secton we present our algorthm for bnary searchng n trees. A crucal observaton employed by the algorthm s that ths problem can be effcently solved when the
More information1 Example 1: Axis-aligned rectangles
COS 511: Theoretcal Machne Learnng Lecturer: Rob Schapre Lecture # 6 Scrbe: Aaron Schld February 21, 2013 Last class, we dscussed an analogue for Occam s Razor for nfnte hypothess spaces that, n conjuncton
More informationLecture 2: Single Layer Perceptrons Kevin Swingler
Lecture 2: Sngle Layer Perceptrons Kevn Sngler kms@cs.str.ac.uk Recap: McCulloch-Ptts Neuron Ths vastly smplfed model of real neurons s also knon as a Threshold Logc Unt: W 2 A Y 3 n W n. A set of synapses
More informationSearching for Interacting Features for Spam Filtering
Searchng for Interactng Features for Spam Flterng Chuanlang Chen 1, Yun-Chao Gong 2, Rongfang Be 1,, and X. Z. Gao 3 1 Department of Computer Scence, Bejng Normal Unversty, Bejng 100875, Chna 2 Software
More informationVehicle Detection and Tracking in Video from Moving Airborne Platform
Journal of Computatonal Informaton Systems 10: 12 (2014) 4965 4972 Avalable at http://www.jofcs.com Vehcle Detecton and Trackng n Vdeo from Movng Arborne Platform Lye ZHANG 1,2,, Hua WANG 3, L LI 2 1 School
More informationUsing Series to Analyze Financial Situations: Present Value
2.8 Usng Seres to Analyze Fnancal Stuatons: Present Value In the prevous secton, you learned how to calculate the amount, or future value, of an ordnary smple annuty. The amount s the sum of the accumulated
More informationAn Alternative Way to Measure Private Equity Performance
An Alternatve Way to Measure Prvate Equty Performance Peter Todd Parlux Investment Technology LLC Summary Internal Rate of Return (IRR) s probably the most common way to measure the performance of prvate
More informationINVESTIGATION OF VEHICULAR USERS FAIRNESS IN CDMA-HDR NETWORKS
21 22 September 2007, BULGARIA 119 Proceedngs of the Internatonal Conference on Informaton Technologes (InfoTech-2007) 21 st 22 nd September 2007, Bulgara vol. 2 INVESTIGATION OF VEHICULAR USERS FAIRNESS
More informationStatistical Approach for Offline Handwritten Signature Verification
Journal of Computer Scence 4 (3): 181-185, 2008 ISSN 1549-3636 2008 Scence Publcatons Statstcal Approach for Offlne Handwrtten Sgnature Verfcaton 2 Debnath Bhattacharyya, 1 Samr Kumar Bandyopadhyay, 2
More informationEye Center Localization on a Facial Image Based on Multi-Block Local Binary Patterns
Eye Center Localzaton on a Facal Image Based on Mult-Bloc Local Bnary Patterns Anatoly tn, Vladmr Khryashchev, Olga Stepanova Yaroslavl State Unversty Yaroslavl, Russa anatolyntnyar@gmal.com, vhr@yandex.ru,
More informationProject Networks With Mixed-Time Constraints
Project Networs Wth Mxed-Tme Constrants L Caccetta and B Wattananon Western Australan Centre of Excellence n Industral Optmsaton (WACEIO) Curtn Unversty of Technology GPO Box U1987 Perth Western Australa
More informationModule 2 LOSSLESS IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur
Module LOSSLESS IMAGE COMPRESSION SYSTEMS Lesson 3 Lossless Compresson: Huffman Codng Instructonal Objectves At the end of ths lesson, the students should be able to:. Defne and measure source entropy..
More information"Research Note" APPLICATION OF CHARGE SIMULATION METHOD TO ELECTRIC FIELD CALCULATION IN THE POWER CABLES *
Iranan Journal of Scence & Technology, Transacton B, Engneerng, ol. 30, No. B6, 789-794 rnted n The Islamc Republc of Iran, 006 Shraz Unversty "Research Note" ALICATION OF CHARGE SIMULATION METHOD TO ELECTRIC
More informationSingle and multiple stage classifiers implementing logistic discrimination
Sngle and multple stage classfers mplementng logstc dscrmnaton Hélo Radke Bttencourt 1 Dens Alter de Olvera Moraes 2 Vctor Haertel 2 1 Pontfíca Unversdade Católca do Ro Grande do Sul - PUCRS Av. Ipranga,
More informationOpen Access A Load Balancing Strategy with Bandwidth Constraint in Cloud Computing. Jing Deng 1,*, Ping Guo 2, Qi Li 3, Haizhu Chen 1
Send Orders for Reprnts to reprnts@benthamscence.ae The Open Cybernetcs & Systemcs Journal, 2014, 8, 115-121 115 Open Access A Load Balancng Strategy wth Bandwdth Constrant n Cloud Computng Jng Deng 1,*,
More informationVRT012 User s guide V0.1. Address: Žirmūnų g. 27, Vilnius LT-09105, Phone: (370-5) 2127472, Fax: (370-5) 276 1380, Email: info@teltonika.
VRT012 User s gude V0.1 Thank you for purchasng our product. We hope ths user-frendly devce wll be helpful n realsng your deas and brngng comfort to your lfe. Please take few mnutes to read ths manual
More informationAn Enhanced Super-Resolution System with Improved Image Registration, Automatic Image Selection, and Image Enhancement
An Enhanced Super-Resoluton System wth Improved Image Regstraton, Automatc Image Selecton, and Image Enhancement Yu-Chuan Kuo ( ), Chen-Yu Chen ( ), and Chou-Shann Fuh ( ) Department of Computer Scence
More informationDEFINING %COMPLETE IN MICROSOFT PROJECT
CelersSystems DEFINING %COMPLETE IN MICROSOFT PROJECT PREPARED BY James E Aksel, PMP, PMI-SP, MVP For Addtonal Informaton about Earned Value Management Systems and reportng, please contact: CelersSystems,
More informationAn Evaluation of the Extended Logistic, Simple Logistic, and Gompertz Models for Forecasting Short Lifecycle Products and Services
An Evaluaton of the Extended Logstc, Smple Logstc, and Gompertz Models for Forecastng Short Lfecycle Products and Servces Charles V. Trappey a,1, Hsn-yng Wu b a Professor (Management Scence), Natonal Chao
More informationA COLLABORATIVE TRADING MODEL BY SUPPORT VECTOR REGRESSION AND TS FUZZY RULE FOR DAILY STOCK TURNING POINTS DETECTION
A COLLABORATIVE TRADING MODEL BY SUPPORT VECTOR REGRESSION AND TS FUZZY RULE FOR DAILY STOCK TURNING POINTS DETECTION JHENG-LONG WU, PEI-CHANN CHANG, KAI-TING CHANG Department of Informaton Management,
More informationDevelopment of an intelligent system for tool wear monitoring applying neural networks
of Achevements n Materals and Manufacturng Engneerng VOLUME 14 ISSUE 1-2 January-February 2006 Development of an ntellgent system for tool wear montorng applyng neural networks A. Antć a, J. Hodolč a,
More informationSupport Vector Machines
Support Vector Machnes Max Wellng Department of Computer Scence Unversty of Toronto 10 Kng s College Road Toronto, M5S 3G5 Canada wellng@cs.toronto.edu Abstract Ths s a note to explan support vector machnes.
More informationFrequency Selective IQ Phase and IQ Amplitude Imbalance Adjustments for OFDM Direct Conversion Transmitters
Frequency Selectve IQ Phase and IQ Ampltude Imbalance Adjustments for OFDM Drect Converson ransmtters Edmund Coersmeer, Ernst Zelnsk Noka, Meesmannstrasse 103, 44807 Bochum, Germany edmund.coersmeer@noka.com,
More informationPerformance Management and Evaluation Research to University Students
631 A publcaton of CHEMICAL ENGINEERING TRANSACTIONS VOL. 46, 2015 Guest Edtors: Peyu Ren, Yancang L, Hupng Song Copyrght 2015, AIDIC Servz S.r.l., ISBN 978-88-95608-37-2; ISSN 2283-9216 The Italan Assocaton
More informationA New Quality of Service Metric for Hard/Soft Real-Time Applications
A New Qualty of Servce Metrc for Hard/Soft Real-Tme Applcatons Shaoxong Hua and Gang Qu Electrcal and Computer Engneerng Department and Insttute of Advanced Computer Study Unversty of Maryland, College
More informationIntelligent Voice-Based Door Access Control System Using Adaptive-Network-based Fuzzy Inference Systems (ANFIS) for Building Security
Journal of Computer Scence 3 (5): 274-280, 2007 ISSN 1549-3636 2007 Scence Publcatons Intellgent Voce-Based Door Access Control System Usng Adaptve-Network-based Fuzzy Inference Systems (ANFIS) for Buldng
More informationSelecting Test Signals for Successful Impairment Classification in VoIP Systems
16 th IMEKO TC4 Symposum Sept. 22-24, 28, Florence, Italy Selectng Test Sgnals for Successful Imparment Classfcaton n VoIP Systems Dors Bao 1,2, Luca De Vto 1, Sergo Rapuano 1 1 Dept. of Engneerng, Unversty
More informationAutomated Network Performance Management and Monitoring via One-class Support Vector Machine
Automated Network Performance Management and Montorng va One-class Support Vector Machne R. Zhang, J. Jang, and S. Zhang Dgtal Meda & Systems Research Insttute, Unversty of Bradford, UK Abstract: In ths
More informationA Dynamic Load Balancing for Massive Multiplayer Online Game Server
A Dynamc Load Balancng for Massve Multplayer Onlne Game Server Jungyoul Lm, Jaeyong Chung, Jnryong Km and Kwanghyun Shm Dgtal Content Research Dvson Electroncs and Telecommuncatons Research Insttute Daejeon,
More informationThe OC Curve of Attribute Acceptance Plans
The OC Curve of Attrbute Acceptance Plans The Operatng Characterstc (OC) curve descrbes the probablty of acceptng a lot as a functon of the lot s qualty. Fgure 1 shows a typcal OC Curve. 10 8 6 4 1 3 4
More informationCHOLESTEROL REFERENCE METHOD LABORATORY NETWORK. Sample Stability Protocol
CHOLESTEROL REFERENCE METHOD LABORATORY NETWORK Sample Stablty Protocol Background The Cholesterol Reference Method Laboratory Network (CRMLN) developed certfcaton protocols for total cholesterol, HDL
More informationInvoicing and Financial Forecasting of Time and Amount of Corresponding Cash Inflow
Dragan Smć Svetlana Smć Vasa Svrčevć Invocng and Fnancal Forecastng of Tme and Amount of Correspondng Cash Inflow Artcle Info:, Vol. 6 (2011), No. 3, pp. 014-021 Receved 13 Janyary 2011 Accepted 20 Aprl
More informationEfficient Project Portfolio as a tool for Enterprise Risk Management
Effcent Proect Portfolo as a tool for Enterprse Rsk Management Valentn O. Nkonov Ural State Techncal Unversty Growth Traectory Consultng Company January 5, 27 Effcent Proect Portfolo as a tool for Enterprse
More informationNEURO-FUZZY INFERENCE SYSTEM FOR E-COMMERCE WEBSITE EVALUATION
NEURO-FUZZY INFERENE SYSTEM FOR E-OMMERE WEBSITE EVALUATION Huan Lu, School of Software, Harbn Unversty of Scence and Technology, Harbn, hna Faculty of Appled Mathematcs and omputer Scence, Belarusan State
More informationA DYNAMIC CRASHING METHOD FOR PROJECT MANAGEMENT USING SIMULATION-BASED OPTIMIZATION. Michael E. Kuhl Radhamés A. Tolentino-Peña
Proceedngs of the 2008 Wnter Smulaton Conference S. J. Mason, R. R. Hll, L. Mönch, O. Rose, T. Jefferson, J. W. Fowler eds. A DYNAMIC CRASHING METHOD FOR PROJECT MANAGEMENT USING SIMULATION-BASED OPTIMIZATION
More informationThe Load Balancing of Database Allocation in the Cloud
, March 3-5, 23, Hong Kong The Load Balancng of Database Allocaton n the Cloud Yu-lung Lo and Mn-Shan La Abstract Each database host n the cloud platform often has to servce more than one database applcaton
More informationTime Delayed Independent Component Analysis for Data Quality Monitoring
IWSSIP 1-17th Internatonal Conference on Systems, Sgnals and Image Processng Tme Delayed Independent Component Analyss for Data Qualty Montorng José Márco Faer Sgnal Processng Laboratory, COE/Pol Federal
More informationDetecting Credit Card Fraud using Periodic Features
Detectng Credt Card Fraud usng Perodc Features Alejandro Correa Bahnsen, Djamla Aouada, Aleksandar Stojanovc and Björn Ottersten Interdscplnary Centre for Securty, Relablty and Trust Unversty of Luxembourg,
More informationAbstract. 1. Introduction
System and Methodology for Usng Moble Phones n Lve Remote Montorng of Physcal Actvtes Hamed Ketabdar and Matt Lyra Qualty and Usablty Lab, Deutsche Telekom Laboratores, TU Berln hamed.ketabdar@telekom.de,
More informationAN APPOINTMENT ORDER OUTPATIENT SCHEDULING SYSTEM THAT IMPROVES OUTPATIENT EXPERIENCE
AN APPOINTMENT ORDER OUTPATIENT SCHEDULING SYSTEM THAT IMPROVES OUTPATIENT EXPERIENCE Yu-L Huang Industral Engneerng Department New Mexco State Unversty Las Cruces, New Mexco 88003, U.S.A. Abstract Patent
More informationFault tolerance in cloud technologies presented as a service
Internatonal Scentfc Conference Computer Scence 2015 Pavel Dzhunev, PhD student Fault tolerance n cloud technologes presented as a servce INTRODUCTION Improvements n technques for vrtualzaton and performance
More informationA Simple Approach to Clustering in Excel
A Smple Approach to Clusterng n Excel Aravnd H Center for Computatonal Engneerng and Networng Amrta Vshwa Vdyapeetham, Combatore, Inda C Rajgopal Center for Computatonal Engneerng and Networng Amrta Vshwa
More informationA Multi-Camera System on PC-Cluster for Real-time 3-D Tracking
The 23 rd Conference of the Mechancal Engneerng Network of Thaland November 4 7, 2009, Chang Ma A Mult-Camera System on PC-Cluster for Real-tme 3-D Trackng Vboon Sangveraphunsr*, Krtsana Uttamang, and
More informationAn Adaptive Cross-layer Bandwidth Scheduling Strategy for the Speed-Sensitive Strategy in Hierarchical Cellular Networks
An Adaptve Cross-layer Bandwdth Schedulng Strategy for the Speed-Senstve Strategy n erarchcal Cellular Networks Jong-Shn Chen #1, Me-Wen #2 Department of Informaton and Communcaton Engneerng ChaoYang Unversty
More informationA DATA MINING APPLICATION IN A STUDENT DATABASE
JOURNAL OF AERONAUTICS AND SPACE TECHNOLOGIES JULY 005 VOLUME NUMBER (53-57) A DATA MINING APPLICATION IN A STUDENT DATABASE Şenol Zafer ERDOĞAN Maltepe Ünversty Faculty of Engneerng Büyükbakkalköy-Istanbul
More information1.1 The University may award Higher Doctorate degrees as specified from time-to-time in UPR AS11 1.
HIGHER DOCTORATE DEGREES SUMMARY OF PRINCIPAL CHANGES General changes None Secton 3.2 Refer to text (Amendments to verson 03.0, UPR AS02 are shown n talcs.) 1 INTRODUCTION 1.1 The Unversty may award Hgher
More informationECE544NA Final Project: Robust Machine Learning Hardware via Classifier Ensemble
1 ECE544NA Fnal Project: Robust Machne Learnng Hardware va Classfer Ensemble Sa Zhang, szhang12@llnos.edu Dept. of Electr. & Comput. Eng., Unv. of Illnos at Urbana-Champagn, Urbana, IL, USA Abstract In
More informationMax-Margin Early Event Detectors
Max-Margn Early Event Detectors Mnh Hoa Fernando De la Torre Robotcs Insttute, Carnege Mellon Unversty Abstract The need for early detecton of temporal events from sequental data arses n a wde spectrum
More informationA Multi-mode Image Tracking System Based on Distributed Fusion
A Mult-mode Image Tracng System Based on Dstrbuted Fuson Ln zheng Chongzhao Han Dongguang Zuo Hongsen Yan School of Electroncs & nformaton engneerng, X an Jaotong Unversty X an, Shaanx, Chna Lnzheng@malst.xjtu.edu.cn
More informationAn artificial Neural Network approach to monitor and diagnose multi-attribute quality control processes. S. T. A. Niaki*
Journal of Industral Engneerng Internatonal July 008, Vol. 4, No. 7, 04 Islamc Azad Unversty, South Tehran Branch An artfcal Neural Network approach to montor and dagnose multattrbute qualty control processes
More informationA spam filtering model based on immune mechanism
Avalable onlne www.jocpr.com Journal of Chemcal and Pharmaceutcal Research, 2014, 6(7):2533-2540 Research Artcle ISSN : 0975-7384 CODEN(USA) : JCPRC5 A spam flterng model based on mmune mechansm Ya-png
More informationPolitecnico di Torino. Porto Institutional Repository
Poltecnco d Torno Porto Insttutonal Repostory [Artcle] A cost-effectve cloud computng framework for acceleratng multmeda communcaton smulatons Orgnal Ctaton: D. Angel, E. Masala (2012). A cost-effectve
More informationDocument image template matching based on component block list
Pattern Recognton Letters 22 2001) 1033±1042 www.elsever.nl/locate/patrec Document mage template matchng based on component block lst Hanchuan Peng a,b,c, *, Fuhu Long b, Zheru Ch b, Wan-Ch Su b a Department
More informationIMPACT ANALYSIS OF A CELLULAR PHONE
4 th ASA & μeta Internatonal Conference IMPACT AALYSIS OF A CELLULAR PHOE We Lu, 2 Hongy L Bejng FEAonlne Engneerng Co.,Ltd. Bejng, Chna ABSTRACT Drop test smulaton plays an mportant role n nvestgatng
More informationData Broadcast on a Multi-System Heterogeneous Overlayed Wireless Network *
JOURNAL OF INFORMATION SCIENCE AND ENGINEERING 24, 819-840 (2008) Data Broadcast on a Mult-System Heterogeneous Overlayed Wreless Network * Department of Computer Scence Natonal Chao Tung Unversty Hsnchu,
More informationProperties of Indoor Received Signal Strength for WLAN Location Fingerprinting
Propertes of Indoor Receved Sgnal Strength for WLAN Locaton Fngerprntng Kamol Kaemarungs and Prashant Krshnamurthy Telecommuncatons Program, School of Informaton Scences, Unversty of Pttsburgh E-mal: kakst2,prashk@ptt.edu
More informationEnabling P2P One-view Multi-party Video Conferencing
Enablng P2P One-vew Mult-party Vdeo Conferencng Yongxang Zhao, Yong Lu, Changja Chen, and JanYn Zhang Abstract Mult-Party Vdeo Conferencng (MPVC) facltates realtme group nteracton between users. Whle P2P
More informationAudio Data Mining Using Multi-perceptron Artificial Neural Network
224 IJCSNS Internatonal Journal of Computer Scence and Network Securty, VOL.8 No.0, October 2008 Audo Data Mnng Usng Mult-perceptron Artfcal Neural Network Surendra Shetty, 2 K.K. Achary Dept of Computer
More informationDamage detection in composite laminates using coin-tap method
Damage detecton n composte lamnates usng con-tap method S.J. Km Korea Aerospace Research Insttute, 45 Eoeun-Dong, Youseong-Gu, 35-333 Daejeon, Republc of Korea yaeln@kar.re.kr 45 The con-tap test has the
More informationReal-Time Process Scheduling
Real-Tme Process Schedulng ktw@cse.ntu.edu.tw (Real-Tme and Embedded Systems Laboratory) Independent Process Schedulng Processes share nothng but CPU Papers for dscussons: C.L. Lu and James. W. Layland,
More informationRESEARCH ON DUAL-SHAKER SINE VIBRATION CONTROL. Yaoqi FENG 1, Hanping QIU 1. China Academy of Space Technology (CAST) yaoqi.feng@yahoo.
ICSV4 Carns Australa 9- July, 007 RESEARCH ON DUAL-SHAKER SINE VIBRATION CONTROL Yaoq FENG, Hanpng QIU Dynamc Test Laboratory, BISEE Chna Academy of Space Technology (CAST) yaoq.feng@yahoo.com Abstract
More informationOffline Verification of Hand Written Signature using Adaptive Resonance Theory Net (Type-1)
Internatonal Journal of Sgnal Processng Systems Vol, No June 203 Offlne Verfcaton of Hand Wrtten Sgnature usng Adaptve Resonance Theory Net (Type-) Trtharaj Dash Veer Surendra Sa Unversty of Technology,
More informationRisk-based Fatigue Estimate of Deep Water Risers -- Course Project for EM388F: Fracture Mechanics, Spring 2008
Rsk-based Fatgue Estmate of Deep Water Rsers -- Course Project for EM388F: Fracture Mechancs, Sprng 2008 Chen Sh Department of Cvl, Archtectural, and Envronmental Engneerng The Unversty of Texas at Austn
More informationPerformance Analysis and Coding Strategy of ECOC SVMs
Internatonal Journal of Grd and Dstrbuted Computng Vol.7, No. (04), pp.67-76 http://dx.do.org/0.457/jgdc.04.7..07 Performance Analyss and Codng Strategy of ECOC SVMs Zhgang Yan, and Yuanxuan Yang, School
More informationA neuro-fuzzy collaborative filtering approach for Web recommendation. G. Castellano, A. M. Fanelli, and M. A. Torsello *
Internatonal Journal of Computatonal Scence 992-6669 (Prnt) 992-6677 (Onlne) Global Informaton Publsher 27, Vol., No., 27-39 A neuro-fuzzy collaboratve flterng approach for Web recommendaton G. Castellano,
More informationDescriptive Models. Cluster Analysis. Example. General Applications of Clustering. Examples of Clustering Applications
CMSC828G Prncples of Data Mnng Lecture #9 Today s Readng: HMS, chapter 9 Today s Lecture: Descrptve Modelng Clusterng Algorthms Descrptve Models model presents the man features of the data, a global summary
More informationNetwork Aware Load-Balancing via Parallel VM Migration for Data Centers
Network Aware Load-Balancng va Parallel VM Mgraton for Data Centers Kun-Tng Chen 2, Chen Chen 12, Po-Hsang Wang 2 1 Informaton Technology Servce Center, 2 Department of Computer Scence Natonal Chao Tung
More informationCalculating the high frequency transmission line parameters of power cables
< ' Calculatng the hgh frequency transmsson lne parameters of power cables Authors: Dr. John Dcknson, Laboratory Servces Manager, N 0 RW E B Communcatons Mr. Peter J. Ncholson, Project Assgnment Manager,
More informationBERNSTEIN POLYNOMIALS
On-Lne Geometrc Modelng Notes BERNSTEIN POLYNOMIALS Kenneth I. Joy Vsualzaton and Graphcs Research Group Department of Computer Scence Unversty of Calforna, Davs Overvew Polynomals are ncredbly useful
More informationMethodology to Determine Relationships between Performance Factors in Hadoop Cloud Computing Applications
Methodology to Determne Relatonshps between Performance Factors n Hadoop Cloud Computng Applcatons Lus Eduardo Bautsta Vllalpando 1,2, Alan Aprl 1 and Alan Abran 1 1 Department of Software Engneerng and
More informationAn interactive system for structure-based ASCII art creation
An nteractve system for structure-based ASCII art creaton Katsunor Myake Henry Johan Tomoyuk Nshta The Unversty of Tokyo Nanyang Technologcal Unversty Abstract Non-Photorealstc Renderng (NPR), whose am
More informationEVALUATING THE PERCEIVED QUALITY OF INFRASTRUCTURE-LESS VOIP. Kun-chan Lan and Tsung-hsun Wu
EVALUATING THE PERCEIVED QUALITY OF INFRASTRUCTURE-LESS VOIP Kun-chan Lan and Tsung-hsun Wu Natonal Cheng Kung Unversty klan@cse.ncku.edu.tw, ryan@cse.ncku.edu.tw ABSTRACT Voce over IP (VoIP) s one of
More informationADVERTISEMENT FOR THE POST OF DIRECTOR, lim TIRUCHIRAPPALLI
ADVERTSEMENT FOR THE POST OF DRECTOR, lm TRUCHRAPPALL The ndan nsttute of Management Truchrappall (MT), establshed n 2011 n the regon of Taml Nadu s a leadng management school n nda. ts vson s "Preparng
More information