ADAPTIVE NETWORK BASED FUZZY INFERENCE SYSTEM FOR SPEECH RECOGNITION THROUGH SUBTRACTIVE CLUSTERING
|
|
- Margaret Vivian Morton
- 6 years ago
- Views:
Transcription
1 Internatonal Journal of Artfcal Intellgence & Applcatons (IJAIA), Vol. 5, No. 6, November 014 ADAPTIVE NETWORK BASED FUZZY INFERENCE SYSTEM FOR SPEECH RECOGNITION THROUGH SUBTRACTIVE CLUSTERING Samya Slarb 1, Bendahmane Abderrahmane ; and Abdelkader Benyettou 3 Faculty of mathematcal and computer scence, department of Computer Scence, Unversty of Scences and Technology Oran USTO-MB, Algera. ABSTRACT Fuzzy modelng requre two man steps whch are structure dentfcaton and parameter optmzaton, the frst one determnes the numbers of membershp functons and fuzzy f-then rules, whle the second dentfes a feasble set of parameters under the gven structure. However, the ncrease of nput dmenson, rule numbers wll have an exponental growth and there wll cause problem of rule dsaster. In ths paper, we have appled adaptve network fuzzy nference system ANFIS for phonemes recognton. The approprate learnng algorthm s performed on TIMIT speech database supervsed type, a pre-processng of the acoustc sgnal and extractng the coeffcents MFCCs parameters relevant to the recognton system. Frst learnng of the network structure by subtractve clusterng, n order to defne an optmal structure and obtan small number of rules, then learnng of parameters network by hybrd learnng whch combne the gradent decent and least square estmaton LSE to fnd a feasble set of antecedents and consequents parameters. The results obtaned show the effectveness of the method n terms of recognton rate and number of fuzzy rules generated. KEYWORDS ANFIS, subtractve clusterng, Phoneme, recognton. 1. INTRODUCTION Automatc Speech Recognton has acheved substantal success n the past few decades but more studes are needed because none of the current methods are fast and precse enough to be comparable wth human recognton abltes [1,]. Many algorthms and schemes based on dfferent mathematcal paradgms have been proposed n an attempt to mprove recognton rates [3,4,5,6]. Neuro-fuzzy modelng s a combnaton of fuzzy logc and neural network that takes advantage of both approaches, process mprecse or vague data by fuzzy logc [7] and at the same tme by ntroducng learnng through neural network. Several archtectures have been proposed dependng on the type of rule they nclude Mamdan or Sugeno [8] [9] one of the most nfluental fuzzy models has been proposed by Robert Jang n [10] called Adaptve Network Based Fuzzy Inference System ANFIS. The rule base of ths model contans the fuzzy f-then rule of Takag and Sugeno's type n whch consequent parts are lnear functons of nputs nstead of fuzzy sets, reducng the number of requred fuzzy rules. The dentfcaton of fuzzy model conssts of two major phases: structure dentfcaton and parameter optmzaton. The frst phase s the determnaton of number of fuzzy f-then rules and DOI : /jaa
2 Internatonal Journal of Artfcal Intellgence & Applcatons (IJAIA), Vol. 5, No. 6, November 014 membershp functons of the premse fuzzy sets whle the second phase s the tunng of the parameter values of the fuzzy model [11]. However, there wll be two problems f we drectly use the tradtonal Takag Sugeno model for the speech recognton system. The frst one s the network reasonng wll fal f the nput dmenson s too large. The second problem s that wth the ncrease of nput dmenson, rule numbers wll have an exponental growth and cause rule dsaster. Thus, determnaton of an approprate structure becomes an mportant ssue, then clusterng technques are appled to solve ths problem. [1] [13] [14]. Ths paper present a neural fuzzy system ANFIS for speech recognton. The approprate learnng algorthm s performed on TIMIT speech database supervsed type, a pre-processng of the acoustc sgnal and extractng the coeffcents MFCCs parameters relevant to the recognton system. Subtractve clusterng s appled n order to defne an optmal structure and obtan small number of rules, then learnng of parameters network by hybrd learnng whch combne the gradent decent and least square estmaton LSE. The paper s organzed as follows: the secton revews the lterature research work, n secton 3 detals the structure and Prncpe of learnng of ANFIS, whle secton 4 descrbes subtractve clusterng; expermental results and dscusson were detaled n secton 5. Secton 6 concludes the paper.. RELATED WORK The ANFIS has the advantage of good applcablty because t can be nterpreted as local lnearzaton modelng, and even as conventonal lnear technques for both state estmaton and state control whch are drectly applcable. Ths adaptve network has good ablty and performance n system dentfcaton, predcton and control has been appled n many dfferent systems. Snce there are not many research works used ANFIS n speech recognton, t s necessary to carry on the exploraton and the thorough research on t. ANFIS was used for Speaker verfcaton [15] usng combnatonal features of MFCC; Lnear Predcton Coeffcents LPC and the frst fve formants; Recognton of dscrete words [16], Speech emoton verfcaton [17] based on MFCC for real tme applcaton. In [18] ANFIS was performed to reduce nose and enhance speech, also for the recognton of solated dgts wth speaker-ndependent [19]. Speaker, language and word recognton was completed by ANFIS [0], furthermore for caller behavor classfcaton [1]. All of these works had used clusterng technques to determne the structure of ANFIS. Another work: an automated gender classfcaton s performed by ANFIS []. 3. ADAPTIVE NETWORK BASED FUZZY INFERENCE SYSTEM ANFIS 3.1. The Concept and Structure ANFIS proposed by Jang n 1993 mult-layered neural network whch connectons are not weghted or all weghts equal 1[10], s alternate method whch combnes the advantages of two ntellgent approaches neural network and fuzzy logc to allow good reasonng n quantty and qualty. A network obtaned has an excellent ablty of tranng by means of neural networks and lngustc nterpretaton of varables va fuzzy logc. The both of them encode the nformaton n parallel and dstrbute archtecture n a numercal framework. ANFIS mplement a frst order Sugeno style fuzzy system; t apples the rule of TSK Takag Sugeno and Kang form n ts archtecture. 44
3 Internatonal Journal of Artfcal Intellgence & Applcatons (IJAIA), Vol. 5, No. 6, November 014 Rule: f x s A1 and y s B1 then f ( x) px + qy + r Where x and y are the nputs, A and B are the fuzzy sets, f are the output, p, q and r are the desgn parameters that determned durng the tranng process. ANFIS s composed of two parts s the frst part s the antecedent and the second part s the concluson, whch are connected to each other by rules n network form. Fve layers are used to construct ths network. Each layer contans several node sts structure shows n fgure 1. layer1: executes a fuzzfcaton process whch denotes membershp functons (MFs) to each nput. In ths paper we choose Gaussan functons as membershp functons: 1 O ( x c ) µ exp A σ (1) layer: executes the fuzzy AND of antecedents part of the fuzzy rules o w µ µ A B ( x ) ( ), 1,,3, 4 1 x () layer3: normalzes the MFs o 3 w w w 4 j 1 j, 1,,3, 4 (3) layer4: executes the concluson part of fuzzy rules O 4 ( x + α x + α ), 1,,3,4. w y w α (4) 1 3 layer5: computes the output of fuzzy system by summng up the outputs of the fourth layer whch s the defuzzfcaton process. O overll _ output w y w w y (5) Fgure1. ANFIS archtecture. Crcles n ANFIS represent fxed nodes that predefned operators to ther nputs and no other parameters but the nput partcpate n ther calculatons. Whle square are the representatve for adaptve nodes that affected by nternal parameters. 45
4 Internatonal Journal of Artfcal Intellgence & Applcatons (IJAIA), Vol. 5, No. 6, November Learnng Algorthm The parameters to be tuned n an ANFIS are the membershp functon parameters of each nput, the consequents parameters also number of rules. nbr rule n _ m (6) Where n s number of nputs and m number of membershp functons by nput and they generate all the possble rules. Two steps of tranng are necessary whch are: Structure learnng whch allows to determnate the approprate structure of network, that s, the best parttonng of the nput space (number of membershp functons for each nput, number of rules). And parametrc learnng carred out to adjust the membershp functons and consequents parameters. In most systems the structure s fxed a pror by experts. In our work we combne both of learnng sequentally The subsequent to the development of ANFIS approach, a number of methods have been proposed for learnng rules and for obtanng an optmal set of rules. For example, Mascol et al [3] have proposed to merge Mn-Max and ANFIS model to obtan neuro-fuzzy network and determne optmal set of fuzzy rules. Jang and Mzutan [4] have presented applcaton of Lavenberg-Marquardt method, whch s essentally a nonlnear least-squares technque, for learnng n ANFIS network. In another paper, Jang [5] has presented a scheme for nput selecton and [6] used Kohonen s map to tranng. Four methods have been proposed by Jang [11] for update the parameters of ANFIS: All parameters are update by only gradent decent. In frst the consequents parameters are obtaned by applcaton of least square estmaton LSE only once and then the gradent decent update all parameters. Sequental LSE that s usng extended Kalman flter to update all parameters. Hybrd learnng: whch combne the gradent decent and LSE to fnd a feasble set of antecedents and consequents parameters. The most common tranng algorthm s the hybrd learnng. ths algorthm s carred out n two steps: forward pass and backward pass, Once all the parameters are ntalzed, n forward pass, functonal sgnals go forward tll fourth layer and the consequents parameters are dentfed by LSE. After dentfyng consequents parameters, the functonal sgnals keep gong forward untl the error measure s calculated. In the backward pass, the error rates propagate backward and the premse parameters are updated by gradent decent. The functon to be mnmzed s Root Mean Square Error (RMSE): RMSE N 1 N 1 ( ) d o (7) Where d s the desred output and O s the ANFIS output for the th sample from tranng data and N s the number of tranng samples. 46
5 Internatonal Journal of Artfcal Intellgence & Applcatons (IJAIA), Vol. 5, No. 6, November SUBTRACTIVE CLUSTERING Clusterng s the classfcaton of objects nto dfferent groups, or more precsely, the parttonng of a data set nto subsets (clusters), so that the data n each subset shares some common features, often proxmty accordng to some defned dstance measure. From the machne learnng perspectve, clusterng can be vewed as unsupervsed learnng of concepts [7][8]. as: 1. Compute the ntal potental value for each data pont x s defned P n j 1 exp x x ( ra ) j (8). Where ra s a postve constant representng a neghborhood radus. Therefore, a pont would have a heght potental value f t has more neghbor ponts close to tself. 3. the pont wth the hghest potental value s selected as le frst cluster center: Frst cluster center x c1 s chosen as the pont havng the largest densty value D c1 4. P ( 1) * ( 1 ( ) max P ( x ) (9) 5. The potental value of each data pont x s reduced as follows: P x x c 1 P Pc 1 exp ( rb ) Where ( rb β ra ) s a postve constant represent the radus of the neghborhood for whch sgnfcant potental reducton wll occur. β s a parameter called as squash factor, whch s multply by radus values to determne the neghborng clusters wthn whch the exstence of other cluster centers are dscouraged. After revsng the densty functon, the next cluster center s selected as the pont havng the greatest potental value. Ths process s repeated to generate the cluster centers untl maxmum potental value n the current teraton s equal or less than the threshold. 5. STRUCTURE LEARNING ALGORITHM FOR T-S FUZZY NEURAL NETWORK ANFIS offers three approaches to dentfy cluster namely grd parttonng, subtractve clusterng and fuzzy c-means clusterng. Grd parttonng approach s useful f the number of features s no more than 6 or 7. If the number of features s too hgh then ths method wll cannot be used as the memory requrement wll be nsuffcent when usng MATLAB. For fuzzy c-means method, the number of clusters for the dataset needs to be specfed. Snce no pror knowledge on the number of clusters s avalable, subtractve clusterng wll be deal. The radus of the clusters wll be used as the bass of the cluster formaton. A range of radus from 0.0 to. has been chosen n order to produce optmum number of clusters for the ntal fuzzy nference system (FIS). The ntal FIS then was fed to the ANFIS for refnng the FIS so that t wll tune the supervsed learnng teratvely wth more detaled rules generated. (10) 47
6 Internatonal Journal of Artfcal Intellgence & Applcatons (IJAIA), Vol. 5, No. 6, November 014 Each cluster center represents a fuzzy f-then-rule. The nth column of cth cluster center s assumed to be the mean value (c n ) of the assocated Gaussan membershp functon defned for cth fuzzy set of nth nput varable. Then, the standard devaton (a n ) of the above mentoned Gaussan functons are calculated as below: Therefore the cluster centers and squash factors may be vewed as parameters, whch the number of fuzzy rules of the ntal FIS depend on, before the rule base parameters of the FIS s tuned by ANNs n ANFIS. 6. EXPERIMENTAL AND RESULTS 6.1. TIMIT Database A reduced subset of TIMIT [9] -Phonetc Contnuous Speech Corpus (TIMIT Texas Instruments (TI) and Massachusetts Insttute of Technology (MIT)) used n our work, whch s the abrdged verson of the complete testng set, conssts of 19 utterances, 8 from each of 4 speakers ( males and 1 female from each dalect regon). In general, phonemes can grouped nto categores based on dstnctve features that ndcate a smlarty n artculatory, acoustc and perceptual. The major classes of phonetc TIMIT database are: 6 vowels {/ah/, /aw/, /ax/, /ax-h/, /uh/, /uw/}, 6 frcatves {/dh/, /f/, /sh/, /v/, /z/, /zh/} and 6 plosves {/b/, /d/, /g/, /p/, /q/, /t/}. 6.. Codng Mel-Frequency Cepstral Coeffcents MFCC Before any calculaton, t s necessary to effect some operatons to shape the speech sgnal. The sgnal s frst fltered and then sampled at a gven frequency (16 KHZ). Pre-emphass s carred out to rase the hgh frequences. Then, the sgnal s segmented nto frames, each frame has a fxed number N5ms of speech samples (perod n whch the speech sgnal can be consdered as statonary).treatng small fragments of sgnal leads to flterng problems (edge effects). In order to reduce edge effects we use weghts wndows (Hammng wndow) these are functons that are appled to set of samples n the wndow of the orgnal sgnal. After sgnal shapng a dscrete Fourer transform DFT partcularly fast Fourer transform FFT s appled to pass n the frequency doman and for extractng the spectrum sgnal. Then, the flterng s performed by multplyng the spectrum obtaned by flters (trangular flters). It s possble to employ the output of the flter bank as nput to the recognton system. However, other factors derved from the outputs of a bank of flters are more dscrmnatve, more robust and less correlated. It s from the cepstral coeffcents derved of the flter bank outputs dstrbuted lnearly on the Mel scale, these are the Mel-frequency Cepstral coeffcents MFCC Each wndow provdes 1 MFCC coeffcents and the correspondng resdual energy Results The corpus that we have used n the classfcaton of phonemes conssts of 6 vowels, 6 frcatves and 6 plosves: nstances for learnng and 1055 nstances for test. Applyng subtractve clusterng to determnate the ntal structure of ANFIS, then Hybrd learnng to fnd a feasble set of antecedents and consequents parameters. (11) 48
7 Internatonal Journal of Artfcal Intellgence & Applcatons (IJAIA), Vol. 5, No. 6, November 014 We appled ANFIS on the classfcaton of phonemes by varyng the radus of the clusterng between [0.0-.5], results are summarzed n the followng tables: Wth a radus between [0-1.4] we acheved a recognton rate of 100 but the number of generated rules was too bg, so that exceeds 00 rules whch has resulted a large runtme tme and especally for real-tme applcaton, and ths for the three classes of phonemes used. For vowels we have obtaned the best recognton rate of 100 and 6 rules wth a radus of.0 and 1.9, we have reduced the number of rules from 6 to 4 wth a radus of.1 and acheved recognton rate of 83 (table 1). For frcatves wth a radus of 1.4 we have obtaned the best recognton rate that reaches 100 and 5 fuzzy rules, we could reduce the number of rules to 4 wth a recognton rate of (table ). For plosves we reached 100 recognton rate and 5 rules wth a radus of 1.4 and 1.5 beyond 1.5 we had 3 rules but reducton of recognton (table 3). Fgure. Rate recognton dependng of radus Fgure3. Error dependng of radus 49
8 Internatonal Journal of Artfcal Intellgence & Applcatons (IJAIA), Vol. 5, No. 6, November 014 Table1. Vowels. Raterecognton error Number-rules Radus tran test tran test tran test Table. Frcatves. Rate-recognton error Number-rules Radus tran test Tran test tran test Radus Table 3. Plosves. Rate-recognton error Number-rules tran test tran test tran test A large number of rules are generated by reducng the radus of clusterng so most of data are correctly classfed but t takes sgnfcant tme. Contrarwse, by ncreasng the radus of clusterng we had very few rules so many alternatves are not consdered by the fuzzy rules, then a lot of data are not correctly classfed. We could see that there was a compromse between the recognton rate and the number of rules generated, wth a radus between [ ] we were able to acheve perfecton on the recognton rate and number of generated rules whch engender a very reasonable computaton tme. 50
9 Internatonal Journal of Artfcal Intellgence & Applcatons (IJAIA), Vol. 5, No. 6, November CONCLUSION AND FUTURE WORK In ths paper, we have appled adaptve network fuzzy nference system for phonemes recognton. Frst learnng of the network structure by subtractve clusterng, n order to defne an optmal structure and obtan small number of rules, then learnng of parameters network by hybrd learnng whch combne the gradent decent and least square estmaton LSE to fnd a feasble set of antecedents and consequents parameters. The approprate learnng algorthm s performed on TIMIT speech database supervsed type, a pre-processng of the acoustc sgnal and extractng the coeffcents MFCCs parameters relevant to the recognton system. Fnally, hybrd learnng combnes the gradent decent and least square estmaton LSE of parameters network. The results obtaned show the effectveness of the method n terms of recognton rate and number of fuzzy rules generated. For future work, consder ths adaptve network for whole dataset TIMIT, and mprovng learnng algorthm by tunng parameters of ANFIS by means of evolutonary algorthms. REFERENCES [1] R. Reddy (001), Spoken Language Processng: A gude to Theory, Algorthm, And System Development, Prentce-Hall, New Jersey. [] X. He, L. Deng (008), Dscrmnatve Learnng for Speech Recognton: Theory and practce, Morgan & Claypool. [3] Cole, Ron, Hrschman, Lynette, Atlas, Les, Beckman, Mary, Bermann, Alan, Bush, Marca, et al (1995) The challenge of spoken language systems: Research drectons for the nnetes. IEEE Transactons on Speech and Audo Processng, vol 3(ssue 1). [4] Scofeld, M. C (1991) Neural networks and speech processng. Amsterdam: Kluwer Academc. [5] Yallop, C. C. (1990) An ntroducton to phonetcs and phonology. Cambrdge, MA: Blackwell. [6] Carla Lopes and Fernando Perdgão. chapter 14 (011) Phone Recognton on the TIMIT Database. Book Speech Technologes Edted by Ivo Ipsc, ISBN , 43 pages, Publsher InTech. [7] George Bojadzev,Mara Bojadzev (1995) Advances n fuzzy systems applcatons and theory, vol 5 book Fuzzy Sets, Fuzzy Logc, Applcatons, World Scentfc. [8] Ajth Abraham (001) Neuro Fuzzy Systems: State-of-the-Art Modelng, Technques n Proceedngs of 6th Internatonal Work-Conference on Artfcal and Natural Neural Networks, IWANN 001 Granada, Span, June 13 15, Sprnger Verlag Germany, vol 084, pp [9] B. Kosko (1991) Neural Networks and Fuzzy Systems A Dynamc Systems Approach, Prentce Hall. [10] J.S.R. Jang (1993) ANFIS: Adaptve Network Based Fuzzy Inference Systems, IEEE Trans, Syst. Man Cybernet. vol 3, No 3, pp [11] J.S.R. Jang, C.T.Sun and E.Mzutan (1997) Neuro-fuuzy and soft computng: a computatonal approach to learnng and machne ntellgence, London: prentce-hall nternatonal, USA. [1] Agus Pryono, Muhammad Rdwan, Ahmad Jas Alas, Rza Atq O. K. Rahmat, Azm Hassan & Mohd. Alauddn Mohd. Al (005) Generaton of fuzzy rules wth subtractve clusterng. Jurnal Teknolog, 43(D). Unverst Teknolog Malaysa, pp [13] Chuen-Tsa Sun (1994) Rule-Base Structure Identfcaton n an Adaptve-Network-Based Fuzzy Inference System. IEEE Transacton on Fuzzy Systems, vol., no. 1. pp [14] Chng-Chang Wong and Cha-Chong Chen (1999) A Hybrd Clusterng and Gradent Descent Approach for Fuzzy Modelng. IEEE Transactons on systems, man, and cybernetcs part b: cybernetcs, vol. 9, no. 6. pp [15] V.Srhar, R.Karthk and R.Antha and S.D.Suganth (010) Speaker verfcaton usng combnatonal features and adaptve neuro-fuzzy nference systems, IIMT 10 December 8-30, Allahabad, UP, Inda, pp
10 Internatonal Journal of Artfcal Intellgence & Applcatons (IJAIA), Vol. 5, No. 6, November 014 [16] N.Helm, B.H.Helm (008 ) Speech recognton wth fuzzy neural network for dscrete words, ICNC Fourth Internatonal Conference on Natural Computaton. vol 07, pp [17] N.Kamaruddn, A.Wahab (008) Speech emoton verfcaton system (sevs) based on mfcc for real tme applcaton, Intellgent Envronments, 008 IET 4th Internatonal Conference on, Seattle, pp [18] Reem Sabah & Raja N. Anon (009) Isolated Dgt Speech Recognton n Malay Language usng Neuro-Fuzzy Approach, Thrd Asa Internatonal Conference on Modellng & Smulaton AMS,, Bal Asa, pp [19] Jasmn Thevarl and H.K.Kwan (005) Speech Enhancement usng Adaptve Neuro-Fuzzy Flterng n Proceedngs of Internatonal Symposum on Intellgent Sgnal Processng and Communcaton Systems ISPACS, December 13-16, pp [0] Bpul Pandey, Alok Ranjan, Rajeev Kumar and Anupam Shukla (010) Multlngual Speaker Recognton Usng ANFIS. n nd Internatonal Conference on Sgnal Processng Systems (ICSPS) vol 3, V V [1] Pretesh. B. Patel, Tshldz Marwala (011) Adaptve Neuro Fuzzy Inference System, Neural Network and Support Vector Machne for caller behavor classfcaton, 10th Internatonal Conference on Machne Learnng and Applcatons ICMLA, 18-1 December, vol 1, pp [] Sachn Lakra, Juh Sngh and Arun Kumar Sngh (013) Automated Ptch-Based Gender Recognton usng an Adaptve Neuro-Fuzzy Inference System, Internatonal Conference on Intellgent Systems and Sgnal Processng (ISSP), pp [3] Gujarat.Mansh Kumar, Devendra P. Garg (004) Intellgent Learnng of Fuzzy Logc Controllers va Neural Network and Genetc Algorthm, Proceedngs of 004 JUSFA Japan USA Symposum on Flexble Automaton. Denver, Colorado, pp [4] Mascol, F.M., Varaz, G.M. and Martnell, G (1997) Constructve Algorthm for Neuro-Fuzzy Networks, Proceedngs of the Sxth IEEE Internatonal Conference on Fuzzy Systems, Vol. 1, pp [5] Jang, J.-S. R., and Mzutan, E (1996) Levenberg-Marquardt Method for ANFIS Learnng, Bennal Conference of the North Amercan Fuzzy Informaton Processng Socety, pp [6] Jang, J.-S.R. (1996) Input Selecton for ANFIS Learnng, Proceedngs of the Ffth IEEE Internatonal Conference on Fuzzy Systems, Vol., pp [7] J.Han, M.Kamber (000) Data Mnng: Concepts and Technques chapter 8. The Morgan Kaufmann Seres n Data Management Systems, Jm Gray, Seres Edtor Morgan Kaufmann Publshers. [8] V.Estvll-Castro, J.Yang (000) A Fast and robust general purpose clusterng algorthm. Pacfc Rm Internatonal Conference on Artfcal Intellgence, pp [9] Zue, V.; Seneff, S. & Glass J (1990) Speech database development at MIT: TIMIT and beyond, Speech Communcaton, Vol. 9, No. 4, pp
Training ANFIS Structure with Modified PSO Algorithm
Proceedngs of the 5th Medterranean Conference on Control & Automaton, July 7-9, 007, Athens - Greece T4-003 Tranng ANFIS Structure wth Modfed PSO Algorthm V.Seyd Ghomsheh *, M. Alyar Shoorehdel **, M.
More informationOptimal Design of Nonlinear Fuzzy Model by Means of Independent Fuzzy Scatter Partition
Optmal Desgn of onlnear Fuzzy Model by Means of Independent Fuzzy Scatter Partton Keon-Jun Park, Hyung-Kl Kang and Yong-Kab Km *, Department of Informaton and Communcaton Engneerng, Wonkwang Unversty,
More informationType-2 Fuzzy Non-uniform Rational B-spline Model with Type-2 Fuzzy Data
Malaysan Journal of Mathematcal Scences 11(S) Aprl : 35 46 (2017) Specal Issue: The 2nd Internatonal Conference and Workshop on Mathematcal Analyss (ICWOMA 2016) MALAYSIAN JOURNAL OF MATHEMATICAL SCIENCES
More informationSum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints
Australan Journal of Basc and Appled Scences, 2(4): 1204-1208, 2008 ISSN 1991-8178 Sum of Lnear and Fractonal Multobjectve Programmng Problem under Fuzzy Rules Constrants 1 2 Sanjay Jan and Kalash Lachhwan
More informationLearning the Kernel Parameters in Kernel Minimum Distance Classifier
Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department
More informationSupport Vector Machines
/9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.
More informationA Binarization Algorithm specialized on Document Images and Photos
A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a
More informationParallelism for Nested Loops with Non-uniform and Flow Dependences
Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr
More informationTuning of Fuzzy Inference Systems Through Unconstrained Optimization Techniques
Tunng of Fuzzy Inference Systems Through Unconstraned Optmzaton Technques ROGERIO ANDRADE FLAUZINO, IVAN NUNES DA SILVA Department of Electrcal Engneerng State Unversty of São Paulo UNESP CP 473, CEP 733-36,
More informationClassifying Acoustic Transient Signals Using Artificial Intelligence
Classfyng Acoustc Transent Sgnals Usng Artfcal Intellgence Steve Sutton, Unversty of North Carolna At Wlmngton (suttons@charter.net) Greg Huff, Unversty of North Carolna At Wlmngton (jgh7476@uncwl.edu)
More informationAn Entropy-Based Approach to Integrated Information Needs Assessment
Dstrbuton Statement A: Approved for publc release; dstrbuton s unlmted. An Entropy-Based Approach to ntegrated nformaton Needs Assessment June 8, 2004 Wllam J. Farrell Lockheed Martn Advanced Technology
More informationDetermining the Optimal Bandwidth Based on Multi-criterion Fusion
Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn
More informationFuzzy Filtering Algorithms for Image Processing: Performance Evaluation of Various Approaches
Proceedngs of the Internatonal Conference on Cognton and Recognton Fuzzy Flterng Algorthms for Image Processng: Performance Evaluaton of Varous Approaches Rajoo Pandey and Umesh Ghanekar Department of
More informationClassifier Selection Based on Data Complexity Measures *
Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.
More informationOutline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1
4/14/011 Outlne Dscrmnatve classfers for mage recognton Wednesday, Aprl 13 Krsten Grauman UT-Austn Last tme: wndow-based generc obect detecton basc ppelne face detecton wth boostng as case study Today:
More informationModule Management Tool in Software Development Organizations
Journal of Computer Scence (5): 8-, 7 ISSN 59-66 7 Scence Publcatons Management Tool n Software Development Organzatons Ahmad A. Al-Rababah and Mohammad A. Al-Rababah Faculty of IT, Al-Ahlyyah Amman Unversty,
More informationS1 Note. Basis functions.
S1 Note. Bass functons. Contents Types of bass functons...1 The Fourer bass...2 B-splne bass...3 Power and type I error rates wth dfferent numbers of bass functons...4 Table S1. Smulaton results of type
More informationCluster Analysis of Electrical Behavior
Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School
More informationNEURO-FUZZY MODELING IN BANKRUPTCY PREDICTION * D. VLACHOS Y. A. TOLIAS
Yugoslav Journal of Operatons Research 3 (23), Number 2, 65-74 NEURO-FUZZY MODELING IN BANKRUPTCY PREDICTION * D. VLACHOS Department of Mechancal Engneerng Arstotle Unversty of Thessalonk, Thessalonk,
More informationA Fast Content-Based Multimedia Retrieval Technique Using Compressed Data
A Fast Content-Based Multmeda Retreval Technque Usng Compressed Data Borko Furht and Pornvt Saksobhavvat NSF Multmeda Laboratory Florda Atlantc Unversty, Boca Raton, Florda 3343 ABSTRACT In ths paper,
More informationComputational Results of Hybrid Learning in Adaptive Neuro Fuzzy Inference System for Optimal Prediction
Internatonal Journal of Appled Engneerng Research ISSN 0973-456 Volume 1, Number 16 (017) pp. 5810-5818 Research Inda Publcatons. http://.rpublcaton.com Computatonal Results of Hybrd Learnng n Adaptve
More informationSubspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points;
Subspace clusterng Clusterng Fundamental to all clusterng technques s the choce of dstance measure between data ponts; D q ( ) ( ) 2 x x = x x, j k = 1 k jk Squared Eucldean dstance Assumpton: All features
More information(1) The control processes are too complex to analyze by conventional quantitative techniques.
Chapter 0 Fuzzy Control and Fuzzy Expert Systems The fuzzy logc controller (FLC) s ntroduced n ths chapter. After ntroducng the archtecture of the FLC, we study ts components step by step and suggest a
More informationImprovement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration
Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,
More informationGA-Based Learning Algorithms to Identify Fuzzy Rules for Fuzzy Neural Networks
Seventh Internatonal Conference on Intellgent Systems Desgn and Applcatons GA-Based Learnng Algorthms to Identfy Fuzzy Rules for Fuzzy Neural Networks K Almejall, K Dahal, Member IEEE, and A Hossan, Member
More informationOutline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like:
Self-Organzng Maps (SOM) Turgay İBRİKÇİ, PhD. Outlne Introducton Structures of SOM SOM Archtecture Neghborhoods SOM Algorthm Examples Summary 1 2 Unsupervsed Hebban Learnng US Hebban Learnng, Cntd 3 A
More informationANFIS Control Double-Inverted Pendulum
895 A publcaton of CHEMICAL ENGINEERING TRANSACTIONS VOL. 46, 05 Guest Edtors: Peyu Ren, Yancang L, Hupng Song Copyrght 05, AIDIC Servz S.r.l., ISBN 978-88-95608-37-; ISSN 83-96 The Italan Assocaton of
More informationAn Image Fusion Approach Based on Segmentation Region
Rong Wang, L-Qun Gao, Shu Yang, Yu-Hua Cha, and Yan-Chun Lu An Image Fuson Approach Based On Segmentaton Regon An Image Fuson Approach Based on Segmentaton Regon Rong Wang, L-Qun Gao, Shu Yang 3, Yu-Hua
More informationHigh-Boost Mesh Filtering for 3-D Shape Enhancement
Hgh-Boost Mesh Flterng for 3-D Shape Enhancement Hrokazu Yagou Λ Alexander Belyaev y Damng We z Λ y z ; ; Shape Modelng Laboratory, Unversty of Azu, Azu-Wakamatsu 965-8580 Japan y Computer Graphcs Group,
More informationMachine Learning 9. week
Machne Learnng 9. week Mappng Concept Radal Bass Functons (RBF) RBF Networks 1 Mappng It s probably the best scenaro for the classfcaton of two dataset s to separate them lnearly. As you see n the below
More informationContent Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers
IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth
More informationIncremental Learning with Support Vector Machines and Fuzzy Set Theory
The 25th Workshop on Combnatoral Mathematcs and Computaton Theory Incremental Learnng wth Support Vector Machnes and Fuzzy Set Theory Yu-Mng Chuang 1 and Cha-Hwa Ln 2* 1 Department of Computer Scence and
More informationA Fast Visual Tracking Algorithm Based on Circle Pixels Matching
A Fast Vsual Trackng Algorthm Based on Crcle Pxels Matchng Zhqang Hou hou_zhq@sohu.com Chongzhao Han czhan@mal.xjtu.edu.cn Ln Zheng Abstract: A fast vsual trackng algorthm based on crcle pxels matchng
More informationPerformance Evaluation of an ANFIS Based Power System Stabilizer Applied in Multi-Machine Power Systems
Performance Evaluaton of an ANFIS Based Power System Stablzer Appled n Mult-Machne Power Systems A. A GHARAVEISI 1,2 A.DARABI 3 M. MONADI 4 A. KHAJEH-ZADEH 5 M. RASHIDI-NEJAD 1,2,5 1. Shahd Bahonar Unversty
More informationStudy on Fuzzy Models of Wind Turbine Power Curve
Proceedngs of the 006 IASME/WSEAS Internatonal Conference on Energy & Envronmental Systems, Chalkda, Greece, May 8-0, 006 (pp-7) Study on Fuzzy Models of Wnd Turbne Power Curve SHU-CHEN WANG PEI-HWA HUANG
More informationTerm Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task
Proceedngs of NTCIR-6 Workshop Meetng, May 15-18, 2007, Tokyo, Japan Term Weghtng Classfcaton System Usng the Ch-square Statstc for the Classfcaton Subtask at NTCIR-6 Patent Retreval Task Kotaro Hashmoto
More informationFuzzy Model Identification Using Support Vector Clustering Method
Fuzzy Model Identfcaton Usng Support Vector Clusterng Method $\úhj OUçar, Yakup Demr, and Cüneyt * ]HOLú Electrcal and Electroncs Engneerng Department, Engneerng Faculty, ÕUDW Unversty, Elazg, Turkey agulucar@eee.org,ydemr@frat.edu.tr
More informationA PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION
1 THE PUBLISHING HOUSE PROCEEDINGS OF THE ROMANIAN ACADEMY, Seres A, OF THE ROMANIAN ACADEMY Volume 4, Number 2/2003, pp.000-000 A PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION Tudor BARBU Insttute
More informationProblem Definitions and Evaluation Criteria for Computational Expensive Optimization
Problem efntons and Evaluaton Crtera for Computatonal Expensve Optmzaton B. Lu 1, Q. Chen and Q. Zhang 3, J. J. Lang 4, P. N. Suganthan, B. Y. Qu 6 1 epartment of Computng, Glyndwr Unversty, UK Faclty
More informationFace Recognition Based on SVM and 2DPCA
Vol. 4, o. 3, September, 2011 Face Recognton Based on SVM and 2DPCA Tha Hoang Le, Len Bu Faculty of Informaton Technology, HCMC Unversty of Scence Faculty of Informaton Scences and Engneerng, Unversty
More informationA Framework of Modified Adaptive Fuzzy Inference Engine (MAFIE) and Its Application
Internatonal Journal of Computer Informaton Systems and Industral Management Applcatons. ISSN 150-7988 Volume 5 (013) pp. 66-670 MIR Labs, www.mrlabs.net/csm/ndex.html A Framework of Modfed Adaptve Fuzzy
More informationNovel Fuzzy logic Based Edge Detection Technique
Novel Fuzzy logc Based Edge Detecton Technque Aborsade, D.O Department of Electroncs Engneerng, adoke Akntola Unversty of Tech., Ogbomoso. Oyo-state. doaborsade@yahoo.com Abstract Ths paper s based on
More informationFace Recognition University at Buffalo CSE666 Lecture Slides Resources:
Face Recognton Unversty at Buffalo CSE666 Lecture Sldes Resources: http://www.face-rec.org/algorthms/ Overvew of face recognton algorthms Correlaton - Pxel based correspondence between two face mages Structural
More informationArtificial Intelligence (AI) methods are concerned with. Artificial Intelligence Techniques for Steam Generator Modelling
Artfcal Intellgence Technques for Steam Generator Modellng Sarah Wrght and Tshldz Marwala Abstract Ths paper nvestgates the use of dfferent Artfcal Intellgence methods to predct the values of several contnuous
More informationCS 534: Computer Vision Model Fitting
CS 534: Computer Vson Model Fttng Sprng 004 Ahmed Elgammal Dept of Computer Scence CS 534 Model Fttng - 1 Outlnes Model fttng s mportant Least-squares fttng Maxmum lkelhood estmaton MAP estmaton Robust
More informationMachine Learning. Topic 6: Clustering
Machne Learnng Topc 6: lusterng lusterng Groupng data nto (hopefully useful) sets. Thngs on the left Thngs on the rght Applcatons of lusterng Hypothess Generaton lusters mght suggest natural groups. Hypothess
More informationA New Approach For the Ranking of Fuzzy Sets With Different Heights
New pproach For the ankng of Fuzzy Sets Wth Dfferent Heghts Pushpnder Sngh School of Mathematcs Computer pplcatons Thapar Unversty, Patala-7 00 Inda pushpndersnl@gmalcom STCT ankng of fuzzy sets plays
More informationMULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION
MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION Paulo Quntlano 1 & Antono Santa-Rosa 1 Federal Polce Department, Brasla, Brazl. E-mals: quntlano.pqs@dpf.gov.br and
More informationSimulation Based Analysis of FAST TCP using OMNET++
Smulaton Based Analyss of FAST TCP usng OMNET++ Umar ul Hassan 04030038@lums.edu.pk Md Term Report CS678 Topcs n Internet Research Sprng, 2006 Introducton Internet traffc s doublng roughly every 3 months
More informationLoad Balancing for Hex-Cell Interconnection Network
Int. J. Communcatons, Network and System Scences,,, - Publshed Onlne Aprl n ScRes. http://www.scrp.org/journal/jcns http://dx.do.org/./jcns.. Load Balancng for Hex-Cell Interconnecton Network Saher Manaseer,
More informationSolving two-person zero-sum game by Matlab
Appled Mechancs and Materals Onlne: 2011-02-02 ISSN: 1662-7482, Vols. 50-51, pp 262-265 do:10.4028/www.scentfc.net/amm.50-51.262 2011 Trans Tech Publcatons, Swtzerland Solvng two-person zero-sum game by
More informationThe Research of Support Vector Machine in Agricultural Data Classification
The Research of Support Vector Machne n Agrcultural Data Classfcaton Le Sh, Qguo Duan, Xnmng Ma, Me Weng College of Informaton and Management Scence, HeNan Agrcultural Unversty, Zhengzhou 45000 Chna Zhengzhou
More informationEdge Detection in Noisy Images Using the Support Vector Machines
Edge Detecton n Nosy Images Usng the Support Vector Machnes Hlaro Gómez-Moreno, Saturnno Maldonado-Bascón, Francsco López-Ferreras Sgnal Theory and Communcatons Department. Unversty of Alcalá Crta. Madrd-Barcelona
More informationNetwork Intrusion Detection Based on PSO-SVM
TELKOMNIKA Indonesan Journal of Electrcal Engneerng Vol.1, No., February 014, pp. 150 ~ 1508 DOI: http://dx.do.org/10.11591/telkomnka.v1.386 150 Network Intruson Detecton Based on PSO-SVM Changsheng Xang*
More informationFUZZY LOGIC AND NEURO-FUZZY MODELING
, pp.-74-84. Avalable onlne at http://www.bonfo.n/contents.php?d=7 FUZZY LOGIC AND NEURO-FUZZY MODELING NIKAM S.R.*, NIKUMBH P.J. AND KULKARNI S.P. RAIT College of Enggneerng, Nerul, Nav Mumba, Inda. *Correspondng
More informationSLAM Summer School 2006 Practical 2: SLAM using Monocular Vision
SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,
More informationOutline. Type of Machine Learning. Examples of Application. Unsupervised Learning
Outlne Artfcal Intellgence and ts applcatons Lecture 8 Unsupervsed Learnng Professor Danel Yeung danyeung@eee.org Dr. Patrck Chan patrckchan@eee.org South Chna Unversty of Technology, Chna Introducton
More informationTsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance
Tsnghua Unversty at TAC 2009: Summarzng Mult-documents by Informaton Dstance Chong Long, Mnle Huang, Xaoyan Zhu State Key Laboratory of Intellgent Technology and Systems, Tsnghua Natonal Laboratory for
More informationClassification of Face Images Based on Gender using Dimensionality Reduction Techniques and SVM
Classfcaton of Face Images Based on Gender usng Dmensonalty Reducton Technques and SVM Fahm Mannan 260 266 294 School of Computer Scence McGll Unversty Abstract Ths report presents gender classfcaton based
More informationFeature Reduction and Selection
Feature Reducton and Selecton Dr. Shuang LIANG School of Software Engneerng TongJ Unversty Fall, 2012 Today s Topcs Introducton Problems of Dmensonalty Feature Reducton Statstc methods Prncpal Components
More informationAn Optimal Algorithm for Prufer Codes *
J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,
More informationCorrelative features for the classification of textural images
Correlatve features for the classfcaton of textural mages M A Turkova 1 and A V Gadel 1, 1 Samara Natonal Research Unversty, Moskovskoe Shosse 34, Samara, Russa, 443086 Image Processng Systems Insttute
More informationMARS: Still an Alien Planet in Soft Computing?
MARS: Stll an Alen Planet n Soft Computng? Ajth Abraham and Dan Stenberg School of Computng and Informaton Technology Monash Unversty (Gppsland Campus), Churchll 3842, Australa Emal: ajth.abraham@nfotech.monash.edu.au
More informationSVM-based Learning for Multiple Model Estimation
SVM-based Learnng for Multple Model Estmaton Vladmr Cherkassky and Yunqan Ma Department of Electrcal and Computer Engneerng Unversty of Mnnesota Mnneapols, MN 55455 {cherkass,myq}@ece.umn.edu Abstract:
More informationCorner-Based Image Alignment using Pyramid Structure with Gradient Vector Similarity
Journal of Sgnal and Informaton Processng, 013, 4, 114-119 do:10.436/jsp.013.43b00 Publshed Onlne August 013 (http://www.scrp.org/journal/jsp) Corner-Based Image Algnment usng Pyramd Structure wth Gradent
More informationTN348: Openlab Module - Colocalization
TN348: Openlab Module - Colocalzaton Topc The Colocalzaton module provdes the faclty to vsualze and quantfy colocalzaton between pars of mages. The Colocalzaton wndow contans a prevew of the two mages
More informationSkew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach
Angle Estmaton and Correcton of Hand Wrtten, Textual and Large areas of Non-Textual Document Images: A Novel Approach D.R.Ramesh Babu Pyush M Kumat Mahesh D Dhannawat PES Insttute of Technology Research
More informationHierarchical clustering for gene expression data analysis
Herarchcal clusterng for gene expresson data analyss Gorgo Valentn e-mal: valentn@ds.unm.t Clusterng of Mcroarray Data. Clusterng of gene expresson profles (rows) => dscovery of co-regulated and functonally
More informationAn Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation
17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 An Iteratve Soluton Approach to Process Plant Layout usng Mxed
More informationAssignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009.
Farrukh Jabeen Algorthms 51 Assgnment #2 Due Date: June 15, 29. Assgnment # 2 Chapter 3 Dscrete Fourer Transforms Implement the FFT for the DFT. Descrbed n sectons 3.1 and 3.2. Delverables: 1. Concse descrpton
More informationDetection of an Object by using Principal Component Analysis
Detecton of an Object by usng Prncpal Component Analyss 1. G. Nagaven, 2. Dr. T. Sreenvasulu Reddy 1. M.Tech, Department of EEE, SVUCE, Trupath, Inda. 2. Assoc. Professor, Department of ECE, SVUCE, Trupath,
More informationSHAPE RECOGNITION METHOD BASED ON THE k-nearest NEIGHBOR RULE
SHAPE RECOGNITION METHOD BASED ON THE k-nearest NEIGHBOR RULE Dorna Purcaru Faculty of Automaton, Computers and Electroncs Unersty of Craoa 13 Al. I. Cuza Street, Craoa RO-1100 ROMANIA E-mal: dpurcaru@electroncs.uc.ro
More informationOptimizing Document Scoring for Query Retrieval
Optmzng Document Scorng for Query Retreval Brent Ellwen baellwe@cs.stanford.edu Abstract The goal of ths project was to automate the process of tunng a document query engne. Specfcally, I used machne learnng
More informationA Comparative Study of Fuzzy Classification Methods on Breast Cancer Data *
Comparatve Study of Fuzzy Classfcaton Methods on Breast Cancer Data * Rav. Jan, th. braham School of Computer & Informaton Scence, Unversty of South ustrala, Mawson Lakes Boulevard, Mawson Lakes, S 5095
More informationStudy of Data Stream Clustering Based on Bio-inspired Model
, pp.412-418 http://dx.do.org/10.14257/astl.2014.53.86 Study of Data Stream lusterng Based on Bo-nspred Model Yngme L, Mn L, Jngbo Shao, Gaoyang Wang ollege of omputer Scence and Informaton Engneerng,
More informationNovel Pattern-based Fingerprint Recognition Technique Using 2D Wavelet Decomposition
Mathematcal Methods for Informaton Scence and Economcs Novel Pattern-based Fngerprnt Recognton Technque Usng D Wavelet Decomposton TUDOR BARBU Insttute of Computer Scence of the Romanan Academy T. Codrescu,,
More informationFuzzy Modeling of the Complexity vs. Accuracy Trade-off in a Sequential Two-Stage Multi-Classifier System
Fuzzy Modelng of the Complexty vs. Accuracy Trade-off n a Sequental Two-Stage Mult-Classfer System MARK LAST 1 Department of Informaton Systems Engneerng Ben-Guron Unversty of the Negev Beer-Sheva 84105
More informationA Saturation Binary Neural Network for Crossbar Switching Problem
A Saturaton Bnary Neural Network for Crossbar Swtchng Problem Cu Zhang 1, L-Qng Zhao 2, and Rong-Long Wang 2 1 Department of Autocontrol, Laonng Insttute of Scence and Technology, Benx, Chna bxlkyzhangcu@163.com
More informationUsing an Adaptive Neuro-Fuzzy Inference System (AnFis) Algorithm for Automatic Diagnosis of Skin Cancer
Journal of Communcaton and Computer 8 (2011) 751-755 Usng an Adaptve Neuro-Fuzzy Inference System (AnFs) Algorthm for Automatc Dagnoss of Skn Cancer Suhal M. Odeh Department of Computer Informaton Systems,
More informationA soft computing approach for modeling of severity of faults in software systems
Internatonal Journal of Physcal Scences Vol. 5 (), pp. 074-085, February, 010 Avalable onlne at http://www.academcjournals.org/ijps ISSN 199-1950 010 Academc Journals Full Length Research Paper A soft
More informationA MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS
Proceedngs of the Wnter Smulaton Conference M E Kuhl, N M Steger, F B Armstrong, and J A Jones, eds A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Mark W Brantley Chun-Hung
More informationUnsupervised Learning and Clustering
Unsupervsed Learnng and Clusterng Why consder unlabeled samples?. Collectng and labelng large set of samples s costly Gettng recorded speech s free, labelng s tme consumng 2. Classfer could be desgned
More informationAn Evolvable Clustering Based Algorithm to Learn Distance Function for Supervised Environment
IJCSI Internatonal Journal of Computer Scence Issues, Vol. 7, Issue 5, September 2010 ISSN (Onlne): 1694-0814 www.ijcsi.org 374 An Evolvable Clusterng Based Algorthm to Learn Dstance Functon for Supervsed
More informationNUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS
ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana
More informationCONTROL SYSTEMS, ROBOTICS, AND AUTOMATION Vol. VI - System Identification Using Fuzzy Models - Robert Babuška
SYSTEM IDENTIFICATION USING FUZZY MODELS obert Delft Unversty of Technology, Faculty of Informaton Technology and Systems, The Netherlands Keywords: Nonlnear system dentfcaton, fuzzy logc, neuro-fuzzy
More informationImage Representation & Visualization Basic Imaging Algorithms Shape Representation and Analysis. outline
mage Vsualzaton mage Vsualzaton mage Representaton & Vsualzaton Basc magng Algorthms Shape Representaton and Analyss outlne mage Representaton & Vsualzaton Basc magng Algorthms Shape Representaton and
More informationLecture 4: Principal components
/3/6 Lecture 4: Prncpal components 3..6 Multvarate lnear regresson MLR s optmal for the estmaton data...but poor for handlng collnear data Covarance matrx s not nvertble (large condton number) Robustness
More informationUser Authentication Based On Behavioral Mouse Dynamics Biometrics
User Authentcaton Based On Behavoral Mouse Dynamcs Bometrcs Chee-Hyung Yoon Danel Donghyun Km Department of Computer Scence Department of Computer Scence Stanford Unversty Stanford Unversty Stanford, CA
More informationCourse Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms
Course Introducton Course Topcs Exams, abs, Proects A quc loo at a few algorthms 1 Advanced Data Structures and Algorthms Descrpton: We are gong to dscuss algorthm complexty analyss, algorthm desgn technques
More informationAn Application of Fuzzy c-means Clustering to FLC Design for Electric Ceramics Kiln
An Applcaton of cmeans Clusterng to FLC Desgn for lectrc Ceramcs Kln Watcharacha Wryasuttwong, Somphop Rodamporn lectrcal ngneerng Department, Faculty of ngneerng, Srnaharnwrot Unversty, Nahornnayo 6,
More informationRecognition of Handwritten Digits Using Optimized Adaptive Neuro-Fuzzy Inference Systems and Effective Features
Journal of Pattern Recognton and Intellgent Systems Aug. 2013, Vol. 1 Iss. 2, PP. 25-37 Recognton of Handwrtten Dgts Usng Optmzed Adaptve Neuro-Fuzzy Inference Systems and Effectve Features Amr Bahador
More informationAn Improved Neural Network Algorithm for Classifying the Transmission Line Faults
1 An Improved Neural Network Algorthm for Classfyng the Transmsson Lne Faults S. Vaslc, Student Member, IEEE, M. Kezunovc, Fellow, IEEE Abstract--Ths study ntroduces a new concept of artfcal ntellgence
More informationCollaboratively Regularized Nearest Points for Set Based Recognition
Academc Center for Computng and Meda Studes, Kyoto Unversty Collaboratvely Regularzed Nearest Ponts for Set Based Recognton Yang Wu, Mchhko Mnoh, Masayuk Mukunok Kyoto Unversty 9/1/013 BMVC 013 @ Brstol,
More informationA new paradigm of fuzzy control point in space curve
MATEMATIKA, 2016, Volume 32, Number 2, 153 159 c Penerbt UTM Press All rghts reserved A new paradgm of fuzzy control pont n space curve 1 Abd Fatah Wahab, 2 Mohd Sallehuddn Husan and 3 Mohammad Izat Emr
More informationLobachevsky State University of Nizhni Novgorod. Polyhedron. Quick Start Guide
Lobachevsky State Unversty of Nzhn Novgorod Polyhedron Quck Start Gude Nzhn Novgorod 2016 Contents Specfcaton of Polyhedron software... 3 Theoretcal background... 4 1. Interface of Polyhedron... 6 1.1.
More informationMeta-heuristics for Multidimensional Knapsack Problems
2012 4th Internatonal Conference on Computer Research and Development IPCSIT vol.39 (2012) (2012) IACSIT Press, Sngapore Meta-heurstcs for Multdmensonal Knapsack Problems Zhbao Man + Computer Scence Department,
More informationLecture 5: Multilayer Perceptrons
Lecture 5: Multlayer Perceptrons Roger Grosse 1 Introducton So far, we ve only talked about lnear models: lnear regresson and lnear bnary classfers. We noted that there are functons that can t be represented
More informationFUZZY LOGIC FUNDAMENTALS
3.fm Page 6 Monday, March 26, 200 0:8 AM C H A P T E R 3 FUZZY LOGIC FUNDAMENTALS 3. INTRODUCTION The past few years have wtnessed a rapd growth n the number and varety of applcatons of fuzzy logc (FL).
More informationDecision Strategies for Rating Objects in Knowledge-Shared Research Networks
Decson Strateges for Ratng Objects n Knowledge-Shared Research etwors ALEXADRA GRACHAROVA *, HAS-JOACHM ER **, HASSA OUR ELD ** OM SUUROE ***, HARR ARAKSE *** * nsttute of Control and System Research,
More informationExtraction of Fuzzy Rules from Trained Neural Network Using Evolutionary Algorithm *
Extracton of Fuzzy Rules from Traned Neural Network Usng Evolutonary Algorthm * Urszula Markowska-Kaczmar, Wojcech Trelak Wrocław Unversty of Technology, Poland kaczmar@c.pwr.wroc.pl, trelak@c.pwr.wroc.pl
More information