Kernelized Radial Basis Probabilistic Neural Network for Classification of River Water Quality
|
|
- Georgiana Rice
- 4 years ago
- Views:
Transcription
1 Kernelzed Radal Bass Probablstc eural etwor for Classfcaton of Rver Water Qualty Lm Eng A a, Zarta Zanuddn b a Insttute of Engneerng Mathematc Unverst Malaysa Perls, 0000 Kangar, Perls Tel : E-mal : ealm@unmap.edu.my b School of Mathematcal Scence Unverst Sans Malaysa, Mnden, Penang E-mal : zarta@cs.usm.my ABSTRACT Radal Bass Probablstc eural etwor (RBP) demonstrates broader and much more generalzed capabltes whch have been successfully appled to dfferent felds. In ths paper, the RBP s extended by calculatng the Eucldean dstance of each data pont based on a ernel-nduced dstance nstead of the conventonal sum-of squares dstance. The ernel functon s a generalzaton of the dstance metrc that measures the dstance between two data ponts as the data ponts are mapped nto a hgh dmensonal space. Through comparng the four constructed classfcaton models wth Kernelzed RBP, Radal Bass Functon networs, RBP and Bac-Propagaton networs as ntended, results showed that, model classfcaton on Rver water qualty of Langat rver n Selangor, Malaysa by Kernelzed RBP exhbted excellent performance n ths regard. Keywords Kernel functon, Radal Bass Probablstc eural etwor, Water Qualty.0 ITRODUCTIO In recent years, water qualty management and assessment model s a very popular and mportant research topc. The socalled management and assessment means to use the related water nformaton to evaluate the level of water qualty of a rver. However, the water qualty status of a rver s usually what the publc concerns about and s usually the most mportant ey for the healthy water supply to household (Gnors et al, 007). Therefore, f the water qualty of a rver can be predcted much earler, the hazardous causes to the publc can be greatly reduced. In the past lterature, some researchers use conventonal statstcal model for model constructon, for example, the Dscrmnant Analyss of (Altman, 968), the Probt of (Ohlson, 980). However, n recent years, some researchers found that the model constructed by data mnng technque has better predcton accuracy than that of the conventonal statstcal model; some of them are, for example, the Bac- Propagaton etwors model of (Odom, 990) and the decson tree classfcaton model of (Breman, 996). Another recent approach done by (Zarta et al, 004) usng bacpropagaton method n classfcaton of rver water qualty. The approach provded a good result n classfcaton, but, t taes a long tmes for tranng the networ. In ths paper, Radal Bass Probablstc eural etwors (RBP) whch s nown to have good generalzaton propertes and s traned faster than the bacpropagaton networ (Ganchev, 003) s adopted for the constructon of water qualty level early warnng classfcaton predcton model. Our proposed method, used ernel-functon to ncrease the separablty of data by worng n a hgh dmensonal space. Thus, the proposed method s characterzed by hgher classfcaton accuracy than the orgnal RBP. The ernelbased classfcaton n the feature space not only preserves the nherent structure of groups n the nput space, but also smplfes the assocated structure of the data (Muller, 00). Snce Grolam frst developed the ernel -means clusterng algorthm for unsupervsed classfcaton (Grolam, 00), several studes have demonstrated the superorty of ernel classfcaton algorthms over other approaches to classfcaton (Zhang, 00; Wu & Xe, 003). In ths paper, we evaluate the performance of our proposed method, ernelzed RBP, wth a comparson to three wellnown classfcaton methods: the Bac-Propagaton (BP), Radal Bass Functon etwor (RBF), and RBP; meanwhle, comparson and analyss on the classfcaton power s done to these three methods. These methods performance are compared usng rver water qualty data sets from Langat Rver of Selangor, Malaysa.
2 .0 RADIAL BASIS PROBABILISTIC EURAL ETWORKS Radal Bass Probablstc eural etwor (RBP) archtecture s based on based on Bayesan decson theory and nonparametrc technque to estmate Probablty Densty Functon (PDF). The form of PDF s a Gaussan dstrbuton. (Specht, 990) had proposed ths functon (Yeh, 998): If X = X, then f ( X ) = f ( X ) = 0 X X ( ) σ σ π = f ( X ) = exp () m m Snce RBP s applcable to general classfcaton problem, and assume that the egenvector to be classfed must belong to one of the nown classfcatons, then the absolute probablstc value of each classfcaton s not mportant and only relatve value needs to be consdered, hence, n equaton (), m m ( π ) σ Can be neglected and equaton () can be smplfed as X X f ( X ) = exp () σ = In equaton (), σ s the smoothng parameter of RBP. After networ tranng s completed, the predcton accuracy can be enhanced through the adustment of the smoothng parameter σ, that s, the larger the value, the smoother the approachng functon. If the smoothng parameter σ s napproprately selected, t wll lead to an excessve or nsuffcent neural unts n the networ desgn, and over fttng or napproprate fttng wll be the result n the functon approachng attempt; fnally, the predcton power wll be reduced. At ths moment, RBP wll depend fully on the nonclassfed sample whch s closest to the classfed sample to decde ts classfcaton. When smoothng parameter σ approaches nfnty, f ( X ) = At ths moment, RBP s close to blnd classfcaton. Usually, the researchers need to try dfferent σ n certan range to obtan one that can reach the optmum accuracy. (Specht, 99) had proposed a method that can be used to adust smoothng parameter σ,that s, assgn each nput neural unt a sngle σ ; durng the test stage, σ that have the optmum classfcaton result are taen through the fne adustment of each σ. RBP s a three-layer feedforward neural networ (as n Fgure ). The frst layer s the nput layer and the number of neural unt s the number of ndependent varable and receves the nput data; the second hdden layer n the mddle s Pattern Layer, whch stores each tranng data; the data sent out by Pattern Layer wll pass through the neural unt of the thrd layer Summaton Layer to correspond to each possble category, n ths layer, the calculaton of equaton (3) wll be performed. The fourth layer s Compettve Layer; the compettve transfer functon of ths layer wll pc up from the output of the last layer the maxmum value from these probabltes and generate the output value. If the output value s, t means t s the category you want; but f the output value s 0, t means t s other unwanted category. Let d = X X Be the square of the Eucldean dstance of two ponts X and X n the sample space, and equaton () can be re-wrtten as d f ( X ) = exp (3) σ = In equaton (3), when smoothng parameter σ approaches zero, Fgure : RBP archtecture
3 3.0 KERELIZED RADIAL BASIS PROBABILISTIC EURAL ETWORKS 3. Kernel-Based Approach Gven an unlabeled data set X { x x } =, K, n the d- n dmensonal space R d d, let Φ : R ᆴ H be a non-lnear mappng functon from ths nput space to a hgh dmensonal feature space H. By applyng the non-lnear mappng functon Φ, the dot product x x n the nput space. The ey noton n ernelbased learnng s that the mappng functon Φ need not be explctly specfed. The dot product Φ( x ) Φ ( x ) n the hgh dmensonal feature space can be calculated through the ernel functon K(x, x ) n the nput space R d (Scholopf & Smola, 00). K ( x, x ) = Φ( x ) Φ ( x ) (4) Three commonly used ernel functons (Scholopf & Smola, 00) are the polynomal ernel functon, K( x, x ) = ( x x + c) d (5) where c ᄈ 0, d ᅫ ; the Gaussan ernel functon, x x K( x, x ) = exp σ where σ > 0; and the sgmodal ernel functon, where κ > 0 and γ < Formulaton (6) K( x, x ) = tanh( κ ( x, x ) + γ ) (7) Gven a data pont x ΣR d ( n) and a non-lnear mappng d Φ : R ᆴ H, the Probablty Densty Functon at data pont x s defned as Φ( x ) Φ( x ) f ( x) = exp (8) σ = where Φ( x ) Φ( x ) s the square of dstance between Φ( x ) and Φ( x ). Thus a hgher value of f ( x) ndcates that x has more data ponts x near to t n the feature space. The dstance n feature space s calculated through the ernel n the nput space as follows: ( ) ( ) Φ( x ) Φ ( x ) = Φ( x ) Φ( x ) Φ( x ) Φ( x ) = Φ( x ) Φ ( x ) Φ( x ) Φ ( x ) + Φ( x ) Φ ( x ) = K ( x, x ) K ( x, x ) + K( x, x ) (9) Therefore, Equaton (8) can be rewrtten as K ( x, x ) K ( x, x ) + K ( x, x ) f ( x) = exp σ = (0) The rest of the computaton procedure s smlar to that of the RBP method. 4.0 COSTRUCTIO OF CLASSIFICATIO PREDICTIO MODEL AD ROC CURVE AALYSIS A sequence of 33 rows of data conssts of Bochemcal Oxygen Demand (BOD), Chemcal Oxygen Demand (COD), Ammona-trate level (A), Suspended Substances level (SS), Dssolved Oxygen (DO), and Acdty (ph) that consttutes to the dataset. The data to ncluded n the database s collected by Department of Envronment of Malaysa and contans the records of Langat Rver water qualty level whch also reported n (Zarta et al, 004). All of data sets of the sx parameters have been Z-transformed for classfcaton purposes and the Gaussan ernel was used n tranng of Kernelzed RBP. The Z-transform s performed as a preprocessng step to mnmze the gap of mnmum and maxmum of data values. Thus, we dvded 33 rows of data nto two separate parts, 00 rows for tranng and 3 rows remanng for testng. So, the rato of tran to test s approxmately :. All tranng s done n a destop PC of Intel duo core GHz wth Gb RAM memory. In the Kernelzed RBP, MATLAB s used to self-wrte the program for the model constructon and we fxed value to the Smoothng Parameter σ to 0. accordng to (Sansone & Vento, 00). In the test phase, each classfer has been tested wth data sets mentoned above. Performance of dfferent classfers n the test phase can be observed n Table. The proposed classfer, whch referred as Kernelzed RBP, acheves the best result amongst all other methods. It can be concluded from Table that the performance of Kernelzed RBP s better than RBP, RBF and BP. In term of tme, Kernelzed
4 RBP s about 89 tmes faster than RBF and 99 tmes faster than GD. The classfcaton accuracy obtaned usng Kernelzed RBP s 6% hgher than RBP. It s also obvous that Kernelzed RBP consumed less tme n comparson to RBP, RBF and BP. RBP outperform RBF n classfcaton accuracy due to ts networ archtecture that mplemented Bayesan decson and nonparametrc technque whch mae t more applcable to general classfcaton problem, ths brng RBF has the worst performance among all models. Although the classfcaton accuracy of the well nown classfcaton method, BP, s hgh, but as we now, hgh processng tme of BP maes t undesrable for many on-lne recognton applcatons Kernelzed RBP RBP RBF GD Reference Lne Method AUC Kernelzed RBP RBP RBF GD 0.69 ROC curve The true postve rate n Fgure means the percentage of the number of class wth predcton result of 0 to the number of class wth real value of 0; false postve rate means the percentage of the number of class wth predcton result of to the number of class wth real value of. In ths paper, the mutual verfcaton of the data among the models s performed. The Kernelzed RBP s verfed as the best performed model and there are 3 rows of data n the test results. The result s drawn as Recever Operator Characterstc (ROC) curve as n Fgure. In the fgure, the farther the ROC curve above the reference lne, the larger the area under the ROC curve (AUC), whch also means the hgher the classfcaton predcton power of ths model (Bradley, 997). Table shows the mutual verfcaton results of the rver water qualty. Fgure shows that the area under the ROC curve of the Kernelzed RBP model of the rver water qualty data are larger than the other three models; from the observaton of Table, t seems that the specfcty of Kernelzed RBP model s less than other three models, but the AUC value are hgher than those of BP, RBP and RBF models; therefore, t can be concluded that Kernelzed RBP model has very good classfcaton predcton capablty. Table : Classfcaton Predcton accuracy of Langat rver water qualty. Method Classfcaton Accuracy (%) CPU Tme (s) Epoch Kernelzed RBP RBP RBF GD true postve rate false postve rate Fgure : The Mutual verfed ROC curve of test data of Langat rver water qualty 5.0 COCLUSIO In ths paper, a new RBP whch conssts of ernel-based approach has been proposed. ot le the conventonal RBP that only based on Eucldean dstance computaton, the Kernelzed RBP appled the ernel-nduced dstance computaton whch can mplctly map the nput data to a hgh dmensonal space n whch data classfcaton s easer. The networ s appled to rver water classfcaton problem and showed an ncreased n classfcaton accuracy n comparson wth other well-nown classfers. The metrcs for consderng performance of a classfer n ths wor were the classfcaton accuracy and processng tme of the classfer. In future wor, we plan to mprove the classfcaton rate by employng dfferent ernel functon such as wavelet functon. Table : Mutual Verfcaton wth AUC. REFERECES
5 Altman, E. I. (968). Fnancal ratos, dscrmnate analyss and the predcton of corporate banruptcy. Journal of Fnance, 3(4), Bradley, A. P. (997). The use of the area under ROC curve n the evaluaton of machne learnng algorthms. Pattern Recognton, 30 (7), Breman, L. (996). Baggng predctors. Machne Learnng, 4(), Zarta, Z., Sharfuddn, M. Z., Lm, E. A. & Hafzan, J. (004). Applcaton of neural networ n rver water qualty classfcaton. Ecologcal and Envronmental Modelng (ECOMOD 004), (), 4-3. Zhang, R. & Rudncy, A. I. (00). A large scale clusterng scheme for ernel -means. The 6 th Internatonal Conference on Pattern Recognton, Ganchev, T., Tasouls, D. K. & Vrahats, M.. (003). Locally Recurrent Probablstc eural etwor for Text- Independent Speaer Verfcaton. In Proc. Of the EuroSpeech, Vol 3, Gnors, Y. P., Amaral, A.L., colau, A., Coelho, M.A.Z., and Ferrera, E.C. (007). Development of an mage analyss procedure for dentfyng protozoa and metazoan typcal of actvated sludge system. Water Research, 4, Grolam, M. (00). Mercer ernel-based clusterng n feature space. IEEE Trans. eural etwors, 3(3), Muller, K. R. (00). An ntroducton to ernel-based learnng algorthms. IEEE Trans. eural etwors, (), 8-0. Ohlson, J. A. (980). Fnancal ratos and the probablstc predcton of banruptcy. Journal of Accountng Research, 8(), Odom, M. D. & Sharda, R. (990). A neural networ model for banruptcy predcton. IEEE IS IJC,, Specht, D. F. (990). Probablstc neural networ and the polynomal adalne as complementary technques for classfcaton. IEEE Trans. eural etwors, (), -. Scholopf, B. & Smola, A. J. (00). Learnng wth Kernels. Cambrdge: MIT Press. Sansone, C. & Vento, M. (00). A classfcaton relablty drve reect rule for mult-expert system. Internatonal Journal of Pattern Recognton and Artfcal Intellgent, 5(6), -9. Wu, Z. D. & Xe, W. X. (003). Fuzzy c-means clusterng algorthm based on ernel method. The Ffth Internatonal Conference on Computatonal Intellgent and Multmeda Applcatons, -6. Yeh, Y. C. (998). Applcaton of neural networ (Scholars Boo, Tawan Tape, 998). Scholars Boos. Tawan.
Forecasting 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 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 informationThe 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 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 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 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 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 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 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 informationL10: Linear discriminants analysis
L0: Lnear dscrmnants analyss Lnear dscrmnant analyss, two classes Lnear dscrmnant analyss, C classes LDA vs. PCA Lmtatons of LDA Varants of LDA Other dmensonalty reducton methods CSCE 666 Pattern Analyss
More informationModelling of Web Domain Visits by Radial Basis Function Neural Networks and Support Vector Machine Regression
Modellng of Web Doman Vsts by Radal Bass Functon Neural Networks and Support Vector Machne Regresson Vladmír Olej, Jana Flpová Insttute of System Engneerng and Informatcs Faculty of Economcs and Admnstraton,
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 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 informationStatistical Methods to Develop Rating Models
Statstcal Methods to Develop Ratng Models [Evelyn Hayden and Danel Porath, Österrechsche Natonalbank and Unversty of Appled Scences at Manz] Source: The Basel II Rsk Parameters Estmaton, Valdaton, and
More informationDesign of Output Codes for Fast Covering Learning using Basic Decomposition Techniques
Journal of Computer Scence (7): 565-57, 6 ISSN 59-66 6 Scence Publcatons Desgn of Output Codes for Fast Coverng Learnng usng Basc Decomposton Technques Aruna Twar and Narendra S. Chaudhar, Faculty of Computer
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 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 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 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 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 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 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 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 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 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 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 informationv a 1 b 1 i, a 2 b 2 i,..., a n b n i.
SECTION 8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS 455 8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS All the vector spaces we have studed thus far n the text are real vector spaces snce the scalars are
More information8.5 UNITARY AND HERMITIAN MATRICES. The conjugate transpose of a complex matrix A, denoted by A*, is given by
6 CHAPTER 8 COMPLEX VECTOR SPACES 5. Fnd the kernel of the lnear transformaton gven n Exercse 5. In Exercses 55 and 56, fnd the mage of v, for the ndcated composton, where and are gven by the followng
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 informationLogistic Regression. Lecture 4: More classifiers and classes. Logistic regression. Adaboost. Optimization. Multiple class classification
Lecture 4: More classfers and classes C4B Machne Learnng Hlary 20 A. Zsserman Logstc regresson Loss functons revsted Adaboost Loss functons revsted Optmzaton Multple class classfcaton Logstc Regresson
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 informationChapter 6. Classification and Prediction
Chapter 6. Classfcaton and Predcton What s classfcaton? What s Lazy learners (or learnng from predcton? your neghbors) Issues regardng classfcaton and Frequent-pattern-based predcton classfcaton Classfcaton
More informationCalculation of Sampling Weights
Perre Foy Statstcs Canada 4 Calculaton of Samplng Weghts 4.1 OVERVIEW The basc sample desgn used n TIMSS Populatons 1 and 2 was a two-stage stratfed cluster desgn. 1 The frst stage conssted of a sample
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 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 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 informationBANKRUPTCY PREDICTION BY USING SUPPORT VECTOR MACHINES AND GENETIC ALGORITHMS
BANKRUPCY PREDICION BY USING SUPPOR VECOR MACHINES AND GENEIC ALGORIHMS SALEHI Mahd Ferdows Unversty of Mashhad, Iran ROSAMI Neda Islamc Azad Unversty Scence and Research Khorasan-e-Razav Branch Abstract:
More informationA study on the ability of Support Vector Regression and Neural Networks to Forecast Basic Time Series Patterns
A study on the ablty of Support Vector Regresson and Neural Networks to Forecast Basc Tme Seres Patterns Sven F. Crone, Jose Guajardo 2, and Rchard Weber 2 Lancaster Unversty, Department of Management
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 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 an Improved BP Neural Network Model in Enterprise Network Security Forecasting
161 A publcaton of VOL. 46, 15 CHEMICAL ENGINEERING TRANSACTIONS Guest Edtors: Peyu Ren, Yancang L, Hupng Song Copyrght 15, AIDIC Servz S.r.l., ISBN 978-88-9568-37-; ISSN 83-916 The Italan Assocaton of
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 informationSupport Vector Machine Model for Currency Crisis Discrimination. Arindam Chaudhuri 1. Abstract
Support Vector Machne Model for Currency Crss Dscrmnaton Arndam Chaudhur Abstract Support Vector Machne (SVM) s powerful classfcaton technque based on the dea of structural rsk mnmzaton. Use of kernel
More informationProduct Quality and Safety Incident Information Tracking Based on Web
Product Qualty and Safety Incdent Informaton Trackng Based on Web News 1 Yuexang Yang, 2 Correspondng Author Yyang Wang, 2 Shan Yu, 2 Jng Q, 1 Hual Ca 1 Chna Natonal Insttute of Standardzaton, Beng 100088,
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 informationData Mining from the Information Systems: Performance Indicators at Masaryk University in Brno
Data Mnng from the Informaton Systems: Performance Indcators at Masaryk Unversty n Brno Mkuláš Bek EUA Workshop Strasbourg, 1-2 December 2006 1 Locaton of Brno Brno EUA Workshop Strasbourg, 1-2 December
More informationNumber of Levels Cumulative Annual operating Income per year construction costs costs ($) ($) ($) 1 600,000 35,000 100,000 2 2,200,000 60,000 350,000
Problem Set 5 Solutons 1 MIT s consderng buldng a new car park near Kendall Square. o unversty funds are avalable (overhead rates are under pressure and the new faclty would have to pay for tself from
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 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 informationAn Efficient and Simplified Model for Forecasting using SRM
HAFIZ MUHAMMAD SHAHZAD ASIF*, MUHAMMAD FAISAL HAYAT*, AND TAUQIR AHMAD* RECEIVED ON 15.04.013 ACCEPTED ON 09.01.014 ABSTRACT Learnng form contnuous fnancal systems play a vtal role n enterprse operatons.
More information320 The Internatonal Arab Journal of Informaton Technology, Vol. 5, No. 3, July 2008 Comparsons Between Data Clusterng Algorthms Osama Abu Abbas Computer Scence Department, Yarmouk Unversty, Jordan Abstract:
More informationCausal, Explanatory Forecasting. Analysis. Regression Analysis. Simple Linear Regression. Which is Independent? Forecasting
Causal, Explanatory Forecastng Assumes cause-and-effect relatonshp between system nputs and ts output Forecastng wth Regresson Analyss Rchard S. Barr Inputs System Cause + Effect Relatonshp The job of
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 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 informationA novel Method for Data Mining and Classification based on
A novel Method for Data Mnng and Classfcaton based on Ensemble Learnng 1 1, Frst Author Nejang Normal Unversty;Schuan Nejang 641112,Chna, E-mal: lhan-gege@126.com Abstract Data mnng has been attached great
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 informationImplementation of Deutsch's Algorithm Using Mathcad
Implementaton of Deutsch's Algorthm Usng Mathcad Frank Roux The followng s a Mathcad mplementaton of Davd Deutsch's quantum computer prototype as presented on pages - n "Machnes, Logc and Quantum Physcs"
More informationTopic Identification based on Bayesian Belief Networks in the context of an Air Traffic Control Task
Procesamento del Lenguaje Natural, núm. 35 (2005), pp. 327-332 recbdo 06-05-2005; aceptado 01-06-2005 Topc Identfcaton based on Bayesan Belef Networs n the context of an Ar Traffc Control Tas F. Fernández,
More informationMining Multiple Large Data Sources
The Internatonal Arab Journal of Informaton Technology, Vol. 7, No. 3, July 2 24 Mnng Multple Large Data Sources Anmesh Adhkar, Pralhad Ramachandrarao 2, Bhanu Prasad 3, and Jhml Adhkar 4 Department of
More informationBayesian Network Based Causal Relationship Identification and Funding Success Prediction in P2P Lending
Proceedngs of 2012 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 25 (2012) (2012) IACSIT Press, Sngapore Bayesan Network Based Causal Relatonshp Identfcaton and Fundng Success
More informationConversion between the vector and raster data structures using Fuzzy Geographical Entities
Converson between the vector and raster data structures usng Fuzzy Geographcal Enttes Cdála Fonte Department of Mathematcs Faculty of Scences and Technology Unversty of Combra, Apartado 38, 3 454 Combra,
More informationFast Fuzzy Clustering of Web Page Collections
Fast Fuzzy Clusterng of Web Page Collectons Chrstan Borgelt and Andreas Nürnberger Dept. of Knowledge Processng and Language Engneerng Otto-von-Guercke-Unversty of Magdeburg Unverstätsplatz, D-396 Magdeburg,
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 informationPrediction of Stock Market Index Movement by Ten Data Mining Techniques
Vol. 3, o. Modern Appled Scence Predcton of Stoc Maret Index Movement by en Data Mnng echnques Phchhang Ou (Correspondng author) School of Busness, Unversty of Shangha for Scence and echnology Rm 0, Internatonal
More informationThe Network flow Motoring System based on Particle Swarm Optimized
The Network flow Motorng System based on Partcle Swarm Optmzed Neural Network Adult Educaton College, Hebe Unversty of Archtecture, Zhangjakou Hebe 075000, Chna Abstract The compatblty of the commercal
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 informationTesting CAB-IDS through Mutations: on the Identification of Network Scans
Testng CAB-IDS through Mutatons: on the Identfcaton of Network Scans Emlo Corchado, Álvaro Herrero, José Manuel Sáz Department of Cvl Engneerng, Unversty of Burgos, Span {escorchado, ahcoso, msaz}@ubu.es
More informationOut-of-Sample Extensions for LLE, Isomap, MDS, Eigenmaps, and Spectral Clustering
Out-of-Sample Extensons for LLE, Isomap, MDS, Egenmaps, and Spectral Clusterng Yoshua Bengo, Jean-Franços Paement, Pascal Vncent Olver Delalleau, Ncolas Le Roux and Mare Oumet Département d Informatque
More informationActive Learning for Interactive Visualization
Actve Learnng for Interactve Vsualzaton Tomoharu Iwata Nel Houlsby Zoubn Ghahraman Unversty of Cambrdge Unversty of Cambrdge Unversty of Cambrdge Abstract Many automatc vsualzaton methods have been. However,
More informationInstitute of Informatics, Faculty of Business and Management, Brno University of Technology,Czech Republic
Lagrange Multplers as Quanttatve Indcators n Economcs Ivan Mezník Insttute of Informatcs, Faculty of Busness and Management, Brno Unversty of TechnologCzech Republc Abstract The quanttatve role of Lagrange
More informationMachine Learning and Software Quality Prediction: As an Expert System
I.J. Informaton Engneerng and Electronc Busness, 2014, 2, 9-27 Publshed Onlne Aprl 2014 n MECS (http://www.mecs-press.org/) DOI: 10.5815/jeeb.2014.02.02 Machne Learnng and Software Qualty Predcton: As
More informationLecture 5,6 Linear Methods for Classification. Summary
Lecture 5,6 Lnear Methods for Classfcaton Rce ELEC 697 Farnaz Koushanfar Fall 2006 Summary Bayes Classfers Lnear Classfers Lnear regresson of an ndcator matrx Lnear dscrmnant analyss (LDA) Logstc regresson
More informationCredit Limit Optimization (CLO) for Credit Cards
Credt Lmt Optmzaton (CLO) for Credt Cards Vay S. Desa CSCC IX, Ednburgh September 8, 2005 Copyrght 2003, SAS Insttute Inc. All rghts reserved. SAS Propretary Agenda Background Tradtonal approaches to credt
More information21 Vectors: The Cross Product & Torque
21 Vectors: The Cross Product & Torque Do not use our left hand when applng ether the rght-hand rule for the cross product of two vectors dscussed n ths chapter or the rght-hand rule for somethng curl
More informationUsing Content-Based Filtering for Recommendation 1
Usng Content-Based Flterng for Recommendaton 1 Robn van Meteren 1 and Maarten van Someren 2 1 NetlnQ Group, Gerard Brandtstraat 26-28, 1054 JK, Amsterdam, The Netherlands, robn@netlnq.nl 2 Unversty of
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 informationMining Feature Importance: Applying Evolutionary Algorithms within a Web-based Educational System
Mnng Feature Importance: Applyng Evolutonary Algorthms wthn a Web-based Educatonal System Behrouz MINAEI-BIDGOLI 1, and Gerd KORTEMEYER 2, and Wllam F. PUNCH 1 1 Genetc Algorthms Research and Applcatons
More informationA FEATURE SELECTION AGENT-BASED IDS
A FEATURE SELECTION AGENT-BASED IDS Emlo Corchado, Álvaro Herrero and José Manuel Sáz Department of Cvl Engneerng, Unversty of Burgos C/Francsco de Vtora s/n., 09006, Burgos, Span Phone: +34 947259395,
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 informationEnterprise Master Patient Index
Enterprse Master Patent Index Healthcare data are captured n many dfferent settngs such as hosptals, clncs, labs, and physcan offces. Accordng to a report by the CDC, patents n the Unted States made an
More informationHow To Know The Components Of Mean Squared Error Of Herarchcal Estmator S
S C H E D A E I N F O R M A T I C A E VOLUME 0 0 On Mean Squared Error of Herarchcal Estmator Stans law Brodowsk Faculty of Physcs, Astronomy, and Appled Computer Scence, Jagellonan Unversty, Reymonta
More informationA Comparative Study of Data Clustering Techniques
A COMPARATIVE STUDY OF DATA CLUSTERING TECHNIQUES A Comparatve Study of Data Clusterng Technques Khaled Hammouda Prof. Fakhreddne Karray Unversty of Waterloo, Ontaro, Canada Abstract Data clusterng s a
More informationTHE METHOD OF LEAST SQUARES THE METHOD OF LEAST SQUARES
The goal: to measure (determne) an unknown quantty x (the value of a RV X) Realsaton: n results: y 1, y 2,..., y j,..., y n, (the measured values of Y 1, Y 2,..., Y j,..., Y n ) every result s encumbered
More informationHow To Create An Emoton Recognzer
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,
More informationProceedings of the Annual Meeting of the American Statistical Association, August 5-9, 2001
Proceedngs of the Annual Meetng of the Amercan Statstcal Assocaton, August 5-9, 2001 LIST-ASSISTED SAMPLING: THE EFFECT OF TELEPHONE SYSTEM CHANGES ON DESIGN 1 Clyde Tucker, Bureau of Labor Statstcs James
More informationForecasting and Modelling Electricity Demand Using Anfis Predictor
Journal of Mathematcs and Statstcs 7 (4): 75-8, 0 ISSN 549-3644 0 Scence Publcatons Forecastng and Modellng Electrcty Demand Usng Anfs Predctor M. Mordjaou and B. Boudjema Department of Electrcal Engneerng,
More informationA cooperative connectionist IDS model to identify independent anomalous SNMP situations
A cooperatve connectonst IDS model to dentfy ndependent anomalous SNMP stuatons Álvaro Herrero, Emlo Corchado, José Manuel Sáz Department of Cvl Engneerng, Unversty of Burgos, Span escorchado@ubu.es Abstract
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 informationBag-of-Words models. Lecture 9. Slides from: S. Lazebnik, A. Torralba, L. Fei-Fei, D. Lowe, C. Szurka
Bag-of-Words models Lecture 9 Sldes from: S. Lazebnk, A. Torralba, L. Fe-Fe, D. Lowe, C. Szurka Bag-of-features models Overvew: Bag-of-features models Orgns and motvaton Image representaton Dscrmnatve
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 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 informationTraffic-light a stress test for life insurance provisions
MEMORANDUM Date 006-09-7 Authors Bengt von Bahr, Göran Ronge Traffc-lght a stress test for lfe nsurance provsons Fnansnspetonen P.O. Box 6750 SE-113 85 Stocholm [Sveavägen 167] Tel +46 8 787 80 00 Fax
More informationMATHEMATICAL ENGINEERING TECHNICAL REPORTS. Sequential Optimizing Investing Strategy with Neural Networks
MATHEMATICAL ENGINEERING TECHNICAL REPORTS Sequental Optmzng Investng Strategy wth Neural Networks Ryo ADACHI and Akmch TAKEMURA METR 2010 03 February 2010 DEPARTMENT OF MATHEMATICAL INFORMATICS GRADUATE
More informationMean Molecular Weight
Mean Molecular Weght The thermodynamc relatons between P, ρ, and T, as well as the calculaton of stellar opacty requres knowledge of the system s mean molecular weght defned as the mass per unt mole of
More informationDecision Tree Model for Count Data
Proceedngs of the World Congress on Engneerng 2012 Vol I Decson Tree Model for Count Data Yap Bee Wah, Norashkn Nasaruddn, Wong Shaw Voon and Mohamad Alas Lazm Abstract The Posson Regresson and Negatve
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 informationPower-of-Two Policies for Single- Warehouse Multi-Retailer Inventory Systems with Order Frequency Discounts
Power-of-wo Polces for Sngle- Warehouse Mult-Retaler Inventory Systems wth Order Frequency Dscounts José A. Ventura Pennsylvana State Unversty (USA) Yale. Herer echnon Israel Insttute of echnology (Israel)
More informationRealistic Image Synthesis
Realstc Image Synthess - Combned Samplng and Path Tracng - Phlpp Slusallek Karol Myszkowsk Vncent Pegoraro Overvew: Today Combned Samplng (Multple Importance Samplng) Renderng and Measurng Equaton Random
More information