Generalised density function estimation using moments and the characteristic function
|
|
- Horace Neal
- 5 years ago
- Views:
Transcription
1 Generalised density function estimation using moments and the characteristic function Gerhard Esterhuizen Thesis presented in partial fulfilment of the requirements for the degree of Master of Science in Electronic Engineering at the University of Stellenbosch Supervisor: Prof. J.A. du Preez April 2003
2 Declaration I, the undersigned, hereby declare that the work contained in this thesis is my own original work and that I have not previously in its entirety or in part submitted it at any university for a degree. Signature: March 2003 i
3 Abstract Probability density functions (PDFs) and cumulative distribution functions (CDFs) playa central role in statistical pattern recognition and verification systems. They allow observations that do not occur according to deterministic rules to be quantified and modelled. An example of such observations would be the voice patterns of a person that is used as input to a biometric security device. In order to model such non-deterministic observations, a density function estimator is employed to estimate a PDF or CDF from sample data. Although numerous density function estimation techniques exist, all the techniques can be classified into one of two groups, parametric and non-parametric, each with its own characteristic advantages and disadvantages. In this research, we introduce a novel approach to density function estimation that attempts to combine some of the advantages of both the parametric and non-parametric estimators. This is done by considering density estimation using an abstract approach in which the density function is modelled entirely in terms of its moments or characteristic function. New density function estimation techniques are first developed in theory, after which a number of practical density function estimators are presented. Experiments are performed in which the performance of the new estimators are compared to two established estimators, namely the Parzen estimator and the Gaussian mixture model (GMM). The comparison is performed in terms of the accuracy, computational requirements and ease of use of the estimators and it is found that the new estimators does combine some of the advantages of the established estimators without the corresponding disadvantages. II
4 Opsomming Waarskynlikheids digtheidsfunksies (WDFs) en Kumulatiewe distribusiefunksies (KDFs) speel 'n sentrale rol in statistiese patroonherkenning en verifikasie stelsels. Hulle maak dit moontlik om nie-deterministiese observasies te kwantifiseer en te modelleer. Die stempatrone van 'n spreker wat as intree tot 'n biometriese sekuriteits stelsel gegee word, is 'n voorbeeld van so 'n observasie. Ten einde sulke observasies te modelleer, word 'n digtheidsfunksie afskatter gebruik om die WDF of KDF vanaf data monsters af te skat. Alhoewel daar talryke digtheidsfunksie afskatters bestaan, kan almal in een van twee katagoriee geplaas word, parametries en nie-parametries, elk met hul eie kenmerkende voordele en nadele. Hierdie werk Ie 'n nuwe benadering tot digtheidsfunksie afskatting voor wat die voordele van beide die parametriese sowel as die nie-parametriese tegnieke probeer kombineer. Dit word gedoen deur digtheidsfunksie afskatting vanuit 'n abstrakte oogpunt te benader waar die digtheidsfunksie uitsluitlik in terme van sy momente en karakteristieke funksie gemodelleer word. Nuwe metodes word eers in teorie ondersoek en ontwikkel waarna praktiese tegnieke voorgele word. Hierdie afskatters het die vermoe om 'n wye verskeidenheid digtheidsfunksies af te skat en is nie net ontwerp om slegs sekere families van digtheidsfunksies optimaal voor te stel nie. Eksperimente is uitgevoer wat die werkverrigting van die nuwe tegnieke met twee gevestigde tegnieke, naamlik die Parzen afskatter en die Gaussiese mengsel model (GMM), te vergelyk. Die werkverrigting word gemeet in terme van akkuraatheid, vereiste numeriese verwerkingsvermoe en die gemak van gebruik. Daar word bevind dat die nuwe afskatters wei voordele van die gevestigde afskatters kombineer sonder die gepaardgaande nadele. iii
5 Acknowledgements I would like to thank the following people: Prof. J.A. Du Preez, my supervisor, for his patient guidance. Zelda Weitz for her unfailing love and encouragement. My parents and family for their education and support. My grandmother and Mr and Mrs Weitz for providing a home away from home. Dr Dave Weber and the DSP Lab for providing a creative learning environment. Chari Botha for his technical advice. Pieter Nel and Koos Hugo for False Bay sailing. iv
6 Contents 1 Introduction Motivation and topicality Density function estimation Our research Background Random variables Pattern classification Hypothesis tests Existing techniques Non-parametric estimators Parametric estimators Other techniques Requirements for a new estimator Objectives Contributions Overview of the document 19 2 Novel probability density function estimators Introduction Motivation Definitions and Background Moments Characteristic function Fourier series Estimators based on moments Motivation. 27 v
7 2.4.2 A PDF in terms of moments The anti-derivative Numerical integration techniques Fourier series approximation Estimating moments from sample data Estimators based on the characteristic function Motivation A PDF in terms of a characteristic function Fourier series Limiting case where N x Conclusions Comparison between characteristic function and moments techniques Comparison with the Parzen estimator and Gaussian mixture model (GMM) 59 3 Novel cumulative distribution function estimators 3.1 Introduction 3.2 Motivation. 3.3 Definitions and background 3.4 Estimators based on moments A CDF in terms of moments Numerical integration techniques Fourier series approximation Estimators based on the characteristic function A CDF in terms of a characteristic function Fourier series 3.6 Conclusions... 4 Experimental results 4.1 Introduction Experimental setup Input data Estimation error measure PDF and CDF estimate using moments PDF and CDF estimate using characteristic function VI
8 4.2.5 Parzen estimator Gaussian Mixture Model Mean estimation error PDF estimators CDF estimators Computational requirements PDF Estimators CDF Estimators Training requirements Application to speaker verification Conclusions Conclusions and recommendations Conclusions Recommendations A A review of the Fourier transform 120 B Summary of algorithms 123 vii
9 List of Tables 4.1 PDF estimators: mllllmum mean estimation errors, minimum combined mean estimation errors and corresponding combined standard deviations (in brackets) from 100 samples (loox Kullback-Leibler divergence is indicated) PDF estimators: minimum mean estimation errors, minimum combined mean estimation errors and corresponding combined standard deviations (in brackets) from 1000 samples (loox Kullback-Leibler divergence is indicated) PDF estimators: the effect of selecting a sub-optimal working point CDF estimators: minimum mean estimation errors, minimum combined mean estimation errors and corresponding combined standard deviation (in brackets) from 100 samples (lox integral absolute difference is indicated) CDF estimators: minimum mean estimation errors, minimum combined mean estimation errors and corresponding combined standard deviation (in brackets) from 1000 samples (lox integral absolute difference is indicated) Gradient (xl00) characterising the relationship between the computation time and the parameter value Comparison between GMM and CF technique in a speaker verification application Feature matrix for all the estimators. 113 A.l Selected Fourier transform properties.. B.l PDF estimate from moments using Fourier series. B.2 PDF estimate from samples using Fourier series.. B.3 CDF estimate from moments using Fourier series. B.4 CDF estimate from samples using Fourier series Vlll
10 List of Figures 2.1 Successive Taylor series approximations to characteristic function of Gaussian PDF using 5, 14, and 32 terms The smoothing effect of a Hamming window on the characteristic function Leakage introduced by rectangular windowing of the characteristic function Discretisation of the characteristic function Reconstructing a PDF directly from data samples, using a triangular windowing function Relationship between f(x), f'(x)and f"(x) Mean estimation error: PDF estimate using GMM from 100 samples Mean estimation error: PDF estimate using Parzen estimator from 100 samples Mean estimation error: PDF estimate using characteristic function from 100 samples Mean estimation error: PDF estimate using moments from 100 samples Typical PDF estimates Examples of over-fitted PDF estimates (obtained from 100 samples) Mean estimation error: CDF estimate using GMM from 100 samples Mean estimation error: CDF estimate using Parzen estimator from 100 samples Mean estimation error: CDF estimate using characteristic function from 100 samples Mean estimation error: CDF estimate using moments from 100 samples Typical CDF estimates Comparison of normalised execution times of PDF estimators: Pentium III 700 MHz 104 ix
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
Oplos van kwadratiese vergelykings: die vind van die vergelyking *
OpenStax-CNX module: m39143 1 Oplos van kwadratiese vergelykings: die vind van die vergelyking * Free High School Science Texts Project Based on Solving quadratic equations: nding the equation by Free
More information3. (d) None of these / Geen van hierdie
SECTION A (24 marks) / AFDELING A (24 punte) The questions in Section A must be completed on SIDE 2 of the optical reader form in SOFT PENCIL. First circle your answers on this paper and then transfer
More informationJUNE 2005 TYD/TIME: 90 min PUNTE / MARKS: 50 VAN/SURNAME: VOORNAME/FIRST NAMES: STUDENTENOMMER/STUDENT NUMBER:
UNIVERSITEIT VAN PRETORIA / UNIVERSITY OF PRETORIA DEPT WISKUNDE EN TOEGEPASTE WISKUNDE DEPT OF MATHEMATICS AND APPLIED MATHEMATICS WTW 63 - NUMERIESE METHODE / NUMERICAL METHODS EKSAMEN / EXAMINATION
More informationEXAMINATION / EKSAMEN 17 JUNE/JUNIE 2011 AT / OM 12:00 Q1 Q2 Q3 Q4 Q5 Q6 TOTAL
UNIVERSITY OF PRETORIA / UNIVERSITEIT VAN PRETORIA FACULTY OF NATURAL AND AGRICULTURAL SCIENCES / FAKULTEIT NATUUR- EN LANDBOUWETENSKAPPE DEPARTMENT OF MATHEMATICS AND APPLIED MATHEMATICS / DEPARTEMENT
More informationUNIVERSITEIT VAN PRETORIA / UNIVERSITY OF PRETORIA WTW263 NUMERIESE METODES WTW263 NUMERICAL METHODS EKSAMEN / EXAMINATION
VAN/SURNAME : UNIVERSITEIT VAN PRETORIA / UNIVERSITY OF PRETORIA VOORNAME/FIRST NAMES : WTW26 NUMERIESE METODES WTW26 NUMERICAL METHODS EKSAMEN / EXAMINATION STUDENTENOMMER/STUDENT NUMBER : HANDTEKENING/SIGNATURE
More informationNovember 2005 TYD/TIME: 90 min PUNTE / MARKS: 35 VAN/SURNAME: VOORNAME/FIRST NAMES: STUDENTENOMMER/STUDENT NUMBER: HANDTEKENING/SIGNATURE:
UNIVERSITEIT VAN PRETORIA / UNIVERSITY OF PRETORIA DEPT WISKUNDE EN TOEGEPASTE WISKUNDE DEPT OF MATHEMATICS AND APPLIED MATHEMATICS WTW 263 - NUMERIESE METODES / NUMERICAL METHODS EKSAMEN / EXAMINATION
More informationUNIVERSITEIT VAN PRETORIA / UNIVERSITY OF PRETORIA DEPT WISKUNDE EN TOEGEPASTE WISKUNDE DEPT OF MATHEMATICS AND APPLIED MATHEMATICS
UNIVERSITEIT VAN PRETORIA / UNIVERSITY OF PRETORIA DEPT WISKUNDE EN TOEGEPASTE WISKUNDE DEPT OF MATHEMATICS AND APPLIED MATHEMATICS VAN/SURNAME: VOORNAME/FIRST NAMES: WTW 218 - CALCULUS SEMESTERTOETS /
More informationDR. J ROSSOUW Pro In9
THE PROBABILITY OF OCCURRENCE AND THE INTENSITY OF TROPICAL CYCLONES ALONG THE SOUTHERN AFRICAN EAST COAST by COSUS ROSSOUW Thesis prepared as part of the requirements For the degree Master in 5ngineering
More informationWTW 263 NUMERIESE METODES / NUMERICAL METHODS
UNIVERSITEIT VAN PRETORIA / UNIVERSITY OF PRETORIA FAKULTEIT NATUUR- EN LANDBOUWETENSKAPPE / FACULTY OF NATURAL AND AGRICULTURAL SCIENCES DEPARTEMENT WISKUNDE EN TOEGEPASTE WISKUNDE DEPARTMENT OF MATHEMATICS
More informationKwadratiese rye - Graad 11
OpenStax-CNX module: m38240 1 Kwadratiese rye - Graad 11 Wiehan Agenbag Free High School Science Texts Project Based on Quadratic Sequences - Grade 11 by Rory Adams Free High School Science Texts Project
More informationMATHEMATICS GRADE 10 TASK 1 INVESTIGATION Marks: 55
WISKUNDE GRAAD 10 TAAK 1 ONDERSOEK Punte: 55 MATHEMATICS GRADE 10 TASK 1 INVESTIGATION Marks: 55 INSTRUKSIES: 1. Die taak moet ingehandig word op 2 Maart 2015. 2. Beantwoord al die vrae. 3. Slegs vrae
More informationEXAMINATION / EKSAMEN 19 JUNE/JUNIE 2013 AT / OM 08:00
UNIVERSITY OF PRETORIA / UNIVERSITEIT VAN PRETORIA FACULTY OF NATURAL AND AGRICULTURAL SCIENCES / FAKULTEIT NATUUR- EN LANDBOUWETENSKAPPE DEPARTMENT OF MATHEMATICS AND APPLIED MATHEMATICS / DEPARTEMENT
More informationEKSAMEN / EXAMINATION Q1 Q2 Q3 Q4 Q5 TOTAL. 2. No pencil work or any work in red ink will be marked.
0 UNIVERSITEIT VAN PRETORIA / UNIVERSITY OF PRETORIA FAKULTEIT NATUUR- EN LANDBOUWETENSKAPPE / FACULTY OF NATURAL AND AGRICULTURAL SCIENCES DEPARTEMENT WISKUNDE EN TOEGEPASTE WISKUNDE DEPARTMENT OF MATHEMATICS
More informationFunksies en Verwantskappe
UITDRUKKINGS Funksies en Verwantskappe n funksie is n verhouding tussen waardes Elke inset waarde het n spesifieke uitset waarde a) Bereken die -waardes (uitset) b) Bereken die -waarde (inset) Die vloeidiagram
More informationFusion of Phoneme Recognisers for South African English
Fusion of Phoneme Recognisers for South African English George Wessel Strydom Thesis presented in partial fulfilment of the requirements for the degree Master of Science in Electronic Engineering at the
More informationDepartment of Mathematics and Applied Mathematics Departement Wiskunde en Toegepaste Wiskunde
Department of Mathematics and Applied Mathematics Departement Wiskunde en Toegepaste Wiskunde GRADES 10 AND 11 GRADE 10 EN 11 30 July 3 Aug 2018 30 Julie 3 Aug 2018 TIME: 2 HOURS TYD: 2 URE 2012 OUTEURSREG
More informationWTW 158 : CALCULUS EKSAMEN / EXAMINATION Eksterne eksaminator / External examiner: Prof NFJ van Rensburg
Outeursreg voorbehou UNIVERSITEIT VAN PRETORIA Departement Wiskunde en Toegepaste Wiskunde Copyright reserved UNIVERSITY OF PRETORIA Department of Mathematics and Applied Maths 4 Junie / June 00 Punte
More informationCAMI EDUCATION. Graad 12 Vraestel I : Rekord eksamen Punte. Lees die volgende instruksies noukeurig deur voordat die vrae beantwoord word:
CAMI Education (Pty) Ltd Reg. No. 1996/017609/07 CAMI House Fir Drive, Northcliff P.O. Bo 160 CRESTA, 118 Tel: +7 (11) 476-00 Fa : 086 601 4400 web: www.camiweb.com e-mail: info@camiweb.com CAMI EDUCATION
More informationGRADE 11 - FINAL ROUND QUESTIONS GRAAD 11 - FINALE RONDTE VRAE
GRE - FINL ROUN QUESTIONS - 9 GR - FINLE RONTE VRE - 9 QUESTION/ VRG 6 Which of the following is not a factor of x? 6 Watter een van die volgende is nie 'n faktor van x nie? x + x x + x + x+ E x QUESTION/
More informationEksterne eksaminator / External examiner: Dr. P Ntumba Interne eksaminatore / Internal examiners: Prof. I Broere, Prof. JE vd Berg, Dr.
VAN / SURNAME: UNIVERSITEIT VAN PRETORIA / UNIVERSITY OF PRETORIA FAKULTEIT NATUUR- EN LANDBOUWETENSKAPPE / FACULTY OF NATURAL AND AGRICULTURAL SCIENCES DEPARTEMENT WISKUNDE EN TOEGEPASTE WISKUNDE / DEPARTMENT
More information3. How many gadgets must he make and sell to make a profit of R1000?
The Entrepreneur An entrepreneur starts a new business. He manufactures gadgets at a cost of 4x + 200 rands and then sells them for 5,50x rands, where x is the number of gadgets produced. Profit = Income
More informationa b
GRDE - FIRST ROUND QUESTIONS - 0 GRD - EERSTE RONDTE VRE - 0 QUESTION/ VRG s a decimal number, 3,% is equal to: s n desimale breuk, word 3,% geskryf as: 0,03 B 0,3 C 3, D 3, E 3 QUESTION/ VRG a63 67 73
More informationHuiswerk Hoofstuk 22 Elektriese velde Homework Chapter 22 Electric fields
1 Huiswerk Hoofstuk Elektriese velde Homework Chapter Electric fields 8 th / 8 ste HRW 1, 5, 7, 10 (0), 43, 45, 47, 53 9 th / 9 de HRW (9.6 10 18 N left, 30 N/C), 3, 8 (0), 11, 39, 41, 47, 49 Elektriese
More informationDepartment of Mathematics and Applied Mathematics Departement Wiskunde en Toegepaste Wiskunde
Department of Mathematics and Applied Mathematics Departement Wiskunde en Toegepaste Wiskunde GRADES 0 AND GRADE 0 EN AUGUST 206 AUGUSTUS 206 TIME: 2 HOURS TYD: 2 URE 202 OUTEURSREG VOORBEHOU, UNIVERSITEIT
More informationGRADE 9 - FINAL ROUND QUESTIONS GRAAD 9 - FINALE RONDTE VRAE
GRADE 9 - FINAL ROUND QUESTIONS - 009 GRAAD 9 - FINALE RONDTE VRAE - 009 QUESTION/ VRAAG Find the final value if an amount of R7500 is increased by 5% and then decreased by 0%. Bepaal die finale waarde
More informationQuestion 1. The van der Waals equation of state is given by the equation: a
Kopiereg voorbehou Universiteit van retoria University of retoria Copyright reserved Departement Chemiese Ingenieurswese Department of Chemical Engineering CHEMICAL ENGINEERING CIR EKSAMEN Volpunte: Tydsduur:
More informationVibration Covariate Regression Analysis of Failure Time Data with the Proportional Hazards Model
Vibration Covariate Regression Analysis of Failure Time Data with the Proportional Hazards Model by Pieter-J an Vlok Submitted as partial fulfillment of the requirements for tbe degree Master in Engineering
More informationHOëRSKOOL STRAND WISKUNDE NOVEMBER 2016 GRAAD 11 VRAESTEL 2
HOëRSKOOL STRAND WISKUNDE NOVEMBER 2016 TOTAAL: 150 Eksaminator: P. Olivier GRAAD 11 VRAESTEL 2 TYD: 3UUR Moderator: E. Loedolff INSTRUKSIES: 1. Hierdie vraestel bestaan uit 8 bladsye en n DIAGRAMBLAD
More information[1a] 1, 3 [1b] 1, 0 [1c] 1, 3 en / and 1, 5 [1d] 1, 0 en / and 1, 0 [1e] Geen van hierdie / None of these
AFDELING A : MEERVOUDIGE KEUSE VRAE. 20 PUNTE Beantwoord vrae 1 tot 10 op die MERKLEESVORM se KANT 2. Indien kant 1 gebruik word sal dit nie nagesien word nie. Gebruik n sagte potlood. Jy mag nie verkeerde
More informationWTW 158 : CALCULUS EKSAMEN / EXAMINATION Eksterne eksaminator / External examiner: Me/Ms R Möller
Outeursreg voorbehou UNIVERSITEIT VAN PRETORIA Departement Wiskunde en Toegepaste Wiskunde Copyright reserved UNIVERSITY OF PRETORIA Department of Mathematics and Applied Maths Junie / June 005 Maksimum
More informationVAN/SURNAME: VOORNAME/FIRST NAMES: STUDENTENOMMER/STUDENT NUMBER: Totaal / Total:
UNIVERSITEIT VAN PRETORIA / UNIVERSITY OF PRETORIA DEPARTMENT OF MATHEMATICS AND APPLIED MATHEMATICS DEPARTEMENT WISKUNDE EN TOEGEPASTE WISKUNDE WTW 15 - WISKUNDIGE MODELLERING / MATHEMATICAL MODELLING
More informationSEMESTERTOETS 1 / SEMESTER TEST 1
UNIVERSITEIT VAN PRETORIA / UNIVERSITY OF PRETORIA FAKULTEIT NATUUR- EN LANDBOUWETENSKAPPE / FACULTY OF NATURAL AND AGRICULTURAL SCIENCES DEPARTEMENT WISKUNDE EN TOEGEPASTE WISKUNDE DEPARTMENT OF MATHEMATICS
More informationVAN / SURNAME: VOORNAME / FIRST NAMES: STUDENTENOMMER / STUDENT NUMBER: FOONNO. GEDURENDE EKSAMENPERIODE / PHONE NO. DURING EXAM PERIOD:
UNIVERSITEIT VAN PRETORIA / UNIVERSITY OF PRETORIA DEPT WISKUNDE EN TOEGEPASTE WISKUNDE DEPT OF MATHEMATICS AND APPLIED MATHEMATICS WTW 220 - ANALISE / ANALYSIS EKSAMEN / EXAM 12 November 2012 TYD/TIME:
More informationDATA MEASURES THAT CHARACTERISE
DATA MEASURES THAT CHARACTERISE CLASSIFICATION PROBLEMS CHRISTIAAN MAARTEN VAN DER WALT DATA MEASURES THAT CHARACTERISE CLASSIFICATION PROBLEMS by CHRISTIAAN MAARTEN VAN DER WALT Submitted in partial fulfilment
More informationExamination Copyright reserved. Eksamen Kopiereg voorbehou. Module EBN122 Elektrisiteit en Elektronika 13 November 2009
Departement Elektriese, Elektroniese en Rekenaar-Ingenieurswese Eksamen Kopiereg voorbehou Module EBN Elektrisiteit en Elektronika 3 November 009 Department of Electrical, Electronic and Computer Engineering
More informationGRAAD 12 SEPTEMBER 2012 WISKUNDE V3 MEMORANDUM
Province of the EASTERN CAPE EDUCATION NASIONALE SENIOR SERTIFIKAAT GRAAD SEPTEMBER 0 WISKUNDE V3 MEMORANDUM PUNTE: 00 Hierdie memorandum bestaan uit 3 bladsye. WISKUNDE V3 (SEPTEMBER 0) VRAAG. ; 4; 0;
More informationGRAAD 11 NOVEMBER 2012 WISKUNDIGE GELETTERDHEID V1 MEMORANDUM
Province of the EASTERN CAPE EDUCATION NASIONALE SENIOR SERTIFIKAAT GRAAD 11 NOVEMBER 2012 WISKUNDIGE GELETTERDHEID V1 MEMORANDUM MARKS: 100 SIMBOOL A CA C J M MA P R RT/RG S SF O VERDUIDELIKING Akkuraatheid
More informationGRADE 11 - FINAL ROUND QUESTIONS GRAAD 11 - FINALE RONDTE VRAE
GR 11 - FINL ROUN QUSTIONS - 007 GR 11 - FINL RONT VR - 007 1 QUSTION/ VRG 1 Three persons, Mr. X, Ms. Y, and Mr. Z, went fishing. Mr. X caught NO fish, Ms. Y caught five fish, and Mr. Z caught three fish
More informationNASIONALE SENIOR SERTIFIKAAT GRAAD 10
NASIONALE SENIOR SERTIFIKAAT GRAAD 10 WISKUNDE V MODEL 01 MEMORANDUM PUNTE: 100 Hierdie memorandum bestaan uit 10 bladsye. Wiskunde/V DBE/01 LET WEL: Indien 'n kandidaat 'n vraag TWEEKEER beantwoord, sien
More informationGraad 12: Rye en Reekse
Graad 1: Rye en Reekse Opgestel vir LitNet deur Jeanne-Mari du Plessis In hierdie inligtingstuk gaan ons kyk na: 1. Rekenkundige rye. Meetkundige rye 3. Rekenkundige reekse 4. Meetkundige reekse 5. Sigma-notasie
More informationHoofstuk 29 Magnetiese Velde a.g.v Elektriese Strome
Hoofstuk 29 Magnetiese Velde a.g.v Elektriese Strome Nadat hierdie hoofstuk afghandel is, moet die student: Magnetiese veld as gevolg van n stroom kan bereken; Die regterhandreëls kan neerskryf en toepas;
More informationSEEKING SPATIAL JUSTICE
SEEKING SPATIAL JUSTICE Empowering the everyday through an architecture that integrates the spatial and social realms ILHAAM TAYOB Submitted in partial fulfilment of the requirements for the degree Master
More informationPunte: Intern Marks: Internal WTW 168 : CALCULUS. EKSAMEN / EXAMINATION Eksterne eksaminator / External examiner: Me / Ms R Möller
Outeursreg voorbehou UNIVERSITEIT VAN PRETORIA Departement Wiskunde en Toegepaste Wiskunde Copright reserved UNIVERSITY OF PRETORIA Department of Mathematics and Applied Maths November 005 Punte / Marks:35
More informationVAN / SURNAME: VOORNAME / FIRST NAMES: STUDENTENOMMER / STUDENT NUMBER: HANDTEKENING / SIGNATURE: TELEFOON / TELEPHONE:
UNIVERSITEIT VAN PRETORIA / UNIVERSITY OF PRETORIA FAKULTEIT NATUUR- EN LANDBOUWETENSKAPPE / FACULTY OF NATURAL AND AGRICULTURAL SCIENCES DEPARTEMENT WISKUNDE EN TOEGEPASTE WISKUNDE DEPARTMENT OF MATHEMATICS
More informationDepartment of Mathematics and Applied Mathematics Departement Wiskunde en Toegepaste Wiskunde
Department of Mathematics and Applied Mathematics Departement Wiskunde en Toegepaste Wiskunde GRADES 10 AND 11 GRADE 10 EN 11 31 July 5 Aug 2017 31 July 5 Aug 2017 TIME: 2 HOURS TYD: 2 URE 2012 OUTEURSREG
More informationNATIONAL SENIOR CERTIFICATE NASIONALE SENIOR SERTIFIKAAT GRADE/GRAAD 12 JUNE/JUNIE 2018 MATHEMATICS P1/WISKUNDE V1 MARKING GUIDELINE/NASIENRIGLYN
NATIONAL SENIOR CERTIFICATE NASIONALE SENIOR SERTIFIKAAT GRADE/GRAAD JUNE/JUNIE 08 MATHEMATICS P/WISKUNDE V MARKING GUIDELINE/NASIENRIGLYN MARKS/PUNTE: 50 This marking guideline consists of 5 pages./ Hierdie
More informationTHE EFFECT OF HABITAT CHANGE ON THE STRUCTURE OF DUNG BEETLE ASSEMBLAGES IN THE NORTH-EASTERN FREE STATE: A COMPARISON OF CONSERVED AND FARMED LAND
THE EFFECT OF HABITAT CHANGE ON THE STRUCTURE OF DUNG BEETLE ASSEMBLAGES IN THE NORTH-EASTERN FREE STATE: A COMPARISON OF CONSERVED AND FARMED LAND by Astrid Jankielsohn Submitted in partial fulfilment
More informationAspects of the determination of the platinum group elements and arsenic by inductively. coupled plasma mass spectrometry
Aspects of the determination of the platinum group elements and arsenic by inductively coupled plasma mass spectrometry by Lilian Olga Schmidt Submitted in partial fulfilment of the requirements for the
More informationBEHAVIOUR OF CEMENTITIOUS SUBBASE LAYERS IN BITUMEN BASE ROAD STRUCTURES
BEHAVIOUR OF CEMENTITIOUS SUBBASE LAYERS IN BITUMEN BASE ROAD STRUCTURES Presented as partial fulfillment of the requirements for the degree M(Eng) in the Faculty of Engineering, University of Pretoria
More informationy =3x2 y 2 x 5 siny x y =6xy2 5x 4 siny
Skeibare PDVs/ Separable PDEs SLIDE 1/9 Parsiële afgeleides f ( x, y) Partial derivatives y f y x f x Voorbeeld/Example:f(x,y)=x 2 y 3 +x 5 cosy f x =2xy3 +5x 4 cosy; f y =3x2 y 2 x 5 siny 2 f x 2=2y3
More informationGovernment Gazette Staatskoerant
Government Gazette Staatskoerant REPUBLIC OF SOUTH AFRICA REPUBLIEK VAN SUID AFRIKA Regulation Gazette No. 10177 Regulasiekoerant Vol. 639 5 September September 2018 No. 41891 N.B. The Government Printing
More informationPERFORMANCE EVALUATION OF WET-COOLING TOWER FILLS WITH COMPUTATIONAL FLUID DYNAMICS
PERFORMANCE EVALUATION OF WET-COOLING TOWER FILLS WITH COMPUTATIONAL FLUID DYNAMICS By Yngvi Gudmundsson Thesis presented in partial fulfilment of the requirements for the degree of Master of Science in
More informationNATIONAL SENIOR CERTIFICATE GRADE 10 MATHEMATICS P3 PREPARATORY EXAMINATION 2008 NOVEMBER 2008
NATIONAL SENIOR CERTIFICATE GRADE 0 MATHEMATICS P3 PREPARATORY EXAMINATION 008 NOVEMBER 008 MARKS: 00 TIME: hours This question paper consists of pages and 4 diagram sheets. Mathematics/P DoE/Preparatory
More informationJAKKALS ROEP KURSUS JUNE 2016
JAKKALS ROEP KURSUS JUNE 2016 Thanks guys for a great time, it was a blast, thanks kindly for the references also, I am glad I could teach you methods that work. We had such good response I never even
More informationDirect Piston Displacement Control of Free-Piston Stirling Engines
Direct Piston Displacement Control of Free-Piston Stirling Engines Johannes Matthias Strauss Dissertation presented for the degree Doctor of Philosophy in the Faculty of Engineering Promoter: Prof M.J.
More informationNATIONAL SENIOR CERTIFICATE/ NASIONALE SENIOR SERTIFIKAAT GRADE/GRAAD 12 SEPTEMBER 2015 MATHEMATICS P1/WISKUNDE V1 MEMORANDUM
NATIONAL SENIOR CERTIFICATE/ NASIONALE SENIOR SERTIFIKAAT GRADE/GRAAD 12 SEPTEMBER 2015 MATHEMATICS P1/WISKUNDE V1 MEMORANDUM MARKS/PUNTE: 150 This memorandum consists of 15 pages./ Hierdie memorandum
More informationUNIVERSITY OF PRETORIA
UNIVERSITY OF PRETORIA FACULTY OF ENGINEERING, BUILT ENVIRONMENT AND INFORMATION TECHNOLOGY Student Number: Name: Surname: Email: Contact Number: Signature: Subject: MPR212: Programming and Date Processing
More informationSome Statistical Aspects of LULU smoothers
Some Statistical Aspects of LULU smoothers Maria Dorothea Jankowitz Dissertation presented for the degree of Doctor of Philosophy in the Department of Statistics and Actuarial Science at Stellenbosch University.
More informationVAN / SURNAME: VOORNAME / FIRST NAMES: STUDENTENOMMER / STUDENT NUMBER: HANDTEKENING / SIGNATURE: SEL NR / CELL NO:
UNIVERSITEIT VAN PRETORIA / UNIVERSITY OF PRETORIA FAKULTEIT NATUUR- EN LANDBOUWETENSKAPPE / FACULTY OF NATURAL AND AGRICULTURAL SCIENCES DEPARTEMENT WISKUNDE EN TOEGEPASTE WISKUNDE / DEPARTMENT OF MATHEMATICS
More informationTW 214 TOETS 2 - VOORBEREIDING 2018 TEST 2 - PREPARATION
TW 214 TOETS 2 - VOORBEREIDING 2018 TEST 2 - PREPARATION Die toets gaan oor die volgende onderwerpe: REFLEKSIES (GEKROMDE SPIEËLS) DETERMINANTE EIEWAARDES EN EIEWAARDE-ONTBINDING (A = SΛS 1 ) STELSELS
More informationA COMPARISON OF PHASE I CONTROL CHARTS. University of Pretoria, South Africa 1 4
A COMPARISON OF PHASE I CONTROL CHARTS M.L.I Coelho 1, S. Chakraborti 2,3 & M.A. Graham 4 1,2,4 Department of Statistics University of Pretoria, South Africa 1 margicoelho@gmail.com, 4 marien.graham@up.ac.za
More informationASSESSMENT OF FREQUENCY DOMAIN FORCE IDENTIFICATION PROCEDURES
ASSESSMENT OF FREQUENCY DOMAIN FORCE IDENTIFICATION PROCEDURES by THOMAS EDWIN SLAUGHTER KRIEL Submitted in partial fulfillment of the requirements for the degree Masters in Engineering in the Faculty
More informationAcoustic Signal Processing. Algorithms for Reverberant. Environments
Acoustic Signal Processing Algorithms for Reverberant Environments Terence Betlehem B.Sc. B.E.(Hons) ANU November 2005 A thesis submitted for the degree of Doctor of Philosophy of The Australian National
More informationSpectral and temporal modulation and characterization of femtosecond ultra-short laser pulses
Spectral and temporal modulation and characterization of femtosecond ultra-short laser pulses By Gaelle Carine Mbanda Nsoungui Thesis presented in partial fulfilment of the academic requirements for the
More informationLIMPOPO DEPARTEMENT VAN ONDERWYS LIMPOPO DEPARTMENT OF EDUCATION- LAERSKOOL WARMBAD
LIMPOPO DEPARTEMENT VAN ONDERWYS LIMPOPO DEPARTMENT OF EDUCATION- LAERSKOOL WARMBAD NSTRUKSIES: kryf jou naam en van op elke antwoordblad en jou nommer SUMMATIEWE ASSESSERING SUMMATIVE ASSESSMENT. Voltooi
More informationDEPRESSIE101. panic attacks - inside the brain TALKING about anxiety attacks. hanteer angstigheid beter snellers vir 'n paniekaanval
9 A P R I L 2 0 1 8 U I T G A W E 2 5 DEPRESSIE101 panic attacks - inside the brain TALKING about anxiety attacks hanteer angstigheid beter snellers vir 'n paniekaanval PANIC ATTACKS: INSIDE THE BRAIN
More informationUNIVERSITY OF PRETORIA DEPT SlVlELE INGENIEURSWESE / DEPT OF CIVIL ENGINEERING
1 UNIVERSCTEIT VAN PRETORIA I UNIVERSITY OF PRETORIA DEPT SlVlELE INGENIEURSWESE / DEPT OF CIVIL ENGINEERING STRENGTH OF MATERIALS SWK210 STERKTELEER SWK210 FINAL EXAMINATION - EiNDEKSAMEN VAN en VOORLETTERS
More informationAlgorithmic Component and System Reliability Analysis of Truss Structures
Algorithmic Component and System Reliability Analysis of Truss Structures by Sepehr Hashemolhosseini Thesis presented in fulfilment of the requirements for the degree of Master of Science in Civil Engineering
More information+ + SEPTEMBER 2016 MATHEMATICS PAPER 1 / WISKUNDE VRAESTEL 1 MEMORANDUM
SEPTEMBER 016 MATHEMATICS PAPER 1 / WISKUNDE VRAESTEL 1 MEMORANDUM NOTE: If a candidate answers a question TWICE, mark only the first one. Consistent accuracy applies in ALL aspects of the marking memorandum.
More informationFAKULTEIT INGENIEURSWESE FACULTY OF ENGINEERING
FAKULTEIT INGENIEURSWESE FACULTY OF ENGINEERING Vakkursus / Subject course: Toegepaste Wiskunde B264 Applied Mathematics B264 Tweede Eksamen: Desember 2010 Second Examination: December 2010 Volpunte /
More informationImproved estimation procedures for a positive extreme value index
Improved estimation procedures for a positive extreme value index Thomas Louw Berning Dissertation presented for the degree of Doctor of Philosophy at Stellenbosch University Promoter: Prof. Tertius de
More informationRecovery Based Error Estimation for the Method of Moments
Recovery Based Error Estimation for the Method of Moments by Willem Jacobus Strydom Thesis presented in partial fullment of the requirements for the degree of Master of Engineering in the Faculty of Engineering
More informationPRODUCTION OF ETHYL ACETATE USING CATALYTIC REACTION METHOD
PRODUCTION OF ETHYL ACETATE USING CATALYTIC REACTION METHOD Bamunusingha Arachchige Nadeeka Niroshinie Bamunusingha (108001 T) Thesis submitted in partial fulfillment of the requirements for the degree
More informationSome Exponential Diophantine Equations. Automan Sibusiso Mabaso
Some Exponential Diophantine Equations by Automan Sibusiso Mabaso Thesis presented in partial fulfilment of the requirements for the degree of Master of Science at Stellebosch University Supervisor: Dr.
More informationCurve Fitting Re-visited, Bishop1.2.5
Curve Fitting Re-visited, Bishop1.2.5 Maximum Likelihood Bishop 1.2.5 Model Likelihood differentiation p(t x, w, β) = Maximum Likelihood N N ( t n y(x n, w), β 1). (1.61) n=1 As we did in the case of the
More informationGovernment Gazette Staatskoerant
Government Gazette Staatskoerant REPUBLIC OF SOUTH AFRICA REPUBLIEK VAN SUID AFRIKA Regulation Gazette No. 10177 Regulasiekoerant Vol. 642 14 December Desember 2018 No. 42116 N.B. The Government Printing
More informationFAKULTEIT INGENIEURSWESE FACULTY OF ENGINEERING. Volpunte: Full marks: Instruksies / Instructions
FAKULTEIT INGENIEURSWESE FACULTY OF ENGINEERING Elektrotegniek 143 Electro-techniques 143 Tydsduur: Duration Eksaminatore: Prof H C Reader Prof J B de Swardt Mnr P-J Randewijk 1.5 h 1 Beantwoord al die
More informationUnsupervised Anomaly Detection for High Dimensional Data
Unsupervised Anomaly Detection for High Dimensional Data Department of Mathematics, Rowan University. July 19th, 2013 International Workshop in Sequential Methodologies (IWSM-2013) Outline of Talk Motivation
More informationDevelopment of a crane load software application for electric driven overhead travelling bridge cranes in accordance with SANS :2010
Development of a crane load software application for electric driven overhead travelling bridge cranes in accordance with SANS 10160-6:010 by Arthur William de Waal Thesis presented in partial fulfilment
More informationQuestion / Vraag 1: [12]
1 Question / Vraag 1: [12] The following questions are multiple choice questions. There is only one correct answer from the choices given. Select the correct option by marking the option with a cross (X).
More informationPattern Recognition. Parameter Estimation of Probability Density Functions
Pattern Recognition Parameter Estimation of Probability Density Functions Classification Problem (Review) The classification problem is to assign an arbitrary feature vector x F to one of c classes. The
More informationUNIVERSITEIT VAN PRETORIA / UNIVERSITY OF PRETORIA DEPT WISKUNDE EN TOEGEPASTE WISKUNDE DEPT OF MATHEMATICS AND APPLIED MATHEMATICS
UNIVERSITEIT VAN PRETORIA / UNIVERSITY OF PRETORIA DEPT WISKUNDE EN TOEGEPASTE WISKUNDE DEPT OF MATHEMATICS AND APPLIED MATHEMATICS WTW 211 : LINEÊRE ALGEBRA / LINEAR ALGEBRA SEMESTERTOETS 1 / SEMESTER
More informationPCA and CVA biplots: A study of their underlying theory and quality measures. Hilmarié Brand
PCA and CVA biplots: A study of their underlying theory and quality measures by Hilmarié Brand Thesis presented in partial fulfillment of the requirements for the degree of Master of Commerce in the faculty
More informationRHEOLOGICAL MODEL FOR PAINT PROPERTIES
RHEOLOGICAL MODEL FOR PAINT PROPERTIES by Pieter Lafras Moolman Dissertation presented for the Degree of DOCTOR OF PHILOSOPHY (Chemical Engineering) in the Department of Process Engineering at the University
More informationNATIONAL SENIOR CERTIFICATE/NASIONALE SENIOR SERTIFIKAAT GRADE/GRAAD 10
NATIONAL SENIOR CERTIFICATE/NASIONALE SENIOR SERTIFIKAAT GRADE/GRAAD 10 TECHNICAL SCIENCES: PHYSICS (P1)/ TEGNIESE WETENSKAPPE: FISIKA (V1) EXEMPLAR/MODEL 2016 MARKS/PUNTE: 150 This memorandum consists
More informationDETECTING CHANGE IN NONLINEAR DYNAMIC PROCESS SYSTEMS
DETECTING CHANGE IN NONLINEAR DYNAMIC PROCESS SYSTEMS - Leon Christo Bezuidenhout - Thesis submitted in partial fulfilment of the requirements for the degree Master of Science in Engineering (Chemical
More informationThe Modelling of IR Emission Spectra and Solid Rocket Motor Parameters using Neural Networks and Partial Least Squares
The Modelling of IR Emission Spectra and Solid Rocket Motor Parameters using Neural Networks and Partial Least Squares By Niko Hamp Thesis presented in partial fulfilment of the requirements for the degree
More informationGRADE 9 - FIRST ROUND QUESTIONS GRAAD 9 - EERSTE RONDTE VRAE
GRADE 9 - FIRST ROUND QUESTIONS - 009 GRAAD 9 - EERSTE RONDTE VRAE - 009 QUESTION/ VRAAG On Saturday, a store gives a discount of 0% on a computer valued at R9 000. On Monday, the store decides to add
More informationNumerical analysis of the flow distribution within packed columns using an explicit approach
Numerical analysis of the flow distribution within packed columns using an explicit approach by William George Joseph Theron B.Eng. Mechanical (North-West University) Student Number: 12900095 Dissertation
More informationProbabilistic Models of Design Wind Loads in South Africa
Probabilistic Models of Design Wind Loads in South Africa by Jacques Botha Dissertation presented for the degree of Doctor of Philosophy in the Faculty of Engineering at Stellenbosch University Department
More informationsoos gemeet in Fase C and D groeitoetse. Selekteer bulle
B E E F M A S T E R 2 0 1 2 51 52 B E E F M A S T E R 2 0 1 2 EFFICIENCY - Feed conversion ratio (FCR) This EBV is an indication of the genetic ability of the animal to efficiently convert feed into body
More informationThe Inverse Finite Element Method: Sensitivity to Measurement Setup
The Inverse Finite Element Method: Sensitivity to Measurement Setup A.J. Maree Thesis presented in partial fulfillment of the requirements for the degree of Master of Engineering at the University of Stellenbosch.
More informationNATIONAL SENIOR CERTIFICATE/ NASIONALE SENIOR SERTIFIKAAT GRADE/GRAAD 10
NATIONAL SENIOR CERTIFICATE/ NASIONALE SENIOR SERTIFIKAAT GRADE/GRAAD 0 MATHEMATICS P/WISKUNDE V NOVEMBER 05 MEMORANDUM MARKS: 50 PUNTE: 50 This memorandum consists of pages./ Hierdie memorandum bestaan
More informationDESIGN OF DOWELS FOR SHEAR TRANSFER AT THE INTERFACE BETWEEN CONCRETE CAST AT DIFFERENT TIMES: A CASE STUDY
DESIGN OF DOWELS FOR SHEAR TRANSFER AT THE INTERFACE BETWEEN CONCRETE CAST AT DIFFERENT TIMES: A CASE STUDY Samayamanthree Mudiyanselage Premasiri Karunarathna 118614J Degree of Master of Engineering in
More informationSupport Vector Machines using GMM Supervectors for Speaker Verification
1 Support Vector Machines using GMM Supervectors for Speaker Verification W. M. Campbell, D. E. Sturim, D. A. Reynolds MIT Lincoln Laboratory 244 Wood Street Lexington, MA 02420 Corresponding author e-mail:
More informationAn Open-Source Implementation to Predict Buckling Behaviour of Cold-Formed Sections
An Open-Source Implementation to Predict Buckling Behaviour of Cold-Formed Sections S-J Bronkhorst Thesis presented in partial fulfilment of the requirements for the degree of Master of Science in Engineering
More informationRole of Assembling Invariant Moments and SVM in Fingerprint Recognition
56 Role of Assembling Invariant Moments SVM in Fingerprint Recognition 1 Supriya Wable, 2 Chaitali Laulkar 1, 2 Department of Computer Engineering, University of Pune Sinhgad College of Engineering, Pune-411
More informationUNIVERSAL PORTFOLIO GENERATED BY IDEMPOTENT MATRIX AND SOME PROBABILITY DISTRIBUTION LIM KIAN HENG MASTER OF MATHEMATICAL SCIENCES
UNIVERSAL PORTFOLIO GENERATED BY IDEMPOTENT MATRIX AND SOME PROBABILITY DISTRIBUTION LIM KIAN HENG MASTER OF MATHEMATICAL SCIENCES FACULTY OF ENGINEERING AND SCIENCE UNIVERSITI TUNKU ABDUL RAHMAN APRIL
More informationSpeaker Representation and Verification Part II. by Vasileios Vasilakakis
Speaker Representation and Verification Part II by Vasileios Vasilakakis Outline -Approaches of Neural Networks in Speaker/Speech Recognition -Feed-Forward Neural Networks -Training with Back-propagation
More informationAlgebra van die vier basiese operasies
OpenStax-CNX module: m31388 1 Algebra van die vier basiese operasies Siyavula Uploaders This work is produced by OpenStax-CNX and licensed under the Creative Commons Attribution License 3.0 1 WISKUNDE
More information