Por favor, use este identificador para citar o enlazar este ítem:
https://hdl.handle.net/20.500.12104/83344Registro completo de metadatos
| Campo DC | Valor | Lengua/Idioma |
|---|---|---|
| dc.contributor.advisor | Sabin, Laurence | |
| dc.contributor.advisor | Corral Escobedo, Luis José Herminio | |
| dc.contributor.advisor | Ramírez Vélez, Julio César | |
| dc.contributor.advisor | Chavoya Peña, Arturo | |
| dc.contributor.author | Córdova Barbosa, Juan Pablo | |
| dc.date.accessioned | 2021-10-02T20:27:08Z | - |
| dc.date.available | 2021-10-02T20:27:08Z | - |
| dc.date.issued | 2019-03-29 | |
| dc.identifier.uri | https://wdg.biblio.udg.mx | |
| dc.identifier.uri | https://hdl.handle.net/20.500.12104/83344 | - |
| dc.description.tableofcontents | Contents 1 Introduction 1 1.1 SpecificAims ............................... 2 1.2 Hypotheses ................................ 3 1.3 Documentdesign ............................. 3 2 Methods 4 2.1 DataMining................................ 5 2.1.1 DataMiningMethods ...................... 7 2.1.1.1 Datacollection ..................... 7 2.1.1.2 Datatransformation .................. 7 2.1.1.3 FieldSelection ..................... 8 2.1.2 MachineLearningAlgorithms .................. 9 2.1.2.1 SupervisedMethods .................. 10 2.1.2.2 UnsupervisedMethods................. 12 2.2 DataMiningonAstronomy ....................... 13 2.2.1 MagneticFieldsonStars..................... 13 2.2.2 TheStokesParameters...................... 15 2.2.3 AstronomicDatabases ...................... 183 Experimental Design 20 3.1 Computationofthesyntheticdatabases . . . . . . . . . . . . . . . . 22 3.1.1 PropertiesSelection........................ 23 3.1.1.1 PhysicalCharacteristics ................ 23 3.1.1.2 SolarandStellarCases................. 28 3.1.2 Properties Delimitation and Parameter Range Selection . . . . 33 3.1.3 DataCollectionandDatabaseCreation. . . . . . . . . . . . . 36 3.2 DatabaseTransformation......................... 37 3.3 ANNTrainingandTesting........................ 41 4 Results 44 4.1 Solarcase ................................. 44 4.2 StellarCase ................................ 52 4.2.1 4.2.2 5 Discussion EffectiveMagneticField ..................... 52 4.2.1.1 FullDatabase...................... 52 4.2.1.2 SectionedDatabase................... 57 Otherparameters......................... 63 4.2.2.1 MagneticMoment ................... 63 4.2.2.2 SuperficialField .................... 71 78 5.1 Conclusions ................................ 79 5.2 Publications................................ 81 5.3 FutureWork................................ 81 A COSSAM configuration file 84B Hardware and Software specifications 85 Bibliography 87List of Figures 2.1 TraditionalANNschematic. ....................... 10 2.2 Ten angstroms fragment of the spectra for the Stokes I parameter for a synthetic object simulated with different Heff. . . . . . . . . .. 16 2.3 Ten angstroms fragment of the spectra for the Stokes V parameter for a synthetic object simulated with different Heff. . . . . . . . . .. 17 3.1 Processflowchart.............................. 22 3.2 COSSAM output example a) Not Normalized and b) Normalized. . . 27 3.3 COSSAM output example: Four stokes parameters (normalized) . . . 28 3.4 Grid configurations: Landstreet (blue circles) and Stift (red crosses). . 30 3.5 54 points grid configurations spectra: Landstreet spectrum (red line), Stift (blue line) and the difference between them (green line). . 31 3.6 Calculated Heff versus total number of points in the grid. . . . 3.7 MZS comparison for two synthetic objects with different Heff. . 3.8 MZScomparisonforcleanandnoisyspectra.. . . . . . . . . . . 3.9 MZS comparison for clean and highly noisy spectra. . . . . . . . 3.10 Artificial Neural Network Structure. The numbers tally with the . . . 32 .. . 39 . . . 40 .. . 41 totalnumberofnodesineachlayer.................... 424.1 Heff Regressionplot............................ 48 4.2 Heff Errors: (a)Absoluteand(b)Relative. . . . . . . . . . . . . . . 49 4.3 Combined case: Relative Errors Histogram and PDF. . . . . . . . . . 50 4.4 Heff>10G,RelativeErrors........................ 51 4.5 Heff >10G,RelativeErrorshistogramandPDF. . . . . . . . . . . . 51 4.6 Heff regression using the complete Stellar Database. . . . . . . . . . 53 4.7 Heff relative errors (complete Stellar Database). . . . . . . . . . . . . 54 4.8 |Hef f | > 10G, Relative errors histogram and PDF (complete Stellar Database).................................. 54 4.9 |Hef f | > 10G, Relative errors zoomed in histogram (complete Stellar Database).................................. 55 4.10 Heff regression using the complete Stellar Database with trimmed MZS..................................... 56 4.11 |Hef f | > 10 relative errors, complete Stellar Database, trimmed MZS. 57 4.12 Heff regression using the low noise Stellar Database with trimmed MZS..................................... 58 4.13 Heff relative errors, low noise Stellar Database, trimmed MZS. . . . . 59 4.14 |Hef f | > 2G Relative errors histogram and PDF, low noise, trimmed MZS,StellarDatabase). ......................... 59 4.15 |Hef f | > 10G Relative errors histogram and PDF, low noise, trimmed MZS,StellarDatabase). ......................... 60 4.16 |Hef f | > 10G Relative errors histogram zoomed in, low noise, trimmed MZS,StellarDatabase). ......................... 61 4.17 Heff regression using the high noise Stellar Database. . . . . . . . . . 614.18 Hef f relative errors, high noise Stellar Database, trimmed MZS. . . . 62 4.19 |Hef f | > 10G Relative errors histogram and PDF, high noise, trimmed MZS, Stellar Database). . . . . . . . . . . . . . . . . . . . . . . . . . 62 4.20 |Hef f | > 10G Relative errors histogram zoomed in high noise, trimmed MZS, Stellar Database). . . . . . . . . . . . . . . . . . . . . 63 4.21 Dipole moment (Fmag) regression using clean IV vectors from the Stellar Database. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 4.22 Fmag using clean IV vectors, relative errors. . . . . . . . . . . . . . . . 65 4.23 Fmag regression, low noise IV vectors. . . . . . . . . . . . . . . . . . . 66 4.24 “QUV” vector examples. . . . . . . . . . . . . . . . . . . . . . . . . . 66 4.25 Fmag regression using clean QUV vectors from the Stellar Database. . 67 4.26 Fmag using clean QUV vectors, relative errors. . . . . . . . . . . . . . 68 4.27 |Hef f | > 10G Relative errors histogram and PDF, clean QUV vectors. 68 4.28 |Hef f | > 10G Relative errors histogram zoomed in, clean QUV vectors. 68 4.29 Fmag regression using low noise QUV vectors. . . . . . . . . . . . . . 69 4.30 Fmag using low noise QUV vectors, relative errors. . . . . . . . . . . . 70 4.31 Fmag > 10G Relative errors histogram and PDF, low noise QUV vectors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 4.32 Fmag > 10G Relative errors histogram zoomed in, low noise QUV vectors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 4.33 Superficial field (Hsup) regression using clean QUV vectors. . . . . . . 72 4.34 Superficial field (Hsup) using clean QUV vectors, relative errors. . . . 72 4.35 Hsup > 50G Relative errors histogram and PDF, clean QUV vectors. . 73 4.36 Superficial field (Hsup) regression using low noise QUV vectors. . . . . 734.37 Superficial field (Hsup) using low noise QUV vectors, relative errors. . 74 4.38 Hsup > 50G Relative errors histogram and PDF, low noise QUV vectors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 4.39 Hsup > 50G Relative errors histogram zoomed in, low noise QUV vectors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 4.40 Superficial field (Hsup) regression using both low and high noise QUV vectors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 4.41 Superficial field (Hsup) using low and high noise QUV vectors, relative errors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 4.42 Hsup > 50G Relative errors histogram and PDF, low and high noise QUV vectors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 4.43 Hsup > 50G Relative errors histogram zoomed in, low and high noise QUV vectors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76List of Tables 3.1 Five lines example of a COSSAM output file. . . . . . . . . . . . . . . 27 3.2 Total number of points in the grid. . . . . . . . . . . . . . . . . . . . 33 3.3 First database: parameter information . . . . . . . . . . . . . . . . . 34 3.4 Second database: parameter information . . . . . . . . . . . . . . . . 35 3.5 NP to SNR equivalents . . . . . . . . . . . . . . . . . . . . . . . . . . 38 4.1 Training Algorithms: Mean Statistical measures. . . . . . . . . . . . . 46 4.2 Hef f regression: Statistical results. . . . . . . . . . . . . . . . . . . . 47 | |
| dc.format | application/PDF | |
| dc.language.iso | eng | |
| dc.publisher | Biblioteca Digital wdg.biblio | |
| dc.publisher | Universidad de Guadalajara | |
| dc.rights.uri | https://www.riudg.udg.mx/info/politicas.jsp | |
| dc.subject | Data Mining | |
| dc.title | Data Mining for Magnetic Field Parameters Determination. | |
| dc.type | Tesis de Doctorado | |
| dc.rights.holder | Universidad de Guadalajara | |
| dc.rights.holder | Córdova Barbosa, Juan Pablo | |
| dc.coverage | ZAPOPAN, JALISCO | |
| dc.type.conacyt | doctoralThesis | |
| dc.degree.name | DOCTORADO EN TECNOLOGIAS DE INFORMACION | |
| dc.degree.department | CUCEA | |
| dc.degree.grantor | Universidad de Guadalajara | |
| dc.degree.creator | DOCTOR EN TECNOLOGIAS DE INFORMACION | |
| dc.contributor.director | Navarro Jiménez, Silvana Guadalupe | |
| Aparece en las colecciones: | CUCEA | |
Ficheros en este ítem:
| Fichero | Tamaño | Formato | |
|---|---|---|---|
| DCUCEA10063.pdf Acceso Restringido | 14.12 MB | Adobe PDF | Visualizar/Abrir Request a copy |
Los ítems de RIUdeG están protegidos por copyright, con todos los derechos reservados, a menos que se indique lo contrario.