Anlisis tiempo-frecuencia a

Si continuamos haciendo lo mismo, lograremos los mismos resultados Albert Einstein

1.

Objetivos
Incorporar el concepto de frecuencia instantnea. a Conocer los detalles del anlisis mediante espectrogramas. a Analizar las ventajas comparativas de diferentes mtodos de anlisis tiempo-frecuencia. e a Revisar el principio de incertidumbre y el teorema de muestreo desde la perspectiva del anlisis tiempoa frecuencia. Utilizar la transformada onditas para analizar algunas seales sencillas. n Explorar aplicaciones del anlisis tiempo frecuencia y la transformada onditas. a

2.

Requisitos
Repase los conceptos estudiados en la unidad Anlisis de Fourier. a Haya asistido y estudiado la teor sobre Tiempo-frecuencia. a

Para alcanzar los objetivos de esta clase se necesita que el alumno:

3.
3.1.

Tcnicas de validacin y comparacin del desempe o de los algoe o o n ritmos

Naive Bayes

Scheme:weka.classifiers.bayes.NaiveBayes Relation: iris Instances: 150 Attributes: 5 sepallength sepalwidth petallength petalwidth class Test mode:split 66.0% train, remainder test === Classifier model (full training set) === Naive Bayes Classifier Class Iris-setosa Iris-versicolor Iris-virginica (0.33) (0.33) (0.33) =============================================================== sepallength mean 4.9913 5.9379 6.5795 std. dev. 0.355 0.5042 0.6353 weight sum 50 50 50 precision 0.1059 0.1059 0.1059 Attribute sepalwidth mean std. dev. weight sum precision petallength mean std. dev. weight sum precision petalwidth mean std. dev. weight sum precision

3.4015 0.3925 50 0.1091

2.7687 0.3038 50 0.1091

2.9629 0.3088 50 0.1091

1.4694 0.1782 50 0.1405

4.2452 0.4712 50 0.1405

5.5516 0.5529 50 0.1405

0.2743 0.1096 50 0.1143

1.3097 0.1915 50 0.1143

2.0343 0.2646 50 0.1143

Time taken to build model: 0 seconds === Evaluation on test split === === Summary === Correctly Classified Instances Incorrectly Classified Instances Kappa statistic Mean absolute error Root mean squared error Relative absolute error Root relative squared error Total Number of Instances === Detailed Accuracy By Class === TP Rate 1 0.947 0.882 0.941 FP Rate 0 0.063 0.029 0.033 Precision 1 0.9 0.938 0.942 Recall 1 0.947 0.882 0.941 F-Measure 1 0.923 0.909 0.941 ROC Area 1 0.988 0.988 0.992 Class Iris-setosa Iris-versicolor Iris-virginica 48 3 0.9113 0.0447 0.1722 10.0365 % 36.4196 % 51 94.1176 % 5.8824 %

Weighted Avg.

=== Confusion Matrix === a b c <-- classified as 15 0 0 | a = Iris-setosa 0 18 1 | b = Iris-versicolor 0 2 15 | c = Iris-virginica

3.2.

Perceptrn multicapa o

Scheme:weka.classifiers.functions.MultilayerPerceptron -L 0.3 -M 0.2 -N 500 -V 0 -S 0 -E 20 -H a Relation: iris Instances: 150 Attributes: 5 sepallength sepalwidth petallength petalwidth class Test mode:evaluate on training data === Classifier model (full training set) === Sigmoid Node 0 Inputs Weights Threshold -3.5015971588434005 Node 3 -1.0058110853859954 Node 4 9.07503844669134 Node 5 -4.107780453339232 Sigmoid Node 1 Inputs Weights Threshold 1.0692845992273172 Node 3 3.898873687789399 Node 4 -9.768910360340262 Node 5 -8.59913449315134 Sigmoid Node 2 Inputs Weights Threshold -1.0071762383436442 Node 3 -4.21840613382704 Node 4 -3.6260596863211187 Node 5 8.805122981737846 Sigmoid Node 3 Inputs Weights Threshold 3.3824855566856806 Attrib sepallength 0.9099827458022274 Attrib sepalwidth 1.5675138827531336 Attrib petallength -5.037338107319896 Attrib petalwidth -4.915469682506095 Sigmoid Node 4 Inputs Weights Threshold -3.3305735922918323 Attrib sepallength -1.1116750023770103 Attrib sepalwidth 3.1250096866676533 Attrib petallength -4.133137022912303 Attrib petalwidth -4.079589727871455 Sigmoid Node 5 Inputs Weights Threshold -7.496091023618096 Attrib sepallength -1.2158878822058805 Attrib sepalwidth -3.533282131753492 Attrib petallength 8.401834252274107 Attrib petalwidth 9.460215580472838 Class Iris-setosa Input Node 0 Class Iris-versicolor Input Node 1 Class Iris-virginica Input Node 2

Time taken to build model: 0.34 seconds === Evaluation on training set === === Summary === Correctly Classified Instances Incorrectly Classified Instances Kappa statistic Mean absolute error Root mean squared error Relative absolute error Root relative squared error Total Number of Instances 148 2 0.98 0.0248 0.0911 5.5779 % 19.3291 % 150 98.6667 % 1.3333 %

=== Detailed Accuracy By Class === TP Rate 1 0.98 0.98 0.987 FP Rate 0 0.01 0.01 0.007 Precision 1 0.98 0.98 0.987 Recall 1 0.98 0.98 0.987 F-Measure 1 0.98 0.98 0.987 ROC Area 1 0.999 0.999 0.999 Class Iris-setosa Iris-versicolor Iris-virginica

Weighted Avg.

=== Confusion Matrix === a b c <-- classified as 50 0 0 | a = Iris-setosa 0 49 1 | b = Iris-versicolor 0 1 49 | c = Iris-virginica

3.3.

Red de funciones de base radial

Scheme:weka.classifiers.functions.RBFNetwork -B 2 -S 1 -R 1.0E-8 -M -1 -W 0.1 Relation: iris Instances: 150 Attributes: 5 sepallength sepalwidth petallength petalwidth class Test mode:10-fold cross-validation === Classifier model (full training set) === Radial basis function network (Logistic regression applied to K-means clusters as basis functions): Logistic Regression with ridge parameter of 1.0E-8 Coefficients... Class Variable Iris-setosa Iris-versicolor ============================================================ pCluster_0_0 132.7289 -37.7972 pCluster_0_1 134.5741 -38.3226 pCluster_1_0 -70.3097 27.9147 pCluster_1_1 -63.3603 29.7589 pCluster_2_0 -67.1796 2.9573 pCluster_2_1 -74.4605 19.0193 Intercept 1.3218 -21.8733

Odds Ratios... Class Variable Iris-setosa Iris-versicolor ============================================================ pCluster_0_0 4.3997318920198195E57 0 pCluster_0_1 2.784886924641381E58 0 pCluster_1_0 0 1.3280364490842412E12 pCluster_1_1 0 8.39722335347585E12 pCluster_2_0 0 19.2459 pCluster_2_1 0 181964100.5948

Time taken to build model: 0.06 seconds === Stratified cross-validation === === Summary === Correctly Classified Instances Incorrectly Classified Instances Kappa statistic Mean absolute error Root mean squared error Relative absolute error Root relative squared error Total Number of Instances === Detailed Accuracy By Class === TP Rate 1 0.92 0.94 0.953 FP Rate 0 0.03 0.04 0.023 Precision 1 0.939 0.922 0.953 Recall 1 0.92 0.94 0.953 F-Measure 1 0.929 0.931 0.953 ROC Area 1 0.987 0.987 0.991 Class Iris-setosa Iris-versicolor Iris-virginica 143 7 0.93 0.034 0.1585 7.6407 % 33.6209 % 150 95.3333 % 4.6667 %

Weighted Avg.

=== Confusion Matrix === a b c <-- classified as 50 0 0 | a = Iris-setosa 0 46 4 | b = Iris-versicolor 0 3 47 | c = Iris-virginica

3.4.

C4.5

Scheme:weka.classifiers.trees.J48 -C 0.25 -M 2 Relation: iris Instances: 150 Attributes: 5 sepallength sepalwidth petallength petalwidth class Test mode:split 66.0% train, remainder test === Classifier model (full training set) === J48 pruned tree -----------------petalwidth <= 0.6: Iris-setosa (50.0) petalwidth > 0.6

| | | | | |

petalwidth <= 1.7 | petallength <= 4.9: Iris-versicolor (48.0/1.0) | petallength > 4.9 | | petalwidth <= 1.5: Iris-virginica (3.0) | | petalwidth > 1.5: Iris-versicolor (3.0/1.0) petalwidth > 1.7: Iris-virginica (46.0/1.0) : 9 5

Number of Leaves Size of the tree :

Time taken to build model: 0.04 seconds === Evaluation on test split === === Summary === Correctly Classified Instances Incorrectly Classified Instances Kappa statistic Mean absolute error Root mean squared error Relative absolute error Root relative squared error Total Number of Instances === Detailed Accuracy By Class === TP Rate 1 1 0.882 0.961 FP Rate 0 0.063 0 0.023 Precision 1 0.905 1 0.965 Recall 1 1 0.882 0.961 F-Measure 1 0.95 0.938 0.961 ROC Area 1 0.969 0.967 0.977 Class Iris-setosa Iris-versicolor Iris-virginica 49 2 0.9408 0.0396 0.1579 8.8979 % 33.4091 % 51 96.0784 % 3.9216 %

Weighted Avg.

=== Confusion Matrix === a b c <-- classified as 15 0 0 | a = Iris-setosa 0 19 0 | b = Iris-versicolor 0 2 15 | c = Iris-virginica

3.5.

Kmeans

Scheme:weka.clusterers.SimpleKMeans -N 2 -A "weka.core.EuclideanDistance -R first-last" -I 500 -S 10 Relation: iris Instances: 150 Attributes: 5 sepallength sepalwidth petallength petalwidth class Test mode:evaluate on training data === Model and evaluation on training set ===

kMeans ====== Number of iterations: 7 Within cluster sum of squared errors: 62.1436882815797 Missing values globally replaced with mean/mode Cluster centroids: Cluster# Full Data 0 1 (150) (100) (50) ================================================================== sepallength 5.8433 6.262 5.006 sepalwidth 3.054 2.872 3.418 petallength 3.7587 4.906 1.464 petalwidth 1.1987 1.676 0.244 class Iris-setosa Iris-versicolor Iris-setosa Attribute

Time taken to build model (full training data) : 0.01 seconds === Model and evaluation on training set === Clustered Instances 0 1 100 ( 67%) 50 ( 33%)