Librería Carat

Existen múltiples modelos de aprendizaje automático que pueden ser empleados para alcanzar los mismos objetivos. Esta variedad nos brinda la oportunidad de realizar pruebas con diferentes algoritmos para determinar cuál se adapta mejor al conjunto de datos en cuestión.

#Ejercicio 1. Modelo de Aprendizaje Automático

#install.packages("mlbench")
library(mlbench)
data(BreastCancer)

Instalar paquetes y llamar librerías

#install.packages("caret") #Algoritmos de aprendizaje
library(caret)

## Loading required package: ggplot2

## Loading required package: lattice

#install.packages("datasets") #Usar la base de datos "Iris"
library(datasets)
#install.packages("ggplot2") #Gráficas con mejor diseño
library(ggplot2)
#install.packages("lattice") #Crear gráficos
library(lattice)
#install.packages("DataExplorer") #Crear gráficos
library(DataExplorer)
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

Crear base de datos

df <- data.frame(BreastCancer)
df <- df %>% select(-Id)
df <- na.omit(df)

df$Cl.thickness <- as.numeric(as.character(df$Cl.thickness))
df$Cell.size <- as.numeric(as.character(df$Cell.size))
df$Cell.shape <- as.numeric(as.character(df$Cell.shape))
df$Marg.adhesion <- as.numeric(as.character(df$Marg.adhesion))
df$Epith.c.size <- as.numeric(as.character(df$Epith.c.size))
df$Bare.nuclei <- as.numeric(as.character(df$Bare.nuclei))
df$Bl.cromatin <- as.numeric(as.character(df$Bl.cromatin))
df$Normal.nucleoli <- as.numeric(as.character(df$Normal.nucleoli))
df$Mitoses <- as.numeric(as.character(df$Mitoses))

plot_histogram(df)

plot_correlation(df)

Partir datos 80/20

set.seed(123)
renglones_entrenamiento <- createDataPartition(df$Class, p=.8, list=FALSE)

entrenamiento <- df[renglones_entrenamiento, ]
prueba <- df[-renglones_entrenamiento, ]

Distintos tipos de métodos para modelar

Los métodos más utilizados para modelar aprendizaje automático son: * SVM. Support Vector Machine o Máquina de Vectores de Soporte. Hay varios subtipos: Lineal (svmLineal), Rdial (svmRadial), Polinómico (svmPoly), etc. * Árbol de Decisión. rpart * Redes Neuronales. nnet * Random Forest o Bosques Aleatorios. rf * Random Forest o Bosques Aleatorios. rf

str(df)

## 'data.frame':    683 obs. of  10 variables:
##  $ Cl.thickness   : num  5 5 3 6 4 8 1 2 2 4 ...
##  $ Cell.size      : num  1 4 1 8 1 10 1 1 1 2 ...
##  $ Cell.shape     : num  1 4 1 8 1 10 1 2 1 1 ...
##  $ Marg.adhesion  : num  1 5 1 1 3 8 1 1 1 1 ...
##  $ Epith.c.size   : num  2 7 2 3 2 7 2 2 2 2 ...
##  $ Bare.nuclei    : num  1 10 2 4 1 10 10 1 1 1 ...
##  $ Bl.cromatin    : num  3 3 3 3 3 9 3 3 1 2 ...
##  $ Normal.nucleoli: num  1 2 1 7 1 7 1 1 1 1 ...
##  $ Mitoses        : num  1 1 1 1 1 1 1 1 5 1 ...
##  $ Class          : Factor w/ 2 levels "benign","malignant": 1 1 1 1 1 2 1 1 1 1 ...
##  - attr(*, "na.action")= 'omit' Named int [1:16] 24 41 140 146 159 165 236 250 276 293 ...
##   ..- attr(*, "names")= chr [1:16] "24" "41" "140" "146" ...

summary(df)

##   Cl.thickness      Cell.size        Cell.shape     Marg.adhesion  
##  Min.   : 1.000   Min.   : 1.000   Min.   : 1.000   Min.   : 1.00  
##  1st Qu.: 2.000   1st Qu.: 1.000   1st Qu.: 1.000   1st Qu.: 1.00  
##  Median : 4.000   Median : 1.000   Median : 1.000   Median : 1.00  
##  Mean   : 4.442   Mean   : 3.151   Mean   : 3.215   Mean   : 2.83  
##  3rd Qu.: 6.000   3rd Qu.: 5.000   3rd Qu.: 5.000   3rd Qu.: 4.00  
##  Max.   :10.000   Max.   :10.000   Max.   :10.000   Max.   :10.00  
##   Epith.c.size     Bare.nuclei      Bl.cromatin     Normal.nucleoli
##  Min.   : 1.000   Min.   : 1.000   Min.   : 1.000   Min.   : 1.00  
##  1st Qu.: 2.000   1st Qu.: 1.000   1st Qu.: 2.000   1st Qu.: 1.00  
##  Median : 2.000   Median : 1.000   Median : 3.000   Median : 1.00  
##  Mean   : 3.234   Mean   : 3.545   Mean   : 3.445   Mean   : 2.87  
##  3rd Qu.: 4.000   3rd Qu.: 6.000   3rd Qu.: 5.000   3rd Qu.: 4.00  
##  Max.   :10.000   Max.   :10.000   Max.   :10.000   Max.   :10.00  
##     Mitoses             Class    
##  Min.   : 1.000   benign   :444  
##  1st Qu.: 1.000   malignant:239  
##  Median : 1.000                  
##  Mean   : 1.603                  
##  3rd Qu.: 1.000                  
##  Max.   :10.000

1. Modelo con el método svmLineal

library(caret)
modelo1 <- train(Class ~ ., data=entrenamiento, 
                method = "svmLinear", 
                preProcess = c("scale", "center"),
                trControl = trainControl(method="cv", number=10),
                tuneGrid = data.frame(C=1) #Cuando es svmLinear
                )

resultado_entrenamiento1 <- predict(modelo1, entrenamiento)
resultado_prueba1 <- predict(modelo1, prueba)

#Matriz de Consufión 
mcre1 <- confusionMatrix(resultado_entrenamiento1,entrenamiento$Class) #Matriz de confusion de resultado de entrenamiento
mcre1

## Confusion Matrix and Statistics
## 
##            Reference
## Prediction  benign malignant
##   benign       347         7
##   malignant      9       185
##                                          
##                Accuracy : 0.9708         
##                  95% CI : (0.953, 0.9832)
##     No Information Rate : 0.6496         
##     P-Value [Acc > NIR] : <2e-16         
##                                          
##                   Kappa : 0.936          
##                                          
##  Mcnemar's Test P-Value : 0.8026         
##                                          
##             Sensitivity : 0.9747         
##             Specificity : 0.9635         
##          Pos Pred Value : 0.9802         
##          Neg Pred Value : 0.9536         
##              Prevalence : 0.6496         
##          Detection Rate : 0.6332         
##    Detection Prevalence : 0.6460         
##       Balanced Accuracy : 0.9691         
##                                          
##        'Positive' Class : benign         
## 

mcrp1 <- confusionMatrix(resultado_prueba1, prueba$Class)  #Matriz de confusion de resultado de prueba
mcrp1

## Confusion Matrix and Statistics
## 
##            Reference
## Prediction  benign malignant
##   benign        87         2
##   malignant      1        45
##                                           
##                Accuracy : 0.9778          
##                  95% CI : (0.9364, 0.9954)
##     No Information Rate : 0.6519          
##     P-Value [Acc > NIR] : <2e-16          
##                                           
##                   Kappa : 0.9508          
##                                           
##  Mcnemar's Test P-Value : 1               
##                                           
##             Sensitivity : 0.9886          
##             Specificity : 0.9574          
##          Pos Pred Value : 0.9775          
##          Neg Pred Value : 0.9783          
##              Prevalence : 0.6519          
##          Detection Rate : 0.6444          
##    Detection Prevalence : 0.6593          
##       Balanced Accuracy : 0.9730          
##                                           
##        'Positive' Class : benign          
## 

2. Modelo con el método svmRadial

modelo2 <- train(Class ~ ., data=entrenamiento, 
                method = "svmRadial", 
                preProcess = c("scale", "center"),
                trControl = trainControl(method="cv", number=10),
                tuneGrid = data.frame(sigma=1,C=1) #Cambiar
                )

resultado_entrenamiento2 <- predict(modelo2, entrenamiento)
resultado_prueba2 <- predict(modelo2, prueba)

#Matriz de Consufión 
mcre2 <- confusionMatrix(resultado_entrenamiento2,entrenamiento$Class) #Matriz de confusion de resultado de entrenamiento
mcre2

## Confusion Matrix and Statistics
## 
##            Reference
## Prediction  benign malignant
##   benign       354         0
##   malignant      2       192
##                                           
##                Accuracy : 0.9964          
##                  95% CI : (0.9869, 0.9996)
##     No Information Rate : 0.6496          
##     P-Value [Acc > NIR] : <2e-16          
##                                           
##                   Kappa : 0.992           
##                                           
##  Mcnemar's Test P-Value : 0.4795          
##                                           
##             Sensitivity : 0.9944          
##             Specificity : 1.0000          
##          Pos Pred Value : 1.0000          
##          Neg Pred Value : 0.9897          
##              Prevalence : 0.6496          
##          Detection Rate : 0.6460          
##    Detection Prevalence : 0.6460          
##       Balanced Accuracy : 0.9972          
##                                           
##        'Positive' Class : benign          
## 

mcrp2 <- confusionMatrix(resultado_prueba2, prueba$Class)  #Matriz de confusion de resultado de prueba
mcrp2

## Confusion Matrix and Statistics
## 
##            Reference
## Prediction  benign malignant
##   benign        82         0
##   malignant      6        47
##                                           
##                Accuracy : 0.9556          
##                  95% CI : (0.9058, 0.9835)
##     No Information Rate : 0.6519          
##     P-Value [Acc > NIR] : < 2e-16         
##                                           
##                   Kappa : 0.9049          
##                                           
##  Mcnemar's Test P-Value : 0.04123         
##                                           
##             Sensitivity : 0.9318          
##             Specificity : 1.0000          
##          Pos Pred Value : 1.0000          
##          Neg Pred Value : 0.8868          
##              Prevalence : 0.6519          
##          Detection Rate : 0.6074          
##    Detection Prevalence : 0.6074          
##       Balanced Accuracy : 0.9659          
##                                           
##        'Positive' Class : benign          
## 

3. Modelo con el método svmPoly

modelo3 <- train(Class ~ ., data = entrenamiento,
                  method = "svmPoly",
                  preProcess = c("scale", "center"),
                  trControl = trainControl(method = "cv", number = 10),
                  tuneGrid = data.frame(degree = 1, scale = 1, C = 1))  # Adjust values as needed

resultado_entrenamiento3 <- predict(modelo3, entrenamiento)
resultado_prueba3 <- predict(modelo3, prueba)

#Matriz de Consufión 
mcre3 <- confusionMatrix(resultado_entrenamiento3,entrenamiento$Class) #Matriz de confusion de resultado de entrenamiento
mcre3

## Confusion Matrix and Statistics
## 
##            Reference
## Prediction  benign malignant
##   benign       347         7
##   malignant      9       185
##                                          
##                Accuracy : 0.9708         
##                  95% CI : (0.953, 0.9832)
##     No Information Rate : 0.6496         
##     P-Value [Acc > NIR] : <2e-16         
##                                          
##                   Kappa : 0.936          
##                                          
##  Mcnemar's Test P-Value : 0.8026         
##                                          
##             Sensitivity : 0.9747         
##             Specificity : 0.9635         
##          Pos Pred Value : 0.9802         
##          Neg Pred Value : 0.9536         
##              Prevalence : 0.6496         
##          Detection Rate : 0.6332         
##    Detection Prevalence : 0.6460         
##       Balanced Accuracy : 0.9691         
##                                          
##        'Positive' Class : benign         
## 

mcrp3 <- confusionMatrix(resultado_prueba3, prueba$Class)  #Matriz de confusion de resultado de prueba
mcrp3

## Confusion Matrix and Statistics
## 
##            Reference
## Prediction  benign malignant
##   benign        87         2
##   malignant      1        45
##                                           
##                Accuracy : 0.9778          
##                  95% CI : (0.9364, 0.9954)
##     No Information Rate : 0.6519          
##     P-Value [Acc > NIR] : <2e-16          
##                                           
##                   Kappa : 0.9508          
##                                           
##  Mcnemar's Test P-Value : 1               
##                                           
##             Sensitivity : 0.9886          
##             Specificity : 0.9574          
##          Pos Pred Value : 0.9775          
##          Neg Pred Value : 0.9783          
##              Prevalence : 0.6519          
##          Detection Rate : 0.6444          
##    Detection Prevalence : 0.6593          
##       Balanced Accuracy : 0.9730          
##                                           
##        'Positive' Class : benign          
## 

4. Modelo con el método rpart

modelo4 <- train(Class ~ ., data=entrenamiento, 
                method = "rpart", 
                preProcess = c("scale", "center"),
                trControl = trainControl(method="cv", number=10),
                tuneLength = 10 #Cambiar
                )

resultado_entrenamiento4 <- predict(modelo4, entrenamiento)
resultado_prueba4 <- predict(modelo4, prueba)

#Matriz de Consufión 
mcre4 <- confusionMatrix(resultado_entrenamiento4,entrenamiento$Class) #Matriz de confusion de resultado de entrenamiento
mcre4

## Confusion Matrix and Statistics
## 
##            Reference
## Prediction  benign malignant
##   benign       345         9
##   malignant     11       183
##                                           
##                Accuracy : 0.9635          
##                  95% CI : (0.9442, 0.9776)
##     No Information Rate : 0.6496          
##     P-Value [Acc > NIR] : <2e-16          
##                                           
##                   Kappa : 0.92            
##                                           
##  Mcnemar's Test P-Value : 0.8231          
##                                           
##             Sensitivity : 0.9691          
##             Specificity : 0.9531          
##          Pos Pred Value : 0.9746          
##          Neg Pred Value : 0.9433          
##              Prevalence : 0.6496          
##          Detection Rate : 0.6296          
##    Detection Prevalence : 0.6460          
##       Balanced Accuracy : 0.9611          
##                                           
##        'Positive' Class : benign          
## 

mcrp4 <- confusionMatrix(resultado_prueba4, prueba$Class)  #Matriz de confusion de resultado de prueba
mcrp4

## Confusion Matrix and Statistics
## 
##            Reference
## Prediction  benign malignant
##   benign        87         5
##   malignant      1        42
##                                           
##                Accuracy : 0.9556          
##                  95% CI : (0.9058, 0.9835)
##     No Information Rate : 0.6519          
##     P-Value [Acc > NIR] : <2e-16          
##                                           
##                   Kappa : 0.9001          
##                                           
##  Mcnemar's Test P-Value : 0.2207          
##                                           
##             Sensitivity : 0.9886          
##             Specificity : 0.8936          
##          Pos Pred Value : 0.9457          
##          Neg Pred Value : 0.9767          
##              Prevalence : 0.6519          
##          Detection Rate : 0.6444          
##    Detection Prevalence : 0.6815          
##       Balanced Accuracy : 0.9411          
##                                           
##        'Positive' Class : benign          
## 

5. Modelo con el método nnet

modelo5 <- train(Class ~ ., data=entrenamiento, 
                method = "nnet", 
                preProcess = c("scale", "center"),
                trControl = trainControl(method="cv", number=10)
                )

## # weights:  12
## initial  value 376.512397 
## iter  10 value 43.109567
## iter  20 value 37.816789
## iter  30 value 37.393307
## iter  40 value 37.202710
## iter  50 value 36.691152
## iter  60 value 36.645519
## iter  70 value 36.203034
## iter  80 value 36.068646
## iter  90 value 36.066816
## iter 100 value 35.990888
## final  value 35.990888 
## stopped after 100 iterations
## # weights:  34
## initial  value 398.785839 
## iter  10 value 40.921332
## iter  20 value 32.226888
## iter  30 value 28.947400
## iter  40 value 27.690881
## iter  50 value 27.258233
## iter  60 value 27.084473
## iter  70 value 26.862289
## iter  80 value 26.362282
## iter  90 value 25.443885
## iter 100 value 24.871772
## final  value 24.871772 
## stopped after 100 iterations
## # weights:  56
## initial  value 392.518725 
## iter  10 value 32.515498
## iter  20 value 12.874502
## iter  30 value 7.985029
## iter  40 value 6.194120
## iter  50 value 5.774182
## iter  60 value 5.718611
## iter  70 value 5.680335
## iter  80 value 5.663894
## iter  90 value 5.652273
## iter 100 value 5.644655
## final  value 5.644655 
## stopped after 100 iterations
## # weights:  12
## initial  value 350.285052 
## iter  10 value 63.176600
## iter  20 value 49.960565
## iter  30 value 49.668121
## final  value 49.667643 
## converged
## # weights:  34
## initial  value 349.498531 
## iter  10 value 115.292299
## iter  20 value 47.167744
## iter  30 value 40.701373
## iter  40 value 39.136713
## iter  50 value 37.993613
## iter  60 value 37.722960
## iter  70 value 37.705185
## iter  80 value 37.705094
## final  value 37.705093 
## converged
## # weights:  56
## initial  value 385.790168 
## iter  10 value 41.568917
## iter  20 value 37.839293
## iter  30 value 37.366085
## iter  40 value 37.244380
## iter  50 value 37.184247
## iter  60 value 37.148147
## iter  70 value 37.147200
## final  value 37.147103 
## converged
## # weights:  12
## initial  value 411.197079 
## iter  10 value 126.101033
## iter  20 value 64.916942
## iter  30 value 62.477071
## iter  40 value 62.434341
## iter  50 value 62.420463
## iter  60 value 60.564723
## iter  70 value 44.580523
## iter  80 value 40.887557
## iter  90 value 37.328363
## iter 100 value 36.937571
## final  value 36.937571 
## stopped after 100 iterations
## # weights:  34
## initial  value 396.809103 
## iter  10 value 36.803808
## iter  20 value 28.222616
## iter  30 value 24.692421
## iter  40 value 21.283836
## iter  50 value 20.693931
## iter  60 value 20.509633
## iter  70 value 20.285770
## iter  80 value 20.104851
## iter  90 value 20.012950
## iter 100 value 19.939195
## final  value 19.939195 
## stopped after 100 iterations
## # weights:  56
## initial  value 470.495322 
## iter  10 value 37.226297
## iter  20 value 28.536698
## iter  30 value 24.831327
## iter  40 value 18.218403
## iter  50 value 16.208980
## iter  60 value 15.441777
## iter  70 value 15.044443
## iter  80 value 14.851473
## iter  90 value 14.766855
## iter 100 value 14.663473
## final  value 14.663473 
## stopped after 100 iterations
## # weights:  12
## initial  value 355.459360 
## iter  10 value 52.710948
## iter  20 value 50.543680
## iter  30 value 42.778082
## iter  40 value 42.575005
## iter  50 value 41.896598
## iter  60 value 39.558006
## iter  70 value 39.540621
## iter  80 value 39.527535
## iter  90 value 39.516675
## iter 100 value 39.510296
## final  value 39.510296 
## stopped after 100 iterations
## # weights:  34
## initial  value 345.094693 
## iter  10 value 38.864054
## iter  20 value 33.098589
## iter  30 value 28.931958
## iter  40 value 28.548481
## iter  50 value 28.498623
## iter  60 value 28.425275
## iter  70 value 28.379912
## iter  80 value 28.347322
## iter  90 value 28.340527
## iter 100 value 28.338452
## final  value 28.338452 
## stopped after 100 iterations
## # weights:  56
## initial  value 318.277821 
## iter  10 value 39.852981
## iter  20 value 20.644864
## iter  30 value 11.484980
## iter  40 value 10.229586
## iter  50 value 9.343020
## iter  60 value 9.253535
## iter  70 value 9.140352
## iter  80 value 3.398547
## iter  90 value 2.576593
## iter 100 value 2.513437
## final  value 2.513437 
## stopped after 100 iterations
## # weights:  12
## initial  value 383.439080 
## iter  10 value 68.936910
## iter  20 value 55.011950
## iter  30 value 53.826306
## iter  40 value 53.551184
## final  value 53.550992 
## converged
## # weights:  34
## initial  value 437.903342 
## iter  10 value 53.347573
## iter  20 value 46.839645
## iter  30 value 42.789234
## iter  40 value 42.318303
## iter  50 value 41.863757
## iter  60 value 41.778420
## iter  70 value 41.771466
## iter  70 value 41.771466
## final  value 41.771466 
## converged
## # weights:  56
## initial  value 432.922085 
## iter  10 value 58.515744
## iter  20 value 43.453015
## iter  30 value 40.439290
## iter  40 value 40.215683
## iter  50 value 40.108368
## iter  60 value 40.091683
## iter  70 value 40.089514
## final  value 40.089511 
## converged
## # weights:  12
## initial  value 370.554674 
## iter  10 value 53.682403
## iter  20 value 46.277541
## iter  30 value 43.347649
## iter  40 value 39.935096
## iter  50 value 39.654913
## iter  60 value 39.643184
## iter  70 value 39.640909
## iter  80 value 39.638620
## iter  90 value 39.638417
## iter 100 value 39.638344
## final  value 39.638344 
## stopped after 100 iterations
## # weights:  34
## initial  value 371.288609 
## iter  10 value 41.762077
## iter  20 value 36.441925
## iter  30 value 35.026523
## iter  40 value 34.445840
## iter  50 value 31.489348
## iter  60 value 31.411756
## iter  70 value 31.378085
## iter  80 value 31.193215
## iter  90 value 30.081311
## iter 100 value 27.611485
## final  value 27.611485 
## stopped after 100 iterations
## # weights:  56
## initial  value 324.078656 
## iter  10 value 35.203759
## iter  20 value 18.713852
## iter  30 value 11.486088
## iter  40 value 11.093492
## iter  50 value 10.415557
## iter  60 value 10.268160
## iter  70 value 10.037759
## iter  80 value 9.978048
## iter  90 value 9.758822
## iter 100 value 9.561419
## final  value 9.561419 
## stopped after 100 iterations
## # weights:  12
## initial  value 376.872469 
## iter  10 value 55.754543
## iter  20 value 48.254280
## iter  30 value 45.399366
## iter  40 value 43.383439
## iter  50 value 39.502387
## iter  60 value 39.486663
## iter  70 value 39.485217
## iter  80 value 39.482703
## iter  90 value 39.482220
## iter 100 value 39.479492
## final  value 39.479492 
## stopped after 100 iterations
## # weights:  34
## initial  value 461.270341 
## iter  10 value 36.971158
## iter  20 value 29.190367
## iter  30 value 19.680057
## iter  40 value 18.250577
## iter  50 value 18.155907
## iter  60 value 18.103269
## iter  70 value 18.056877
## iter  80 value 17.989927
## iter  90 value 17.791648
## iter 100 value 17.741487
## final  value 17.741487 
## stopped after 100 iterations
## # weights:  56
## initial  value 291.158632 
## iter  10 value 32.488217
## iter  20 value 26.334718
## iter  30 value 23.977332
## iter  40 value 18.600272
## iter  50 value 17.936452
## iter  60 value 17.807248
## iter  70 value 17.734333
## iter  80 value 17.560468
## iter  90 value 17.500910
## iter 100 value 17.412769
## final  value 17.412769 
## stopped after 100 iterations
## # weights:  12
## initial  value 350.758871 
## iter  10 value 61.913337
## iter  20 value 52.799784
## iter  30 value 52.782791
## final  value 52.782280 
## converged
## # weights:  34
## initial  value 325.534606 
## iter  10 value 53.573710
## iter  20 value 47.224232
## iter  30 value 45.210505
## iter  40 value 42.197093
## iter  50 value 41.624769
## iter  60 value 41.618364
## iter  70 value 41.576205
## iter  80 value 41.564176
## final  value 41.564175 
## converged
## # weights:  56
## initial  value 290.753871 
## iter  10 value 82.927340
## iter  20 value 50.542865
## iter  30 value 44.382185
## iter  40 value 42.431680
## iter  50 value 41.058181
## iter  60 value 39.768515
## iter  70 value 38.984534
## iter  80 value 38.964683
## iter  90 value 38.963632
## final  value 38.963626 
## converged
## # weights:  12
## initial  value 429.336451 
## iter  10 value 47.883049
## iter  20 value 43.499831
## iter  30 value 41.880976
## iter  40 value 39.366043
## iter  50 value 39.341667
## iter  60 value 39.316204
## iter  70 value 39.290956
## iter  80 value 39.284362
## iter  90 value 39.281555
## iter 100 value 39.275839
## final  value 39.275839 
## stopped after 100 iterations
## # weights:  34
## initial  value 476.298677 
## iter  10 value 42.819519
## iter  20 value 31.517941
## iter  30 value 27.595750
## iter  40 value 26.682247
## iter  50 value 24.366580
## iter  60 value 23.065664
## iter  70 value 22.701349
## iter  80 value 22.590481
## iter  90 value 22.495564
## iter 100 value 22.349801
## final  value 22.349801 
## stopped after 100 iterations
## # weights:  56
## initial  value 378.896960 
## iter  10 value 36.842685
## iter  20 value 18.639488
## iter  30 value 7.801945
## iter  40 value 7.569890
## iter  50 value 7.337884
## iter  60 value 7.167823
## iter  70 value 7.036394
## iter  80 value 6.917578
## iter  90 value 6.759553
## iter 100 value 6.626162
## final  value 6.626162 
## stopped after 100 iterations
## # weights:  12
## initial  value 344.820755 
## iter  10 value 55.774708
## iter  20 value 51.813829
## iter  30 value 48.483388
## iter  40 value 48.360414
## iter  50 value 45.090975
## iter  60 value 44.999497
## iter  70 value 44.811518
## iter  80 value 44.788230
## iter  90 value 44.779972
## iter 100 value 44.779306
## final  value 44.779306 
## stopped after 100 iterations
## # weights:  34
## initial  value 324.434349 
## iter  10 value 35.073614
## iter  20 value 28.309883
## iter  30 value 24.379088
## iter  40 value 19.858114
## iter  50 value 18.851557
## iter  60 value 18.478818
## iter  70 value 18.354701
## iter  80 value 18.329160
## iter  90 value 18.317284
## iter 100 value 18.315578
## final  value 18.315578 
## stopped after 100 iterations
## # weights:  56
## initial  value 301.471018 
## iter  10 value 31.048488
## iter  20 value 26.683856
## iter  30 value 22.148084
## iter  40 value 19.147597
## iter  50 value 17.815825
## iter  60 value 17.614277
## iter  70 value 17.408252
## iter  80 value 17.337243
## iter  90 value 17.136861
## iter 100 value 17.034617
## final  value 17.034617 
## stopped after 100 iterations
## # weights:  12
## initial  value 441.591031 
## iter  10 value 52.286962
## iter  20 value 47.424877
## iter  30 value 46.692740
## final  value 46.692073 
## converged
## # weights:  34
## initial  value 400.902943 
## iter  10 value 38.230663
## iter  20 value 36.476461
## iter  30 value 36.435445
## iter  40 value 36.381102
## iter  50 value 36.366204
## final  value 36.366111 
## converged
## # weights:  56
## initial  value 354.704046 
## iter  10 value 40.110429
## iter  20 value 35.734147
## iter  30 value 34.752683
## iter  40 value 34.670163
## iter  50 value 34.663302
## final  value 34.662961 
## converged
## # weights:  12
## initial  value 382.292257 
## iter  10 value 51.368504
## iter  20 value 40.145722
## iter  30 value 36.402339
## iter  40 value 36.363779
## iter  50 value 36.353899
## iter  60 value 36.352843
## iter  70 value 36.351267
## iter  80 value 36.350961
## iter  90 value 36.350758
## iter 100 value 36.350596
## final  value 36.350596 
## stopped after 100 iterations
## # weights:  34
## initial  value 384.939391 
## iter  10 value 62.462506
## iter  20 value 26.735137
## iter  30 value 22.711353
## iter  40 value 19.620734
## iter  50 value 18.236719
## iter  60 value 17.387882
## iter  70 value 17.273982
## iter  80 value 16.910083
## iter  90 value 16.867087
## iter 100 value 16.833337
## final  value 16.833337 
## stopped after 100 iterations
## # weights:  56
## initial  value 517.139303 
## iter  10 value 208.864116
## iter  20 value 21.380448
## iter  30 value 19.541237
## iter  40 value 18.126321
## iter  50 value 17.710978
## iter  60 value 17.540169
## iter  70 value 17.450041
## iter  80 value 17.277237
## iter  90 value 17.091628
## iter 100 value 16.957556
## final  value 16.957556 
## stopped after 100 iterations
## # weights:  12
## initial  value 292.888398 
## iter  10 value 51.333516
## iter  20 value 47.789397
## iter  30 value 42.795220
## iter  40 value 42.495016
## iter  50 value 42.488166
## iter  60 value 42.486191
## iter  70 value 42.482677
## iter  80 value 42.481495
## iter  90 value 42.480611
## iter 100 value 42.479688
## final  value 42.479688 
## stopped after 100 iterations
## # weights:  34
## initial  value 370.309048 
## iter  10 value 38.906729
## iter  20 value 33.553344
## iter  30 value 27.802055
## iter  40 value 18.987398
## iter  50 value 17.329191
## iter  60 value 17.253432
## iter  70 value 17.235331
## iter  80 value 17.229974
## iter  90 value 17.216500
## iter 100 value 17.198776
## final  value 17.198776 
## stopped after 100 iterations
## # weights:  56
## initial  value 330.962845 
## iter  10 value 41.726571
## iter  20 value 25.590695
## iter  30 value 18.909810
## iter  40 value 15.860228
## iter  50 value 14.256631
## iter  60 value 13.820280
## iter  70 value 13.049504
## iter  80 value 12.975596
## iter  90 value 12.926084
## iter 100 value 12.904999
## final  value 12.904999 
## stopped after 100 iterations
## # weights:  12
## initial  value 445.887679 
## iter  10 value 50.646446
## iter  20 value 49.089459
## iter  30 value 48.772241
## iter  30 value 48.772241
## iter  30 value 48.772241
## final  value 48.772241 
## converged
## # weights:  34
## initial  value 539.285830 
## iter  10 value 72.414357
## iter  20 value 44.581740
## iter  30 value 40.041652
## iter  40 value 39.144273
## iter  50 value 38.875064
## iter  60 value 38.797750
## iter  70 value 38.788990
## iter  80 value 38.780930
## final  value 38.780763 
## converged
## # weights:  56
## initial  value 300.532220 
## iter  10 value 93.758304
## iter  20 value 46.410806
## iter  30 value 38.930739
## iter  40 value 37.801074
## iter  50 value 37.469783
## iter  60 value 37.373706
## iter  70 value 37.369824
## final  value 37.369791 
## converged
## # weights:  12
## initial  value 342.181028 
## iter  10 value 40.288158
## iter  20 value 37.523293
## iter  30 value 37.249012
## iter  40 value 37.119649
## iter  50 value 36.475494
## iter  60 value 36.287119
## iter  70 value 35.941428
## iter  80 value 35.813799
## iter  90 value 35.812278
## iter 100 value 35.809003
## final  value 35.809003 
## stopped after 100 iterations
## # weights:  34
## initial  value 379.266270 
## iter  10 value 43.021067
## iter  20 value 38.833754
## iter  30 value 35.832750
## iter  40 value 32.652108
## iter  50 value 32.441961
## iter  60 value 32.066898
## iter  70 value 31.783041
## iter  80 value 31.566600
## iter  90 value 31.468979
## iter 100 value 31.424753
## final  value 31.424753 
## stopped after 100 iterations
## # weights:  56
## initial  value 369.470832 
## iter  10 value 35.821117
## iter  20 value 15.908660
## iter  30 value 8.328573
## iter  40 value 3.877744
## iter  50 value 1.903260
## iter  60 value 1.023840
## iter  70 value 0.898991
## iter  80 value 0.836879
## iter  90 value 0.751292
## iter 100 value 0.685881
## final  value 0.685881 
## stopped after 100 iterations
## # weights:  12
## initial  value 321.325968 
## iter  10 value 41.748205
## iter  20 value 36.690287
## iter  30 value 35.466382
## iter  40 value 34.924324
## iter  50 value 34.879107
## iter  60 value 34.873452
## iter  70 value 34.871806
## iter  80 value 34.870474
## iter  90 value 34.869605
## iter 100 value 34.868966
## final  value 34.868966 
## stopped after 100 iterations
## # weights:  34
## initial  value 389.665964 
## iter  10 value 29.388989
## iter  20 value 20.570303
## iter  30 value 11.034016
## iter  40 value 9.374116
## iter  50 value 9.301317
## iter  60 value 9.298310
## final  value 9.298307 
## converged
## # weights:  56
## initial  value 266.877699 
## iter  10 value 23.487453
## iter  20 value 12.065476
## iter  30 value 9.874584
## iter  40 value 7.267801
## iter  50 value 7.086451
## iter  60 value 7.028914
## iter  70 value 7.005241
## iter  80 value 6.991960
## iter  90 value 6.987269
## iter 100 value 6.982161
## final  value 6.982161 
## stopped after 100 iterations
## # weights:  12
## initial  value 320.534221 
## iter  10 value 63.421444
## iter  20 value 45.728887
## iter  30 value 44.538101
## iter  40 value 44.535136
## iter  40 value 44.535136
## iter  40 value 44.535136
## final  value 44.535136 
## converged
## # weights:  34
## initial  value 407.770635 
## iter  10 value 54.542195
## iter  20 value 34.350959
## iter  30 value 34.057233
## iter  40 value 34.053718
## final  value 34.053086 
## converged
## # weights:  56
## initial  value 454.249533 
## iter  10 value 70.206337
## iter  20 value 34.001490
## iter  30 value 31.528974
## iter  40 value 31.231718
## iter  50 value 30.920129
## iter  60 value 30.750788
## iter  70 value 30.678612
## iter  80 value 30.673839
## final  value 30.673837 
## converged
## # weights:  12
## initial  value 400.898661 
## iter  10 value 33.253193
## iter  20 value 32.914174
## iter  30 value 32.889662
## iter  40 value 32.884246
## iter  50 value 32.879964
## iter  60 value 32.874967
## iter  70 value 32.873948
## iter  80 value 32.873462
## iter  90 value 32.873192
## iter 100 value 32.873123
## final  value 32.873123 
## stopped after 100 iterations
## # weights:  34
## initial  value 382.556717 
## iter  10 value 36.280934
## iter  20 value 17.930640
## iter  30 value 15.813059
## iter  40 value 13.913675
## iter  50 value 6.643322
## iter  60 value 6.165713
## iter  70 value 6.140107
## iter  80 value 6.128952
## iter  90 value 6.096947
## iter 100 value 6.079612
## final  value 6.079612 
## stopped after 100 iterations
## # weights:  56
## initial  value 297.748314 
## iter  10 value 28.861229
## iter  20 value 14.571782
## iter  30 value 9.282334
## iter  40 value 7.902590
## iter  50 value 7.783095
## iter  60 value 7.749937
## iter  70 value 7.659874
## iter  80 value 6.613719
## iter  90 value 6.584836
## iter 100 value 6.571720
## final  value 6.571720 
## stopped after 100 iterations
## # weights:  12
## initial  value 339.317539 
## iter  10 value 183.974334
## iter  20 value 53.549125
## iter  30 value 46.256763
## iter  40 value 42.712294
## iter  50 value 42.570629
## iter  60 value 42.551576
## iter  70 value 42.541568
## iter  80 value 42.539761
## iter  90 value 42.538272
## iter 100 value 42.535652
## final  value 42.535652 
## stopped after 100 iterations
## # weights:  34
## initial  value 396.418010 
## iter  10 value 42.095411
## iter  20 value 33.884891
## iter  30 value 27.376367
## iter  40 value 25.038883
## iter  50 value 21.657024
## iter  60 value 20.737880
## iter  70 value 17.361819
## iter  80 value 14.592029
## iter  90 value 12.616978
## iter 100 value 12.189204
## final  value 12.189204 
## stopped after 100 iterations
## # weights:  56
## initial  value 325.357705 
## iter  10 value 34.777626
## iter  20 value 22.528115
## iter  30 value 15.106759
## iter  40 value 13.386605
## iter  50 value 13.367140
## iter  60 value 13.366071
## final  value 13.366069 
## converged
## # weights:  12
## initial  value 328.046979 
## iter  10 value 63.438203
## iter  20 value 49.793236
## iter  30 value 49.132896
## final  value 49.110519 
## converged
## # weights:  34
## initial  value 323.935299 
## iter  10 value 46.209800
## iter  20 value 40.752360
## iter  30 value 39.742886
## iter  40 value 39.518575
## iter  50 value 39.442077
## final  value 39.442069 
## converged
## # weights:  56
## initial  value 449.071180 
## iter  10 value 81.019611
## iter  20 value 47.380764
## iter  30 value 43.872064
## iter  40 value 41.699791
## iter  50 value 41.065785
## iter  60 value 40.939268
## iter  70 value 40.900264
## iter  80 value 40.896946
## iter  90 value 40.896722
## iter 100 value 40.896104
## final  value 40.896104 
## stopped after 100 iterations
## # weights:  12
## initial  value 326.760562 
## iter  10 value 43.955248
## iter  20 value 36.831036
## iter  30 value 33.451374
## iter  40 value 33.386855
## iter  50 value 33.379406
## iter  60 value 33.378513
## iter  70 value 33.377686
## iter  80 value 33.376411
## iter  90 value 33.375892
## iter 100 value 33.375764
## final  value 33.375764 
## stopped after 100 iterations
## # weights:  34
## initial  value 428.856325 
## iter  10 value 44.885332
## iter  20 value 28.619876
## iter  30 value 25.215948
## iter  40 value 24.038104
## iter  50 value 23.847601
## iter  60 value 23.800001
## iter  70 value 23.777096
## iter  80 value 23.742404
## iter  90 value 23.683407
## iter 100 value 23.646538
## final  value 23.646538 
## stopped after 100 iterations
## # weights:  56
## initial  value 408.831512 
## iter  10 value 35.150293
## iter  20 value 19.962417
## iter  30 value 6.999703
## iter  40 value 6.410355
## iter  50 value 6.254994
## iter  60 value 6.161725
## iter  70 value 6.108996
## iter  80 value 6.081319
## iter  90 value 6.067144
## iter 100 value 6.056036
## final  value 6.056036 
## stopped after 100 iterations
## # weights:  12
## initial  value 321.701873 
## iter  10 value 43.680934
## iter  20 value 36.648538
## iter  30 value 36.456665
## iter  40 value 36.332539
## iter  50 value 36.318616
## iter  60 value 36.310392
## iter  70 value 36.306801
## iter  80 value 36.305603
## iter  90 value 36.305059
## final  value 36.304997 
## converged
## # weights:  34
## initial  value 300.159570 
## iter  10 value 39.055026
## iter  20 value 33.840220
## iter  30 value 30.603859
## iter  40 value 28.952140
## iter  50 value 27.795766
## iter  60 value 26.925555
## iter  70 value 25.599296
## iter  80 value 23.045157
## iter  90 value 21.495400
## iter 100 value 19.704341
## final  value 19.704341 
## stopped after 100 iterations
## # weights:  56
## initial  value 282.174286 
## iter  10 value 32.145654
## iter  20 value 18.082127
## iter  30 value 12.459379
## iter  40 value 10.960882
## iter  50 value 10.652498
## iter  60 value 10.601143
## iter  70 value 10.584815
## iter  80 value 10.582475
## iter  90 value 10.574826
## iter 100 value 10.544862
## final  value 10.544862 
## stopped after 100 iterations
## # weights:  12
## initial  value 338.491261 
## iter  10 value 57.656398
## iter  20 value 48.425059
## iter  30 value 46.955982
## iter  40 value 46.919184
## iter  40 value 46.919184
## iter  40 value 46.919184
## final  value 46.919184 
## converged
## # weights:  34
## initial  value 388.872597 
## iter  10 value 64.351406
## iter  20 value 47.266682
## iter  30 value 41.781574
## iter  40 value 38.213982
## iter  50 value 37.343387
## iter  60 value 37.030934
## iter  70 value 36.890786
## iter  80 value 36.846767
## iter  90 value 36.845837
## final  value 36.845835 
## converged
## # weights:  56
## initial  value 314.139149 
## iter  10 value 39.705645
## iter  20 value 37.016868
## iter  30 value 36.067629
## iter  40 value 35.824429
## iter  50 value 35.783243
## iter  60 value 35.587943
## iter  70 value 34.866619
## iter  80 value 34.810830
## final  value 34.809798 
## converged
## # weights:  12
## initial  value 317.804136 
## iter  10 value 55.228707
## iter  20 value 42.605093
## iter  30 value 41.648684
## iter  40 value 36.608355
## iter  50 value 36.485480
## iter  60 value 36.471912
## iter  70 value 36.464600
## iter  80 value 36.461373
## iter  90 value 36.458796
## iter 100 value 36.457700
## final  value 36.457700 
## stopped after 100 iterations
## # weights:  34
## initial  value 315.846299 
## iter  10 value 38.473415
## iter  20 value 26.848986
## iter  30 value 20.270260
## iter  40 value 18.296945
## iter  50 value 18.100182
## iter  60 value 17.907687
## iter  70 value 17.876447
## iter  80 value 17.846870
## iter  90 value 17.822413
## iter 100 value 17.767418
## final  value 17.767418 
## stopped after 100 iterations
## # weights:  56
## initial  value 335.972417 
## iter  10 value 29.936780
## iter  20 value 19.271285
## iter  30 value 13.712456
## iter  40 value 12.629135
## iter  50 value 12.481253
## iter  60 value 12.426979
## iter  70 value 12.145665
## iter  80 value 11.385390
## iter  90 value 11.080059
## iter 100 value 10.601426
## final  value 10.601426 
## stopped after 100 iterations
## # weights:  12
## initial  value 373.437219 
## iter  10 value 46.335578
## iter  20 value 44.162371
## iter  30 value 43.714313
## iter  40 value 42.541834
## iter  50 value 42.525357
## final  value 42.525324 
## converged
## # weights:  34
## initial  value 303.287895 
## iter  10 value 39.238993
## iter  20 value 36.067627
## iter  30 value 32.632827
## iter  40 value 31.610994
## iter  50 value 30.855448
## iter  60 value 30.341467
## iter  70 value 30.135302
## iter  80 value 30.035437
## iter  90 value 29.931110
## iter 100 value 29.770333
## final  value 29.770333 
## stopped after 100 iterations
## # weights:  56
## initial  value 297.621298 
## iter  10 value 34.972015
## iter  20 value 18.943944
## iter  30 value 11.586677
## iter  40 value 11.282534
## iter  50 value 11.021814
## iter  60 value 10.410503
## iter  70 value 9.605172
## iter  80 value 9.522676
## iter  90 value 8.214681
## iter 100 value 8.175033
## final  value 8.175033 
## stopped after 100 iterations
## # weights:  12
## initial  value 323.345105 
## iter  10 value 58.209433
## iter  20 value 54.315007
## iter  30 value 54.270182
## iter  30 value 54.270182
## iter  30 value 54.270182
## final  value 54.270182 
## converged
## # weights:  34
## initial  value 422.076971 
## iter  10 value 46.170339
## iter  20 value 42.350684
## iter  30 value 41.441216
## iter  40 value 41.012147
## iter  50 value 40.896813
## iter  60 value 40.881822
## iter  70 value 40.881225
## iter  80 value 40.875106
## final  value 40.874913 
## converged
## # weights:  56
## initial  value 306.668049 
## iter  10 value 60.690236
## iter  20 value 46.284930
## iter  30 value 42.735875
## iter  40 value 41.007019
## iter  50 value 40.491250
## iter  60 value 39.621355
## iter  70 value 39.060881
## iter  80 value 39.033623
## iter  90 value 39.032187
## final  value 39.032074 
## converged
## # weights:  12
## initial  value 302.761041 
## iter  10 value 83.560537
## iter  20 value 62.563210
## iter  30 value 49.839441
## iter  40 value 46.705631
## iter  50 value 46.581316
## iter  60 value 46.530918
## iter  70 value 46.373933
## iter  80 value 44.511199
## iter  90 value 43.966029
## iter 100 value 43.743383
## final  value 43.743383 
## stopped after 100 iterations
## # weights:  34
## initial  value 357.720110 
## iter  10 value 48.561801
## iter  20 value 42.404880
## iter  30 value 41.543935
## iter  40 value 39.408952
## iter  50 value 35.645359
## iter  60 value 35.145528
## iter  70 value 35.064690
## iter  80 value 34.964785
## iter  90 value 34.635062
## iter 100 value 32.106136
## final  value 32.106136 
## stopped after 100 iterations
## # weights:  56
## initial  value 401.677397 
## iter  10 value 35.902627
## iter  20 value 23.675744
## iter  30 value 11.245062
## iter  40 value 10.976112
## iter  50 value 10.822860
## iter  60 value 9.424841
## iter  70 value 8.921258
## iter  80 value 8.775222
## iter  90 value 8.483119
## iter 100 value 7.500143
## final  value 7.500143 
## stopped after 100 iterations
## # weights:  12
## initial  value 341.489758 
## iter  10 value 48.563817
## iter  20 value 39.788874
## iter  30 value 39.579703
## iter  40 value 39.551215
## iter  50 value 39.529592
## iter  60 value 39.507280
## iter  70 value 39.499314
## iter  80 value 39.496387
## iter  90 value 39.486742
## iter 100 value 39.483188
## final  value 39.483188 
## stopped after 100 iterations
## # weights:  34
## initial  value 330.979350 
## iter  10 value 33.960451
## iter  20 value 22.919974
## iter  30 value 18.884964
## iter  40 value 18.551028
## iter  50 value 18.550439
## final  value 18.550299 
## converged
## # weights:  56
## initial  value 366.530519 
## iter  10 value 31.983994
## iter  20 value 19.072783
## iter  30 value 12.356932
## iter  40 value 12.069920
## iter  50 value 11.858794
## iter  60 value 11.801794
## iter  70 value 11.561430
## iter  80 value 11.512375
## iter  90 value 11.508395
## iter 100 value 11.504984
## final  value 11.504984 
## stopped after 100 iterations
## # weights:  12
## initial  value 360.748343 
## iter  10 value 83.858557
## iter  20 value 57.325467
## iter  30 value 50.454849
## iter  40 value 48.691569
## final  value 48.690256 
## converged
## # weights:  34
## initial  value 421.251608 
## iter  10 value 79.608387
## iter  20 value 41.007630
## iter  30 value 38.956423
## iter  40 value 37.784770
## iter  50 value 37.553882
## iter  60 value 36.905560
## iter  70 value 36.859086
## final  value 36.859032 
## converged
## # weights:  56
## initial  value 303.982587 
## iter  10 value 43.509859
## iter  20 value 37.730531
## iter  30 value 36.815942
## iter  40 value 36.664768
## iter  50 value 36.515796
## iter  60 value 36.370700
## iter  70 value 36.303726
## iter  80 value 36.296592
## final  value 36.296559 
## converged
## # weights:  12
## initial  value 328.637795 
## iter  10 value 47.299042
## iter  20 value 39.924365
## iter  30 value 39.651773
## iter  40 value 39.632332
## iter  50 value 39.622021
## iter  60 value 39.619574
## iter  70 value 39.618946
## iter  80 value 39.617697
## iter  90 value 39.616645
## iter 100 value 39.616475
## final  value 39.616475 
## stopped after 100 iterations
## # weights:  34
## initial  value 420.846713 
## iter  10 value 41.417072
## iter  20 value 33.677250
## iter  30 value 29.984111
## iter  40 value 29.832642
## iter  50 value 29.811944
## iter  60 value 29.799675
## iter  70 value 29.786340
## iter  80 value 29.779071
## iter  90 value 29.773013
## iter 100 value 29.756576
## final  value 29.756576 
## stopped after 100 iterations
## # weights:  56
## initial  value 305.656107 
## iter  10 value 33.063694
## iter  20 value 28.876505
## iter  30 value 24.388957
## iter  40 value 20.162492
## iter  50 value 19.530642
## iter  60 value 19.093866
## iter  70 value 18.709888
## iter  80 value 18.638073
## iter  90 value 18.591180
## iter 100 value 18.416507
## final  value 18.416507 
## stopped after 100 iterations
## # weights:  12
## initial  value 362.923629 
## iter  10 value 45.663543
## iter  20 value 44.513763
## iter  30 value 43.151358
## iter  40 value 41.514797
## iter  50 value 39.813640
## iter  60 value 39.808652
## iter  70 value 39.808294
## final  value 39.808266 
## converged

resultado_entrenamiento5 <- predict(modelo5, entrenamiento)
resultado_prueba5 <- predict(modelo5, prueba)

#Matriz de Consufión 
mcre5 <- confusionMatrix(resultado_entrenamiento5,entrenamiento$Class) #Matriz de confusion de resultado de entrenamiento
mcre5

## Confusion Matrix and Statistics
## 
##            Reference
## Prediction  benign malignant
##   benign       346         0
##   malignant     10       192
##                                           
##                Accuracy : 0.9818          
##                  95% CI : (0.9667, 0.9912)
##     No Information Rate : 0.6496          
##     P-Value [Acc > NIR] : < 2.2e-16       
##                                           
##                   Kappa : 0.9604          
##                                           
##  Mcnemar's Test P-Value : 0.004427        
##                                           
##             Sensitivity : 0.9719          
##             Specificity : 1.0000          
##          Pos Pred Value : 1.0000          
##          Neg Pred Value : 0.9505          
##              Prevalence : 0.6496          
##          Detection Rate : 0.6314          
##    Detection Prevalence : 0.6314          
##       Balanced Accuracy : 0.9860          
##                                           
##        'Positive' Class : benign          
## 

mcrp5 <- confusionMatrix(resultado_prueba5, prueba$Class)  #Matriz de confusion de resultado de prueba
mcrp5

## Confusion Matrix and Statistics
## 
##            Reference
## Prediction  benign malignant
##   benign        85         1
##   malignant      3        46
##                                           
##                Accuracy : 0.9704          
##                  95% CI : (0.9259, 0.9919)
##     No Information Rate : 0.6519          
##     P-Value [Acc > NIR] : <2e-16          
##                                           
##                   Kappa : 0.9354          
##                                           
##  Mcnemar's Test P-Value : 0.6171          
##                                           
##             Sensitivity : 0.9659          
##             Specificity : 0.9787          
##          Pos Pred Value : 0.9884          
##          Neg Pred Value : 0.9388          
##              Prevalence : 0.6519          
##          Detection Rate : 0.6296          
##    Detection Prevalence : 0.6370          
##       Balanced Accuracy : 0.9723          
##                                           
##        'Positive' Class : benign          
## 

6. Modelo con el método rf

modelo6 <- train(Class ~ ., data=entrenamiento, 
                method = "rf", 
                preProcess = c("scale", "center"),
                trControl = trainControl(method="cv", number=10),
                tuneGrid = expand.grid(mtry=c(2,4,6))
                )

resultado_entrenamiento6 <- predict(modelo6, entrenamiento)
resultado_prueba6 <- predict(modelo6, prueba)

#Matriz de Consufión 
mcre6 <- confusionMatrix(resultado_entrenamiento6,entrenamiento$Class) #Matriz de confusion de resultado de entrenamiento
mcre6

## Confusion Matrix and Statistics
## 
##            Reference
## Prediction  benign malignant
##   benign       356         1
##   malignant      0       191
##                                      
##                Accuracy : 0.9982     
##                  95% CI : (0.9899, 1)
##     No Information Rate : 0.6496     
##     P-Value [Acc > NIR] : <2e-16     
##                                      
##                   Kappa : 0.996      
##                                      
##  Mcnemar's Test P-Value : 1          
##                                      
##             Sensitivity : 1.0000     
##             Specificity : 0.9948     
##          Pos Pred Value : 0.9972     
##          Neg Pred Value : 1.0000     
##              Prevalence : 0.6496     
##          Detection Rate : 0.6496     
##    Detection Prevalence : 0.6515     
##       Balanced Accuracy : 0.9974     
##                                      
##        'Positive' Class : benign     
## 

mcrp6 <- confusionMatrix(resultado_prueba6, prueba$Class)  #Matriz de confusion de resultado de prueba
mcrp6

## Confusion Matrix and Statistics
## 
##            Reference
## Prediction  benign malignant
##   benign        85         1
##   malignant      3        46
##                                           
##                Accuracy : 0.9704          
##                  95% CI : (0.9259, 0.9919)
##     No Information Rate : 0.6519          
##     P-Value [Acc > NIR] : <2e-16          
##                                           
##                   Kappa : 0.9354          
##                                           
##  Mcnemar's Test P-Value : 0.6171          
##                                           
##             Sensitivity : 0.9659          
##             Specificity : 0.9787          
##          Pos Pred Value : 0.9884          
##          Neg Pred Value : 0.9388          
##              Prevalence : 0.6519          
##          Detection Rate : 0.6296          
##    Detection Prevalence : 0.6370          
##       Balanced Accuracy : 0.9723          
##                                           
##        'Positive' Class : benign          
## 

Resumen de resultados

resultados <- data.frame(
  "svmLinear" = c(mcre1$overall["Accuracy"], mcrp1$overall["Accuracy"]),
  "svmRadial" = c(mcre2$overall["Accuracy"], mcrp2$overall["Accuracy"]),
  "svmPoly" = c(mcre3$overall["Accuracy"], mcrp3$overall["Accuracy"]),
  "rpart" = c(mcre4$overall["Accuracy"], mcrp4$overall["Accuracy"]),
  "nnet" = c(mcre5$overall["Accuracy"], mcrp5$overall["Accuracy"]),
  "rf" = c(mcre6$overall["Accuracy"], mcrp6$overall["Accuracy"])
)
rownames(resultados) <- c("Precision de entrenamiento", "Precision de prueba")
resultados

##                            svmLinear svmRadial   svmPoly     rpart      nnet
## Precision de entrenamiento 0.9708029 0.9963504 0.9708029 0.9635036 0.9817518
## Precision de prueba        0.9777778 0.9555556 0.9777778 0.9555556 0.9703704
##                                   rf
## Precision de entrenamiento 0.9981752
## Precision de prueba        0.9703704

Conclusión

Al analizar los resultados de los modelos de aprendizaje automático, se puede concluir que el modelo más sobresaliente sería el de NNET. Esto se debe a que su precisión durante el entrenamiento es alta, pero no alcanza el valor de 1, lo que sugiere que no hay sobreajuste. Además, su precisión en la prueba es la más alta de todos los modelos evaluados.