Diferente de um extrator de caracteristicas, nesse arquivo apenas focamos em entender se apenas contarmos com apenas alguns lags anteriores é suficiente para realizar uma classifciação de qualidade.

Basicamente, apenas pegamos os últimos momentos capturados pelo percurso e tranformamos esses momentos finais em atributos de uma instância de um possível modelo. Nessa simples abordagem é muito utilizada em regressão para previsão dos próximos passos ou valores.

Com essa abordagem podemos simplificar e deixar por conta dos classificadores extrair ou linearmente separar as classes binárias, como é o nosso caso.

Abaixo apenas consideramos os últimos 2 lags e assim fazer com que o classificador, SVM, separe cada instância.



In [12]:

    
library(caret)
library(kernlab)
library(pROC)

groups <- read.csv(file="./MovementAAL/groups/MovementAAL_DatasetGroup.csv",head=TRUE,sep=",")
targetAll <- read.csv(file="./MovementAAL/dataset/MovementAAL_target.csv",head=TRUE,sep=",")



In [2]:

    
#Group 1
allDataGroup1<-list()
allDataGroup1Target<-list()
groups1 = groups[groups$dataset_ID==1, ]

index<-1
for (id in groups1$X.sequence_ID){
    caminho <-paste("./MovementAAL/dataset/MovementAAL_RSS_",id,".csv",sep="")
    allDataGroup1[[index]]<-read.csv(file=caminho,head=TRUE,sep=",")
    allDataGroup1Target[index]<-targetAll[[2]][id]
    index<-index+1
}
wtData <- NULL
minStepsBack = 1
for (i in 1:length(allDataGroup1)){
     aa<- t(unlist(allDataGroup1[[i]][(nrow(allDataGroup1[[i]])-minStepsBack):nrow(allDataGroup1[[i]]),]))
    wtData <- rbind(wtData, aa)
}
wtData <- as.data.frame(wtData)
data = unlist(allDataGroup1Target)
target = factor(data,labels=c("No","Yes"))
frameDataFinal <- data.frame(cbind(target, wtData))
head(frameDataFinal)
##use only lagged data









    Out[2]:





target X.RSS_anchor11 X.RSS_anchor12 RSS_anchor21 RSS_anchor22 RSS_anchor31 RSS_anchor32 RSS_anchor41 RSS_anchor42

	1 Yes 0 -0.14286 0.04 -0.6 -0.047619 -0.28571 -0.05 -0.1
	2 Yes -0.33333 -0.14286 0.04 0.04 0.095238 0.14286 -0.1 0.05
	3 Yes -0.28571 -0.14286 -0.04 -0.08 -0.095238 0.14286 -0.05 0.35
	4 Yes -0.42857 -0.57143 -0.2 -0.52 0 0.047619 0.4 -0.15
	5 Yes -0.57143 -0.52381 -0.6 -0.44 0.57143 0.28571 0.9 0.8
	6 Yes -1 -0.47619 -0.32 -0.2 0.71429 0.52381 0.65 0.95

Média e Desvio padrão respectivamente.

Group 1, com Cross Validation tipo 10-fold



In [3]:

    
inTraining <- createDataPartition(frameDataFinal$target, p = .7, list = TRUE,times=10)
allAccuracyGroup1 <- c()

for( i in 1:length(inTraining)){

    training <- frameDataFinal[ inTraining[[i]],]
    testing  <- frameDataFinal[-inTraining[[i]],]
    fitControl <- trainControl(method = "none", classProbs = TRUE)

    svmLinearFit <- train(target ~ ., data = training,
                     method = "svmLinear",
                     trControl = fitControl,
                     family=binomial)
    preds<- predict(svmLinearFit, newdata = testing)
    matrix <- confusionMatrix(preds,frameDataFinal$target[-inTraining[[i]]])
    allAccuracyGroup1 <- c(allAccuracyGroup1,matrix[3]$overall[[1]])
}

mean(allAccuracyGroup1)
sd(allAccuracyGroup1)









    Out[3]:




0.673333333333333






    Out[3]:




0.0644061188719531



In [4]:

    
#Group 2
allDataGroup2<-list()
allDataGroup2Target<-list()
groups2 = groups[groups$dataset_ID==2, ]

index<-1
for (id in groups2$X.sequence_ID){
    caminho <-paste("./MovementAAL/dataset/MovementAAL_RSS_",id,".csv",sep="")
    allDataGroup2[[index]]<-read.csv(file=caminho,head=TRUE,sep=",")
    allDataGroup2Target[index]<-targetAll[[2]][id]
    index<-index+1
}
wtData <- NULL
minStepsBack = 1
for (i in 1:length(allDataGroup2)){
     aa<- t(unlist(allDataGroup2[[i]][(nrow(allDataGroup2[[i]])-minStepsBack):nrow(allDataGroup2[[i]]),]))
    wtData <- rbind(wtData, aa)
}
wtData <- as.data.frame(wtData)
data = unlist(allDataGroup2Target)
target = factor(data,labels=c("No","Yes"))
frameDataFinal <- data.frame(cbind(target, wtData))
head(frameDataFinal)
##use only lagged data









    Out[4]:





target X.RSS_anchor11 X.RSS_anchor12 RSS_anchor21 RSS_anchor22 RSS_anchor31 RSS_anchor32 RSS_anchor41 RSS_anchor42

	1 Yes 0.066667 0.066667 -0.48718 -0.17949 -0.45455 -0.36364 -0.047619 -0.14286
	2 Yes 0.86667 1 0.38462 0.84615 -0.59091 -0.81818 -0.7619 -0.7619
	3 Yes 0.73333 0.73333 0.53846 0.53846 -0.81818 -0.90909 -0.61905 -0.57143
	4 Yes 0.2 0.2 0.53846 0.23077 -0.86364 -0.72727 -1 -0.57143
	5 Yes 0.68889 0.68889 0.69231 0.38462 -0.86364 -0.90909 -0.80952 -0.61905
	6 Yes 0.64444 0.64444 0.64103 0.69231 -0.59091 -0.81818 -0.61905 -0.57143

Média e Desvio padrão respectivamente.

Group 2, com Cross Validation tipo 10-fold



In [5]:

    
inTraining <- createDataPartition(frameDataFinal$target, p = .7, list = TRUE,times=10)
allAccuracyGroup2 <- c()

for( i in 1:length(inTraining)){

    training <- frameDataFinal[ inTraining[[i]],]
    testing  <- frameDataFinal[-inTraining[[i]],]
    fitControl <- trainControl(method = "none", classProbs = TRUE)

    svmLinearFit <- train(target ~ ., data = training,
                     method = "svmLinear",
                     trControl = fitControl,
                     family=binomial)
    preds<- predict(svmLinearFit, newdata = testing)
    matrix <- confusionMatrix(preds,frameDataFinal$target[-inTraining[[i]]])
    allAccuracyGroup2 <- c(allAccuracyGroup2,matrix[3]$overall[[1]])
}

mean(allAccuracyGroup2)
sd(allAccuracyGroup2)









    Out[5]:




0.683870967741935






    Out[5]:




0.0757282445662753



In [6]:

    
#Group 3
allDataGroup3<-list()
allDataGroup3Target<-list()
groups3 = groups[groups$dataset_ID==3, ]

index<-1
for (id in groups3$X.sequence_ID){
    caminho <-paste("./MovementAAL/dataset/MovementAAL_RSS_",id,".csv",sep="")
    allDataGroup3[[index]]<-read.csv(file=caminho,head=TRUE,sep=",")
    allDataGroup3Target[index]<-targetAll[[2]][id]
    index<-index+1
}
wtData <- NULL
minStepsBack = 1
for (i in 1:length(allDataGroup3)){
     aa<- t(unlist(allDataGroup3[[i]][(nrow(allDataGroup3[[i]])-minStepsBack):nrow(allDataGroup3[[i]]),]))
    wtData <- rbind(wtData, aa)
}
wtData <- as.data.frame(wtData)
data = unlist(allDataGroup3Target)
target = factor(data,labels=c("No","Yes"))
frameDataFinal <- data.frame(cbind(target, wtData))
head(frameDataFinal)
##use only lagged data









    Out[6]:





target X.RSS_anchor11 X.RSS_anchor12 RSS_anchor21 RSS_anchor22 RSS_anchor31 RSS_anchor32 RSS_anchor41 RSS_anchor42

	1 Yes 0.090909 0.090909 0.066667 0.066667 -0.33333 -0.33333 -0.31915 -0.3617
	2 Yes 0.77273 0.77273 -0.15556 0.6 -0.82222 -1 -0.61702 -0.61702
	3 Yes 0.5 0.5 0.2 0.33333 -0.86667 -0.68889 -0.57447 -0.57447
	4 Yes 0.86364 0.86364 0.24444 -0.68889 -0.95556 -0.95556 -0.53191 -0.53191
	5 Yes 0.77273 0.86364 -0.066667 -0.11111 -0.68889 -0.64444 -0.3617 -0.3617
	6 Yes 0.31818 0.72727 0.15556 -0.11111 -0.6 -0.51111 -0.48936 -0.48936

Média e Desvio padrão respectivamente.

Group 3, com Cross Validation tipo 10-fold



In [7]:

    
inTraining <- createDataPartition(frameDataFinal$target, p = .7, list = TRUE,times=10)
allAccuracyGroup3 <- c()

for( i in 1:length(inTraining)){

    training <- frameDataFinal[ inTraining[[i]],]
    testing  <- frameDataFinal[-inTraining[[i]],]
    fitControl <- trainControl(method = "none", classProbs = TRUE)

    svmLinearFit <- train(target ~ ., data = training,
                     method = "svmLinear",
                     trControl = fitControl,
                     family=binomial)
    preds<- predict(svmLinearFit, newdata = testing)
    matrix <- confusionMatrix(preds,frameDataFinal$target[-inTraining[[i]]])
    allAccuracyGroup3 <- c(allAccuracyGroup3,matrix[3]$overall[[1]])
}

mean(allAccuracyGroup3)
sd(allAccuracyGroup3)









    Out[7]:




0.516129032258065






    Out[7]:




0.056897877657303



In [8]:

    
#All Groups
allData<-list()
allDataTarget<-list()
targetAll <- read.csv(file="./MovementAAL/dataset/MovementAAL_target.csv",head=TRUE,sep=",")

index<-1
for (id in targetAll$X.sequence_ID){
    caminho <-paste("./MovementAAL/dataset/MovementAAL_RSS_",id,".csv",sep="")
    allData[[index]]<-read.csv(file=caminho,head=TRUE,sep=",")
    allDataTarget[index]<-targetAll[[2]][id]
    index<-index+1
}
wtData <- NULL
minStepsBack = 1
for (i in 1:length(allData)){
     aa<- t(unlist(allData[[i]][(nrow(allData[[i]])-minStepsBack):nrow(allData[[i]]),]))
    wtData <- rbind(wtData, aa)
}
wtData <- as.data.frame(wtData)
data = unlist(allDataTarget)
target = factor(data,labels=c("No","Yes"))
frameDataFinal <- data.frame(cbind(target, wtData))
head(frameDataFinal)









    Out[8]:





target X.RSS_anchor11 X.RSS_anchor12 RSS_anchor21 RSS_anchor22 RSS_anchor31 RSS_anchor32 RSS_anchor41 RSS_anchor42

	1 Yes 0 -0.14286 0.04 -0.6 -0.047619 -0.28571 -0.05 -0.1
	2 Yes -0.33333 -0.14286 0.04 0.04 0.095238 0.14286 -0.1 0.05
	3 Yes -0.28571 -0.14286 -0.04 -0.08 -0.095238 0.14286 -0.05 0.35
	4 Yes -0.42857 -0.57143 -0.2 -0.52 0 0.047619 0.4 -0.15
	5 Yes -0.57143 -0.52381 -0.6 -0.44 0.57143 0.28571 0.9 0.8
	6 Yes -1 -0.47619 -0.32 -0.2 0.71429 0.52381 0.65 0.95

Média e Desvio padrão respectivamente.

Todos os Groups em uma base apenas, com Cross Validation tipo 10-fold



In [9]:

    
inTraining <- createDataPartition(frameDataFinal$target, p = .7, list = TRUE,times=10)
allAccuracy <- c()

for( i in 1:length(inTraining)){

    training <- frameDataFinal[ inTraining[[i]],]
    testing  <- frameDataFinal[-inTraining[[i]],]
    fitControl <- trainControl(method = "none", classProbs = TRUE)

    svmLinearFit <- train(target ~ ., data = training,
                     method = "svmLinear",
                     trControl = fitControl,
                     family=binomial)
    preds<- predict(svmLinearFit, newdata = testing)
    matrix <- confusionMatrix(preds,frameDataFinal$target[-inTraining[[i]]])
    allAccuracy <- c(allAccuracy,matrix[3]$overall[[1]])
}

mean(allAccuracy)
sd(allAccuracy)









    Out[9]:




0.610752688172043






    Out[9]:




0.0350448933665385

Matrix de confusão

Todos os Groups em uma base apenas



In [10]:

    
#All groups datasets Confusion Matrix 
inTraining <- createDataPartition(frameDataFinal$target, p = .7, list = TRUE,times=1)
training <- frameDataFinal[ inTraining[[1]],]
testing  <- frameDataFinal[-inTraining[[1]],]
fitControl <- trainControl(method = "none", classProbs = TRUE)

svmLinearFit <- train(target ~ ., data = training,
                     method = "svmLinear",
                     trControl = fitControl,
                     family=binomial)
preds<- predict(svmLinearFit, newdata = testing)
matrix <- confusionMatrix(preds,frameDataFinal$target[-inTraining[[1]]])
matrix









    Out[10]:





Confusion Matrix and Statistics

          Reference
Prediction No Yes
       No  31  16
       Yes 15  31
                                          
               Accuracy : 0.6667          
                 95% CI : (0.5613, 0.7611)
    No Information Rate : 0.5054          
    P-Value [Acc > NIR] : 0.001211        
                                          
                  Kappa : 0.3334          
 Mcnemar's Test P-Value : 1.000000        
                                          
            Sensitivity : 0.6739          
            Specificity : 0.6596          
         Pos Pred Value : 0.6596          
         Neg Pred Value : 0.6739          
             Prevalence : 0.4946          
         Detection Rate : 0.3333          
   Detection Prevalence : 0.5054          
      Balanced Accuracy : 0.6667          
                                          
       'Positive' Class : No

Curva ROC e AUC

Todos os Groups em uma base apenas



In [11]:

    
#ROC CURVE AND AUC
predsProb<- predict(svmLinearFit, newdata = testing,type="prob")
outcome<- predsProb[,2]
classes <- frameDataFinal$target[-inTraining[[1]]]
rocobj <- roc(classes, outcome,levels=c("No","Yes"))
plot(rocobj)









    Out[11]:





Call:
roc.default(response = classes, predictor = outcome, levels = c("No",     "Yes"))

Data: outcome in 46 controls (classes No) < 47 cases (classes Yes).
Area under the curve: 0.7068

	target	X.RSS_anchor11	X.RSS_anchor12	RSS_anchor21	RSS_anchor22	RSS_anchor31	RSS_anchor32	RSS_anchor41	RSS_anchor42
1	Yes	0	-0.14286	0.04	-0.6	-0.047619	-0.28571	-0.05	-0.1
2	Yes	-0.33333	-0.14286	0.04	0.04	0.095238	0.14286	-0.1	0.05
3	Yes	-0.28571	-0.14286	-0.04	-0.08	-0.095238	0.14286	-0.05	0.35
4	Yes	-0.42857	-0.57143	-0.2	-0.52	0	0.047619	0.4	-0.15
5	Yes	-0.57143	-0.52381	-0.6	-0.44	0.57143	0.28571	0.9	0.8
6	Yes	-1	-0.47619	-0.32	-0.2	0.71429	0.52381	0.65	0.95

	target	X.RSS_anchor11	X.RSS_anchor12	RSS_anchor21	RSS_anchor22	RSS_anchor31	RSS_anchor32	RSS_anchor41	RSS_anchor42
1	Yes	0.066667	0.066667	-0.48718	-0.17949	-0.45455	-0.36364	-0.047619	-0.14286
2	Yes	0.86667	1	0.38462	0.84615	-0.59091	-0.81818	-0.7619	-0.7619
3	Yes	0.73333	0.73333	0.53846	0.53846	-0.81818	-0.90909	-0.61905	-0.57143
4	Yes	0.2	0.2	0.53846	0.23077	-0.86364	-0.72727	-1	-0.57143
5	Yes	0.68889	0.68889	0.69231	0.38462	-0.86364	-0.90909	-0.80952	-0.61905
6	Yes	0.64444	0.64444	0.64103	0.69231	-0.59091	-0.81818	-0.61905	-0.57143

	target	X.RSS_anchor11	X.RSS_anchor12	RSS_anchor21	RSS_anchor22	RSS_anchor31	RSS_anchor32	RSS_anchor41	RSS_anchor42
1	Yes	0.090909	0.090909	0.066667	0.066667	-0.33333	-0.33333	-0.31915	-0.3617
2	Yes	0.77273	0.77273	-0.15556	0.6	-0.82222	-1	-0.61702	-0.61702
3	Yes	0.5	0.5	0.2	0.33333	-0.86667	-0.68889	-0.57447	-0.57447
4	Yes	0.86364	0.86364	0.24444	-0.68889	-0.95556	-0.95556	-0.53191	-0.53191
5	Yes	0.77273	0.86364	-0.066667	-0.11111	-0.68889	-0.64444	-0.3617	-0.3617
6	Yes	0.31818	0.72727	0.15556	-0.11111	-0.6	-0.51111	-0.48936	-0.48936