AdTracking_Fraud_Detection/modelTrainingSave.R at master · WPI-DS502-FinalProject/AdTracking_Fraud_Detection · GitHub

1
2
3
4
5
6
7
8
9
10
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
library(leaps)
library(MASS)
library(class)

#Constants
BRANCH_TOTAL = 7 #Number of branches
MODEL_TOTAL = 5  #Number of models per branch
PER = 0.8        #Percent for training/testing set

extractFeature <- function(origData){
  origData$attributed_time <- NULL
  origData$hour <- as.numeric(format(as.POSIXct(origData$click_time) ,format = "%H"))
  origData$ip_app <- (origData$ip + 1) * origData$app
  origData$channel_app <- (origData$channel + 1) * origData$app
  origData$channel_ip <- (origData$channel + 1) * origData$ip
  origData$channel_ip_app <- (origData$channel_ip + 1) * origData$app
  origData$click_time <- NULL
  return (origData)
}

#Load Data
origData=read.csv(file=paste("./data/t1p60_subsamples/sub_", BEST_FILE,".csv", sep=""))
names(origData) = c('ip', 'app', 'device', 'os', 'channel','click_time', 'attributed_time', 'is_attributed')
origData <- extractFeature(origData)

training_models=list()

for(branch_num in c(1:BRANCH_TOTAL)){
  #Create Branches:
  branch_data = switch(
    branch_num,
    origData[((origData$app< 18.5) & (origData$channel< 114.5) & (origData$channel< 112)),],
    origData[((origData$app< 18.5) & (origData$channel< 114.5) & (origData$channel>=112)),],
    origData[((origData$app< 18.5) & (origData$channel>=114.5)),],
    origData[((origData$app>=18.5) & (origData$app< 19.5)),],
    origData[((origData$app>=18.5) & (origData$app>=19.5) & (origData$app< 28.5)),],
    origData[((origData$app>=18.5) & (origData$app>=19.5) & (origData$app>=28.5) & (origData$channel< 345)),],
    origData[((origData$app>=18.5) & (origData$app>=19.5) & (origData$app>=28.5) & (origData$channel>=345)),])

  #Split data: training and testing set
  branch_index = sample(floor(nrow(branch_data) * PER))
  branch_train = branch_data[branch_index,]
  branch_test  = branch_data[-branch_index,]

  #Logistic Regression:
  #Logistic Regression: Assign Predictor
  pred=pred_per_model_table[(pred_per_model_table$model=="Logistic Regression")&(pred_per_model_table$file==BEST_FILE)&(pred_per_model_table$branch==branch_num),]$predictors

  msg=sprintf("Training and testing -> Branch: %d -> Predictor: %s -> Model: %s",branch_num, pred, "Logistic Regression")
  message(msg)

  #Logistic Regression: Train Model
  branch_logreg = glm(paste("is_attributed~",pred,collapse=""), data=branch_data, family = "binomial", subset = branch_index)

  #Logistic Regression: Save Model
  training_models[[(branch_num-1)*MODEL_TOTAL+1]]=branch_logreg

  #LDA:
  #LDA: Assign Predictor
  pred=pred_per_model_table[(pred_per_model_table$model=="LDA")&(pred_per_model_table$file==BEST_FILE)&(pred_per_model_table$branch==branch_num),]$predictors

  msg=sprintf("Training and testing -> Branch: %d -> Predictor: %s -> Model: %s",branch_num, pred, "LDA")
  message(msg)

  #LDA: Train Model
  branch_LDA=lda(as.formula(paste("is_attributed~",pred,collapse="")), data=branch_data, subset=branch_index)

  #LDA: Save Result
  training_models[[(branch_num-1)*MODEL_TOTAL+2]]=branch_LDA

  #QDA:
  #QDA: Assign Predictor
  pred=pred_per_model_table[(pred_per_model_table$model=="QDA")&(pred_per_model_table$file==BEST_FILE)&(pred_per_model_table$branch==branch_num),]$predictors

  msg=sprintf("Training and testing -> Branch: %d -> Predictor: %s -> Model: %s",branch_num, pred, "QDA")
  message(msg)

  #QDA: Train Model
  branch_QDA=qda(as.formula(paste("is_attributed~",pred,collapse="")), data=branch_data, subset=branch_index)

  #QDA: Save Result
  training_models[[(branch_num-1)*MODEL_TOTAL+3]]=branch_QDA

  #SVM:
  #SVM: Assign Predictor
  pred=pred_per_model_table[(pred_per_model_table$model=="SVM")&(pred_per_model_table$file==BEST_FILE)&(pred_per_model_table$branch==branch_num),]$predictors

  msg=sprintf("Training and testing -> Branch: %d -> Predictor: %s -> Model: %s",branch_num, pred, "SVM")
  message(msg)

  #SVM: Train Model
  branch_SVM=svm(as.formula(paste("is_attributed~",pred,collapse="")), data=branch_data, subset=branch_index)
  #branch_SVM=glm(paste("is_attributed~",pred,collapse=""), data=branch_data, subset=branch_index)

  #SVM: Save Result
  training_models[[(branch_num-1)*MODEL_TOTAL+4]]=branch_SVM

  #NaiveBayes:
  #NaiveBayes: Assign Predictor
  msg=sprintf("Training and testing -> Branch: %d -> Predictor: %s -> Model: %s",branch_num, pred, "NaiveBayes")
  pred=pred_per_model_table[(pred_per_model_table$model=="NaiveBayes")&(pred_per_model_table$file==BEST_FILE)&(pred_per_model_table$branch==branch_num),]$predictors

  msg=sprintf("Training and testing -> Branch: %d -> Predictor: %s -> Model: %s",branch_num, pred, "NaiveBayes")
  message(msg)

  #NaiveBayes: Train Model
  branch_NB=naiveBayes(as.formula(paste("as.factor(is_attributed)~",pred,collapse="")), data=branch_data, subset=branch_index)

  #NaiveBayes: Save Result
  training_models[[(branch_num-1)*MODEL_TOTAL+5]]=branch_NB

  if(FALSE){
    # KNN
    pred=pred_per_model_table[(pred_per_model_table$model=="KNN 12")&(pred_per_model_table$file==BEST_FILE)&(pred_per_model_table$branch==branch_num),]$predictors

    msg=sprintf("Training and testing -> Branch: %d -> Predictor: %s -> Model: %s",branch_num, pred, "KNN 12")
    message(msg)

    #grid = expand.grid(k = c(12)) #in this case data.frame(k = c(3, 9, 12)) will do
    branch_knn = train(as.formula(paste("is_attributed~",pred,collapse="")), method= "knn",
                       data = branch_train,
                       #trControl = trainControl(method = "repeatedcv", number = 10, repeats = 3),
                       #preProcess = c("center", "scale"),
                       tuneGrid = expand.grid(k = c(12)))

    test_pred = predict(branch_knn, newdata = branch_test)
    #branch_mean=mean(test_pred == branch_test$is_attributed)
    x=confusionMatrix(test_pred, branch_test$is_attributed )

    #message(branch_mean)
    #train.Attributed=cbind(branch_data$is_attributed[branch_index])
    #predix = unlist(strsplit(as.character(pred), "+", fixed=TRUE))
    #branch_knn = knn(branch_train[predix], branch_test[predix], train.Attributed, k = 12)

    #KNN: Save Result
    training_models[[(branch_num-1)*MODEL_TOTAL+4]]=branch_knn
  }
}