Figure 5 6 Original Distribution of AP_HI Parameter
Related Figures (17)
Figure: 2 Importing the Libraries and Dataset As you can see in figure 3 that there are only few entries on left most part of part of the graph, so for normalizing what we did is we eliminated the lowest 0.01 percentile data to make it a normal distribution. Figure: 3 Before & After Data Distribution of Age Parameter Figure: 4 Data Distribution of Height Parameter As you can see height parameter is evenly distributed as you can see in figure 4, so there is no need of normalizing As you can see in figure 5 that the graph is extended from the bottom part on the right, so we simply eliminated top 0.01 percentile data to make it a normal distribution. Figure: 5 Before & After Data Distribution of Weight Parameter As you can see in figure 6 that the original graph is extended from both the sides, so, we eliminated both top and lowest 0.01 percentile to achieve normal distribution as shown in figure 7. As you can see in figure 8 that the original graph is extended from right side so, we eliminated both top and 0.014 percentile to achieve normal distribution. Also, this is the last parameter for normalizing as there are no more value based parameters remaining. Figure: 8 Before and After Data Distribution of AP_LO Parameter Figure: 12 Logistic Regression Training Accuracy Code Figure 11 below shows the algorithm for implementing the logistic regression model on our dataset. Figure: 10 Logistic Regression Formula Figure: 13 Logistic Regression Testing Accuracy Code Figure: 15 Implementing Naive Bayes Classifier Model Figure: 17 Implementing Decision Tree Classifier Model e Figure 17 below shows the algorithm for implementing the Decision Tree model on our dataset. Figure: 21 Implementing Random Forest Model Figure: 19 Implementing Support Vector Machine Model e Figure 19 below shows the algorithm for implementing the SVM model on our dataset. Figure: 20 Random Forest Model A. Accuracy Comparison
