Page 88 - Big Data Analytics for Intelligent Healthcare Management

4.6 RESULT AND ANALYSIS 81

98
96
94 Accuracy
Precision
92
Recall
90 F1score
88
86
Logistic regression Support vector K-NN
FIG. 4.26
Performance of ResNet50 with three different classifiers for 400 .

Table 4.20 Result of Inception V3 with LR, SVM, and K-NN on 400×

Feature Extractor Classifier Accuracy (%) Precision (%) Recall (%) F1-Score (%)
Inception V3 LR 91.21 91.60 96.00 93.75
SVM 92.86 91.79 98.40 94.98
K-NN 89.01 89.02 95.53 92.16

100
98
96
Accuracy
94
92 Precision
90 Recall
88 F1score
86
84
Logistic regression Support vector K-NN

FIG. 4.27
Performance of InceptionV3 with three different classifiers for 400 .

4.6.3.7 Overall performance on 400×
Interpretation: Resnet50 with the Support Vector classifier and InceptionV3 with the Support Vector
classifier had the highest accuracy but InceptionV3 with the Support Vector classifier gave the highest
recall value and ResNet-50 with the Support Vector gave the highest precision (Fig. 4.30).