Machine Learning; Compare performance of model of both before and after hyperparameter tuning data stated below. Please analyze for below data only. Not the general. Suggest the ML model that, demonstrates the most favorable performance and justify.

Before hyperparameter Tuning of KNN model:

Accuracy: $0\mathrm{.}5180,$ Precision: $0\mathrm{.}5319,$ Recall: $0\mathrm{.}4883,$ F$1-$Score: $0\mathrm{.}5092$

Confusion Matrix:

$[[134110]$

$[131125]]$

Complete Report of KNN Model:

precision recall f$1-$score support

$00\mathrm{.}510\mathrm{.}550\mathrm{.}53244$

$10\mathrm{.}530\mathrm{.}490\mathrm{.}51256$

accuracy $0\mathrm{.}52500$

macro avg $0\mathrm{.}520\mathrm{.}520\mathrm{.}52500$

weighted avg $0\mathrm{.}520\mathrm{.}520\mathrm{.}52500$

$---------------------------------------$

After hyperparameter Tuning of KNN model:

Accuracy: $0\mathrm{.}5280,$ Precision: $0\mathrm{.}5417,$ Recall: $0\mathrm{.}5078,$ F$1-$Score: $0\mathrm{.}5242$

Confusion Matrix:

$[[134110]$

$[126130]]$

Complete Report of KNN Model:

precision recall f$1-$score support

$00\mathrm{.}520\mathrm{.}550\mathrm{.}53244$

$10\mathrm{.}540\mathrm{.}510\mathrm{.}52256$

accuracy $0\mathrm{.}53500$

macro avg $0\mathrm{.}530\mathrm{.}530\mathrm{.}53500$

weighted avg $0\mathrm{.}530\mathrm{.}530\mathrm{.}53500$

$---------------------------------------$

Before hyperparameter Tuning of Naive Bayes model:

Accuracy: $0\mathrm{.}4640,$ Precision: $0\mathrm{.}4744,$ Recall: $0\mathrm{.}4336,$ F$1-$Score: $0\mathrm{.}4531$

Confusion Matrix:

$[[121123]$

$[145111]]$

Complete Report of Naive Bayes Model:

precision recall f$1-$score support

$00\mathrm{.}450\mathrm{.}500\mathrm{.}47244$

$10\mathrm{.}470\mathrm{.}430\mathrm{.}45256$

accuracy $0\mathrm{.}46500$

macro avg $0\mathrm{.}460\mathrm{.}460\mathrm{.}46500$

weighted avg $0\mathrm{.}460\mathrm{.}460\mathrm{.}46500$

$---------------------------------------$

After hyperparameter Tuning of Naive Bayes model:

Accuracy: $0\mathrm{.}4680,$ Precision: $0\mathrm{.}4728,$ Recall: $0\mathrm{.}3398,$ F$1-$Score: $0\mathrm{.}3955$

Confusion Matrix:

$[[14797]$

$[16987]]$

Complete Report of Naive Bayes Model:

precision recall f$1-$score support

$00\mathrm{.}470\mathrm{.}600\mathrm{.}53244$

$10\mathrm{.}470\mathrm{.}340\mathrm{.}40256$

accuracy $0\mathrm{.}47500$

macro avg $0\mathrm{.}470\mathrm{.}470\mathrm{.}46500$

weighted avg $0\mathrm{.}470\mathrm{.}470\mathrm{.}46500$

$---------------------------------------$

Before hyperparameter Tuning of Random Forest model:

Accuracy: $0\mathrm{.}4960,$ Precision: $0\mathrm{.}5085,$ Recall: $0\mathrm{.}4688,$ F$1-$Score: $0\mathrm{.}4878$

Confusion Matrix:

$[[128116]$

$[136120]]$

Complete Report of Random Forest Model:

precision recall f$1-$score support

$00\mathrm{.}480\mathrm{.}520\mathrm{.}50244$

$10\mathrm{.}510\mathrm{.}470\mathrm{.}49256$

accuracy $0\mathrm{.}50500$

macro avg $0\mathrm{.}500\mathrm{.}500\mathrm{.}50500$

weighted avg $0\mathrm{.}500\mathrm{.}500\mathrm{.}50500$

$---------------------------------------$

After hyperparameter Tuning of Random Forest model:

Accuracy: $0\mathrm{.}5220,$ Precision: $0\mathrm{.}5341,$ Recall: $0\mathrm{.}5195,$ F$1-$Score: $0\mathrm{.}5267$

Confusion Matrix:

$[[128116]$

$[123133]]$

Complete Report of Random Forest Model:

precision recall f$1-$score support

$00\mathrm{.}510\mathrm{.}520\mathrm{.}52244$