https://informationcorners.com/ml-001-name-text-gender-predictor-classifier/

Evaluation of the model used.md

This manual mode where you can test this predicition model with runtime names.

def model_evaluation(classifier):
    print('<<<  Testing Module   >>> ')
    print('Enter "q" or "quit" to end testing module')
    while 1:
        test_name = input('\n Enter name to classify: ')
        if test_name.lower() == 'q' or test_name.lower() == 'quit':
            print('End')
            exit(1)
        if not test_name:
            continue

        result = classifier.classify(extract_feature(test_name))
        if gender_map[result.upper()]:
            result = 'Female'
        else:
            result = 'Male'
        print('{} is classified as {}'.format(test_name, result))

The final output window - We able to acheive accuracy around 97% (close). By choosing more efficient features we can improve the accuracy level further.

Dataset Overview.
 Total names - 96223 
 Total males names - 35323 
 Total female names - 60900
Feature matrix shape -  (96223, 7)

Naive Bayes Accuracy- 0.9688230709275136

Most informative features
	 last_3 = LIA 
	 last_3 = INA 
	 last_3 = RAH 
	 last_3 = SHA 
	 last_3 = SIA 
     
<<<  Testing Module   >>> 

Enter "q" or "quit" to end testing module

 Enter name to classify: Vijay
Vijay is classified as Male

 Enter name to classify: Umaa
Umaa is classified as Female

 Enter name to classify: Kala
Kala is classified as Female

 Enter name to classify: Ashok
Ashok is classified as Male

 Enter name to classify: John
John is classified as Male

 Enter name to classify: Vimlesh
Vimlesh is classified as Male