Wiki Report for ICP4 - NagaSurendraBethapudi/Python-ICP GitHub Wiki

Video Link : https://drive.google.com/file/d/1OJOhbIhw8HobeStKBkv7NfuZwepLFM6c/view?usp=sharing

Question 1 :

find the correlation between ‘survived’(target column) and ‘sex’ column for the Titanic use case in class. Do you think we should keep this feature?

Explanation:

By using below code, found correlation

Question 2 :

Implement Naïve Bayes method using scikit-learn library

Explanation:

  • By using below code, implemented naive bayes and got accuracy of 51%

  • After doing correlation accuracy was decreased to 37%

Question 3 :

Implement linear SVM method using scikit-learn library

Explanation:

  • By using below code, implemented linear SVM and got accuracy of 67.44%

  • After doing correlation accuracy was reduced to 65%

After running two models, could see below accuracies:

  • Naive Bayes accuracy = 51%
  • linear SVM accuracy = 67.44%

Could see linear SVM algorithm got more accuracy since Naive Bayes deals the output to a particular class (Classification) where as linear SVM tries to find a hyperplane which will maximises the margin and also has optimization function.