Data Science Tutorial | Creating Text Classifier Model using Naive Bayes Algorithm

In this third video text analytics in R, I've talked about modeling process using the naive bayes classifier that helps us creating a statistical text classifier model which helps classifying the data in ham or spam sms message. You will see how you can tune the parameters also and make the best use of naive bayes classifier model.

[towardsmachinelearning]

Where and when did you create "ham_or_spam_train" dataset before writing the line spam<- subset(ham_or_spam_train, type="spam") ? I didn't find "ham_or_spam_train" object.

[rajendrapandey]

Hi i am getting the following error: table(sms_test_pred, sms_test_labels)
Error in table(sms_test_pred, sms_test_labels) :
all arguments must have the same length.
Can you please tell me how to sort this

[r_pydatascience]

Would hv bn great if you did the tokenization after you split the data. Splitting the data after preprocessing is not a recommended approach.

[tableauvizwithvineet]

Hi, thanks for great learning video, I need to confirm in last part - crosstable, is 97.7% the prediction probability that ham will be ham and 96.6% is the actual prob that ham is ham?

[muralidhara]

Hey Abi,
Thank you for your time to share the videos. Need small favor. Can you please share the csv file.
Thanks in advance.

[uwerich]

Hi, can you make a video on how to apply the spam filter on unknown Mails?
So right now you know which words or combination of words are the best predictors for a spam mail. The question is which words have which probability concerning a spam or ham mail? Moreover, how can I use this Information for further prediction.

[niladrighosh]

Where and when did you create "ham_or_spam_train" dataset before writing the line spam<- subset(ham_or_spam_train, type="spam") ? I didn't find "ham_or_spam_train" object.