I don’t understand what probability of a new message means. Does this mean the probability you are looking at a new message rather than a message you have already received? How would one calculate that? Unopened/opened messages? In the previous mission, the denominator was p(secret), which makes sense; the probability that a message has the word secret. My explanation doesn’t sit comfortably with me.
Probability of a new message is probably the probability of recent unopened message.
Here is the main concern if it’s a spam or not.
Reading the course, I understand too the concern with
new message: not clear how it’s calculated.
I guess @raisa.jerin.sristy79
new message is the probability that the message is coming from an unknown sender?
I’m not sure about that too, but I think it’s better to concentrate on rest of the course.
I know when we call into questions about something, we can’t move from there until we get the clarification, but sometimes it can resist our progress.
Hope you’ll find out the solution very soon.
I’m sure you probably figured it out as you went further in the mission, but this gave me trouble with thinking through the examples until I figured it out myself. When they use New Message in the algorithm, it is a generic stand-in for any characteristic that is being used to determine the probability of the message being spam. So New Message could stand for the message containing the word “secret” or any other characteristic that is used to determine if the message is spam.
I hope this helps anyone else who is confused by this. This is one of the very few times I feel a mission could be worded more clearly.
Hey, everybody, an answer to this question is already here: I'm having trouble putting some parameters in the naive Bayes algorithm into comprehensive terms