Please enable JavaScript.
Coggle requires JavaScript to display documents.
Artificial Intelligence - Coggle Diagram
Artificial Intelligence
Odds and probability
Probability
The history of dealing with uncertainty
Various competing paradigms for handling uncertain and imprecise information
Fuzzy logic
the best approach for reasoning under uncertainty
Host of unknown possibilities
ranging from missing information to deliberate deception
Naive Bayes classification
Spam filters
Why we call it “naive”
a crude simplification of the process
no dependency between adjacent words
Estimating parameters
Zero means trouble
zero counts lead to zero estimates
specify the prior odds for spam
on the average half of the incoming messages are spam
To get likelihood ratios
two different probabilities for any word occurring
a machine learning technique that can be used to classify objects
determine the probabilities of the classes given a number of different observations
The Bayes rule
this particular formula is both simple and elegant as well as incredibly powerful
How odd change
encounter this information in alternative situations
Prior and posterior odds
The Bayes rule can be expressed in many forms
simplest one is in terms of odds
the odds for something happening
refers to our assessment of the odds before obtaining some new information
Likelihood ratio
the probability of the observation in case the event of interest