Bayesian classification is based on Baye's Theorem. It is a statistical classifier that predicts class membership probabilities such as the probability that a given tuple belongs to a particular class.
P(A/B) = P(B/A) P(A)/ P(B)
- Has high accuracy and speed for large databases.
- Has minimum error rate in comparison to all other classifier.
- Bayesian Belief Networks (Graphical Method)
- Bayesian Belief Network specifies joint conditional probability distributions.
- Bayesian Networks and Probabilistic Network are known as belief network.
- It allows class conditional independencies to be defined between subsets of variables.
- It provides a graphical model of causal relationship on which learning can be performed.
- It represents a set of random variables and their conditional dependencies via a directed acyclic graph
2. Naïve Bayesian Classifier
- The Naive Bayes Classifier technique is based on the so-called Bayesian theorem and is particularly suited when the dimensionality of the inputs is high.
- It simplifies the computational complexity.
- Naïve Bayesian Classifier assumes that the effect of an attribute value on a given class is independent of the value of other attributes e. class conditional independence.
- Let D be a training set of tuples and C1, C2,.......... , Cm are their associated classes.
- Given a tuple X, the classifier will predict that X belongs to the class having highest posterior probability conditioned on X i.e. the Naïve Bayesian classifier predicts that tuple x belong to the class Ci if and only if
P(Ci/X) > P(Cj/X) for 1≤ j ≥m, j G i
i.e. P(Ci/X) = P(X/Ci)P(Ci)/ P(X) maximum
P(X) = Constant
P(Ci) => P(C1) = P(C2) =.............. = P(Cm)
So we need to maximize P(X/Ci)
Naïve assumption is class condition independence,
= P(x1/Ci) * P (x2/Ci) *.................. * P(xn/Ci)
These probabilities can be calculated from training tuples.