Skip to main content

TableĀ 3 Notation use for feature selection

From: An ensemble approach to stabilize the features for multi-domain sentiment analysis using supervised machine learning

Symbol Description
P(ci) Probability that a document d in class ci
P(f) Probability that document d contains feature f
P(\(\bar{f}\)) Probability that a document d does not contains feature f
P(ci/f) Probability that document d contains feature f in class ci
P(ci/\(\bar{f}\)) Probability that document d does not contains feature f in class ci