A Feature-Driven Decision Support System for Heart Failure Prediction Based on Statistical Model and Gaussian Naive Bayes

Table 2

Table to compute test score.

	Positive class	Negative class	Total
Feature occurs	α	β
Feature does not occur	λ	γ
Total			τ

The sum of instances comprising feature is denoted by μ, the sum of instances without feature is denoted by , the sum of instances that are positive is expressed as ω, and the sum of instances that are negative are represented by . Let the observed values be α, β, λ, and γ with the expected values , , , and . Based on the hypothesis that the two events are independent, the expected value can be evaluated as follows: