A Metaheuristic Approach to Map Driving Pattern for Analyzing Driver Behavior Using Big Data Analysis

Algorithm 6

Postlearning on a feed-forward network to classify the patterns.

	Input: Its features as Current Sample scs, Adapted feature Xi and Yi at K patterns
	Output: optimized class pattern
	Step 1: Start: compute the driver access pattern DIAR rate and occurrence of feature Max weight term
	Step 2: Read Apt ← DIAR. Data values and Scs. Data values
	Step 3: For each feeded layer class Pc with logical Inter Class-ReLU Hidden unit
	Step 4: Compute the hidden layer neurons weight to c as set =
	Closest pattern Pps = Closest pattern (Cset).
	For each closest pattern on the relative link from DIAR for, each pattern p
	By each similarity, features are classified Max access patterns
	Behavioral Pattern feature selection BPC =
	End
	Compute cumulative rate BPfs =
	End
	Step 5: Optimized Driver pattern class recommendation (Dpcr) = BPFs return set maximumvalues
	Stop