A Model for Trend Analysis in the Online Shopping Scenario Using Multilevel Hesitation Pattern Mining

Algorithm 1

Mining multilevel hesitated pattern algorithm.

	Input: transactional dataset, minimum threshold support, minimum threshold confidence, number of hesitation status.
	Output: hesitated patterns, hesitated association patterns
	TD: Initial Transactional Dataset
	MD: Multilevel Transactional Dataset//after transforming TD into multilevel taxonomy
	M: highest level in the concept hierarchy//input
	: store the currently processing level
	: candidate pattern of size i//
	: hesitated Patterns of size i//
	: minimal threshold support as
	//different for each level in the hierarchy
	// is the attractiveness support and is the hesitation support of an itemset.
	= minimal threshold confidence as
	// is the attractiveness confidence and is the hesitation confidence of an itemset.
	: hesitation status//
	Initialize:  = 1
	Whiledo
	begin
	//for each class at each hesitation status
	Initialize: i = 1
	Support_calculation for i-candidate patterns
	= {candidate patterns}
	= {hesitated patterns}//after comparing Support with minimum threshold support
	i = i + 1
	Whiledo
	Begin
//gen_candidate_patterns from ;//according to hesitation status
	for all pattern hp1 belongs to do
	for all pattern hp2 belongs to do
	if
	then
	Prune;
	for all CP belongs to
	for all subsets b of CP do
	if b does not belong to
	then
	= {candidate patterns}
	Calculate the support of each prune candidate patterns at each Hs.
	Where x and y are the two individual hesitated frequent patterns.
	= {hesitated patterns}
	end
	i = i + 1;
	end
	end
	P = P + 1;
	end
	Association Pattern Generation
	for all item in HP do
	Construct association
	Calculate confidence
	conf
	if confidence then
	Output