id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_3klorapqwnd2pmpmvo24rrzmta	Timur Osadchiy	Recommender system based on pairwise association rules	2019	8	.pdf	application/pdf	8018	1119	72	We describe a recommender algorithm that builds a model of collective preferences independently of personal user interests and does not require a complex system of ratings. the algorithm is analyzed on a large transactional data set generated by a real-world dietary intake recall The algorithm calculates the likeliAlgorithm 1: Recommendations based on association rules. (2010) for our food recommendation task, which reulted in a recommender algorithm based on transactional item Recommender algorithm based on association rules applied to the example data set. bserved IF , the recommender algorithm based on pairwise assocition rules (PAR) recommends foods that are likely to be observed Algorithm 4: Training the model based on pairwise association rules. Recommender algorithm based on transactional confidence applied to the example data Recommender algorithm based on pairwise association rules applied to the example data PAR is selected to be used for the implementation of the associated foods recommender algorithm. The evaluation of the associated foods recommender algorithm	./cache/work_3klorapqwnd2pmpmvo24rrzmta.pdf	./txt/work_3klorapqwnd2pmpmvo24rrzmta.txt