id	author	title	date	pages	extension	mime	words	sentence	flesch	summary	cache	txt
american_scientific_journal-2592	Moe, Win Win; Htwe, Nang Aye Aye	Performance Comparison of Collaborative Filtering Prediction Methods on Recommendation System	2017	12	.pdf	application/pdf	4208	176	40	Different techniques for computing item-item similarities are presented (e.g., item-item correlation and cosine similarities between item vectors) and different techniques for rating prediction (e.g., weighted sum and regression model). Firstly, the double mean centering collaborative filtering is proposed that tries to capture user biases and item biases and separates them from the whole effects so as to better concentrating on user-item interaction.	cache/american_scientific_journal-2592.pdf	txt/american_scientific_journal-2592.txt