Paper Info
Title
Recommendation Systems in Real Applications: Algorithm and Parallel Architecture.
Abstract
Recommendation systems are popular both in business and in academia. A series of works have been reported. In this paper, we briefly introduce the background and some basic concepts of recommendation systems, especially the applications in mainstream websites, most of them built upon parallel processing systems. However, how the recommendation algorithm works in real applications? We investigate (1) the key ideas of recommendation algorithms that are being used in real applications and (2) the parallel architecture in those real recommendation systems. In addition, the performance of recommendation system for those sites are also being analyzed and compared. We also analyze their features and compare their performances. Finally, we outline the challenges and opportunities that all recommendation systems are facing. It is anticipated that the present review will deepen people's understanding of the field and hence contribute to guide the future research of recommendation systems. Our work can help people to better understand the literature and guide the future directions.
Year
DOI
Venue
2016
10.1007/978-3-319-49148-6_5
Lecture Notes in Computer Science
Keywords
DocType
Volume
Recommendation system,Real application,Parallel architecture,Google news,Netflix,Meituan,Facebook
Conference
10066
ISSN
Citations 
PageRank 
0302-9743
0
0.34
References 
Authors
28
3
Name
Order
Citations
PageRank
Mengxian Li100.68
Wenjun Jiang235624.25
Kenli Li31389124.28