Select Lab Publications

Efficient Multicore Collaborative Filtering (2011)

By: Yao Wu, Qiang Yan, Danny Bickson, Yucheng Low, and Qing Yang

Abstract: This paper describes the solution method taken by LeBu- SiShu team for track1 in ACM KDD CUP 2011 contest (re- sulting in the 5th place). We identified two main challenges: the unique item taxonomy characteristics as well as the large data set size. To handle the item taxonomy, we present a novel method called Matrix Factorization Item Taxonomy Regularization (MFITR). MFITR obtained the 2nd best prediction result out of more then ten implemented algorithms. For rapidly computing multiple solutions of various algo- rithms, we have implemented an open source parallel col- laborative filtering library on top of the GraphLab machine learning framework. We report some preliminary perfor- mance results obtained using the BlackLight supercomputer

Download Information
Yao Wu, Qiang Yan, Danny Bickson, Yucheng Low, and Qing Yang (2011). "Efficient Multicore Collaborative Filtering." ACM KDD CUP workshop. pdf            
BibTeX citation

Author = {Yao Wu and Qiang Yan and Danny Bickson and Yucheng Low and Qing Yang},
booktitle = {{ACM KDD CUP workshop}},
title = {Efficient Multicore Collaborative Filtering},
year = 2011,
wwwfilebase = {kdd2011-cup-workshop-bickson},
wwwtopic = {Collaborative Filtering},

full list