Matches in ScholarlyData for { <https://w3id.org/scholarlydata/inproceedings/www2010/paper/main/430> ?p ?o. }
Showing items 1 to 13 of
13
with 100 items per page.
- 430 creator chao-liu.
- 430 creator hung-chih-yang.
- 430 creator jinliang-fan.
- 430 creator li-wei-he.
- 430 creator yi-min-wang.
- 430 type InProceedings.
- 430 label "Distributed Non-negative Matrix Factorization for Web-Scale Dyadic Data Analysis on MapReduce".
- 430 sameAs 430.
- 430 abstract "Web abounds with dyadic data that keeps increasing by every single second. Previous work has repeatedly shown the usefulness of extracting the interaction structure inside dyadic data [20, 10, 9]. A commonly used tool in extracting the underlying structure is the matrix factorization, whose fame was recently boosted in the Netflix challenge [25]. When we were trying to replicate the same success on real-world Web dyadic data, we were seriously challenged by the scalability of available tools. We therefore in this paper report our efforts on scaling up the nonnegative matrix factorization (NMF) technique. We show that by carefully partitioning the data and arranging the computations to maximize data locality and parallelism, factorizing tens of millions by hundreds of millions matrices with billions of nonzero cells can be accomplished within tens of hours. Besides scalability, we also study the effectiveness of NMF on different Web dyadic data, which demonstrates the versatility of the proposed approach.".
- 430 hasAuthorList authorList.
- 430 isPartOf proceedings.
- 430 keyword "Efficient algorithms for large-scale analysis".
- 430 title "Distributed Non-negative Matrix Factorization for Web-Scale Dyadic Data Analysis on MapReduce".