Documents

Presentation

Description
Distributed Matrix Factorization with MapReduce
Categories
Published
of 22
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Distributed Matrix Factorization with MapReduce   Agenda ã  Recommender System (RS) ã  Recommendation System Techniques ã  MapReduce ã  Matrix Factorization (MF) ã  Effect of MF on Time Complexity ã  Conclusion ã  References 15 October 20142  Literature Survey 15 October 20143 Sr. No.Title of PaperName of AuthorFindings 1  Distributed Matrix Factorization with Mapreduce Using a Series of Broadcast-Joins Sebastian Schelter, Christoph Boden, Martin Schenck, Alexander Alexandrov and Volker MarklAuthors have proposed an efficient, data-parallellow-rank matrix factorization with Alternating LeastSquares which uses a series of broadcast-joins thatcan be efficiently executed with MapReduce. 2  Comparing the Effect of Matrix Factorization Techniques in Reducing the Time Complexity for Traversing the Big Data of Recommendation System Animesh Pandeyand SiddharthShrotriyaAuthors discussed that in this fast trendingtechnological era, data is growing very fast allaround the globe. Today Big  Data”  analytics ischallenging and dependent on time complexity. Theycompared the time complexity of different MatrixFactorization model. 3  Matrix Factorization Techniques for Recommender Systems Yehuda Koren,Robert Bell andChris VolinskyAuthors have discussed various learning algorithmslike SGD and ALS. Adding biases allows theincorporation of additional information such asimplicit feedback, temporal effects, and confidencelevels for faster and scalable processing of largedatasets using the matrix factorization  Literature Survey (Cont.) 15 October 20144 Sr. No.Title of PaperName of AuthorFindings 4  Recommender Systems Handbook Yehuda Koren andRobert BellIn Chapter 5, authors have discussed the matrixfactorization techniques, which combineimplementation convenience with a relatively highaccuracy. This has made them the preferredtechnique for addressing the largest publiclyavailable dataset - the Netflix data. 5  Large-Scale Matrix Factorization using MapreduceZhengguo Sun, TaoLi and NaphtaliRisheAuthors discussed about the feasibility to factorize amillion-by-million matrix with billions of nonzeroelements on a MapReduce cluster. In this work, they presented three different matrix multiplicationimplementations and scale up three types of nonnegative matrix factorizations on MapReduce. 6  Mapreduce: Simplied Data Processing on Large Clusters Jeffrey Dean andSanjay GhemawatAuthors discussed the MapReduce as a programming model and an associatedimplementation for processing and generating largedatasets that is amenable to a broad variety of real-world tasks.
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks