• Corpus ID: 51822041

Netflix Prize and SVD

@inproceedings{Gower2014NetflixPA,
  title={Netflix Prize and SVD},
  author={Stephen M. Gower},
  year={2014}
}
Singular Value Decompositions (SVD) have become very popular in the field of Collaborative Filtering. The winning entry for the famed Netflix Prize had a number of SVD models including SVD++ blended with Restricted Boltzmann Machines. Using these methods they achieved a 10 percent increase in accuracy over Netflix’s existing algorithm. In this paper I explore the different facets of a successful recommender model. I also will explore a few of the more prominent SVD based models such as… 

Figures from this paper

Novel GPU-Based Approach for Matrix Factorization using Stochastic Gradient Descent
TLDR
This research proposes a novel GPU approach for parallelizing Stochastic Gradient Descent, which is more efficient than recent parallel approaches because it utilizes GPU, reducing non-coalesced access of global memory and achieving load balance of threads.
Recommender Systems in Light of Big Data
TLDR
This implementation is intended to validate the applicability of, existing contributions to the field of, SVD-based RSs as well as validated the effectiveness of Hadoop and spark in developing large-scale systems.
Embedding and latent variable models using maximal correlation
TLDR
This work explores using maximal correlation and the alternating conditional expectation algorithm to construct embeddings one dimensional at a time to maximally preserve the linear correlation in the embedding space to induce informative soft clustering and mixture models.
ContextMF : A Fast and Context-aware Embedding Learning Method for Recommendation Systems
TLDR
ContextMF is proposed, a novel linear embedding method for sparse contextual features, thus, allowing us to quickly convert these sparse features to latent vector space while still preserving the comparable embedding quality to neural network embedding models.
Mining place-time affinity to improve POI recommendation
TLDR
A novel transfer learning model is proposed to learn affinity between the time and places, and the mined features are used to improve the performance of a content-based POI recommendation system.
An Ontology-Based Recommender System with an Application to the Star Trek Television Franchise
TLDR
An ontology-based recommender system that integrates the knowledge represented in a large ontology of literary themes to produce fiction content recommendations and is found to be a convenient alternative to collaborative filtering approaches for collections of mostly similar items, particularly when other content-based approaches are not applicable or otherwise unavailable.
Media content personalisation brokerage platform
TLDR
This thesis tackles the problem of near real time media content personalisation and designed a platform meeting the above mentioned requirements, focussed on the three main designed components: a Profiling Service; a Recommendation Service; and a Multi-Agent System (MAS) Brokerage Platform.
Predicting Drugs Adverse Side-Effects Using a Recommender-System
TLDR
This work proposes using prior information on existing interactions through recommendation systems algorithms to pinpoint possible ADRs during the drug development process, using data from the ADReCS database with promising results.
Temporally Stable Clusters of Movie Series : A Machine Learning Approach to Content Segmentation
TLDR
Adaptive evolutionary spectral clustering is a state-of-the-art method to obtain temporally stableClustering techniques have been shown to provide insight in various domains and applications.
New Exploration to Identify the Naive T Cell-Specific Gene Using Gene Similarity
TLDR
T-SNE using gene similarity calculation together with the selection of the candidate genes encoding the proteins is the most effective strategy to identify the novel genes specific to the target cells.
...
1
2
...

References

SHOWING 1-10 OF 10 REFERENCES
Improving regularized singular value decomposition for collaborative filtering
TLDR
Different efficient collaborative filtering techniques and a framework for combining them to obtain a good prediction are described, predicting users’ preferences for movies with error rate 7.04% better on the Netflix Prize dataset than the reference algorithm Netflix Cinematch.
Restricted Boltzmann machines for collaborative filtering
TLDR
This paper shows how a class of two-layer undirected graphical models, called Restricted Boltzmann Machines (RBM's), can be used to model tabular data, such as user's ratings of movies, and demonstrates that RBM's can be successfully applied to the Netflix data set.
Applying SVD on Generalized Item-based Filtering
TLDR
The results show that a reduction in the dimension of the item neighborhood via SVD, either by itself or combined with the usage of relevant demographic information, is promising, since it does not only tackle some of the recorded problems of Recommender Systems, but also assists in increasing the accuracy of systems employing it.
Factorization meets the neighborhood: a multifaceted collaborative filtering model
TLDR
The factor and neighborhood models can now be smoothly merged, thereby building a more accurate combined model and a new evaluation metric is suggested, which highlights the differences among methods, based on their performance at a top-K recommendation task.
The BellKor Solution to the Netflix Grand Prize
TLDR
Part of the contribution to the “BellKor’s Pragmatic Chaos” final solution, which won the Netflix Grand Prize, is described, which improved the baseline predictors and introduced a new blending algorithm based on gradient boosted decision trees.
Eigentaste: A Constant Time Collaborative Filtering Algorithm
TLDR
This work compares Eigentaste to alternative algorithms using data from Jester, an online joke recommending system, and uses the Normalized Mean Absolute Error (NMAE) measure to compare performance of different algorithms.
The Advantage of Careful Imputation Sources in Sparse Data-Environment of Recommender Systems: Generating Improved SVD-based Recommendations
TLDR
The Advantage of Careful Imputation Sources in Sparse Data-Environment of Recommender Systems: Generating Improved SVD-based Recommendations and how to choose the best sources for these sources.
Matrix Factorization Techniques for Recommender Systems
Jahrer, The BigChaos Solution to the Netflix Grand Prize
  • AT&T Labs, New Jersey,
  • 2009
The Pragmatic Theory Solution to the Netflix Grand Prize
  • Pragmatic Theory inc,
  • 2009