<p>Designing algorithms to recommend items such as news articles and movies to users is a challenging task in numerous web applications. The crux of the problem is to rank items based on users' responses to different items to optimize for multiple objectives. Major technical challenges are high dimensional prediction with sparse data and constructing high dimensional sequential designs to collect data for user modeling and system design. This comprehensive treatment of the statistical issues that arise in recommender systems includes detailed, in-depth discussions of current state-of-the-art methods such as adaptive sequential designs (multi-armed bandit methods), bilinear random-effects models (matrix factorization) and scalable model fitting using modern computing paradigms like MapReduce. The authors draw upon their vast experience working with such large-scale systems at Yahoo! and LinkedIn, and bridge the gap between theory and practice by illustrating complex concepts with examples from applications they are directly involved with.</p>

Statistical Methods for Recommender Systems

Computers & Technology

Agarwal, Deepak K.

Chen, Bee-Chung

Computer Science

Cambridge University Press

This monograph is dedicated to those who have something to hide. It is a monograph about "privacy-preserving data publishing" - the art of publishing sensitive personal data, collected from a group of individuals, in a form that does not violate their privacy. This problem has numerous and diverse areas of application, including releasing Census data, search logs, medical records, and interactions on a social network. The purpose of this monograph is to provide a detailed overview of the current state of the art as well as open challenges, focusing particular attention on four key themes: • RIGOROUS PRIVACY POLICIES: Repeated and highly-publicized attacks on published data have demonstrated that simplistic approaches to data publishing do not work. Significant recent advances have exposed the shortcomings of naive (and not-so-naive) techniques. They have also led to the development of mathematically rigorous definitions of privacy that publishing techniques must satisfy. • METRICS FOR DATA UTILITY: While it is necessary to enforce stringent privacy policies, it is equally important to ensure that the published version of the data is useful for its intended purpose. The authors provide an overview of diverse approaches to measuring data utility. • ENFORCEMENT MECHANISMS: This monograph describes in detail various key data publishing mechanisms that guarantee privacy and utility. • EMERGING APPLICATIONS: The problem of privacy-preserving data publishing arises in diverse application domains with unique privacy and utility requirements. The authors elaborate on the merits and limitations of existing solutions, based on which we expect to see many advances in years to come.

Privacy-Preserving Data Publishing

Kifer, Daniel

LeFevre, Kristen

Software

Now Publishers Inc

See the best price to sell, buy, or rent books by Chen, Bee-Chung. BookScouter helps to compare book prices from 25+ online bookstores and 30+ buyback vendors with a single search.

Books by Chen, Bee-Chung