Muutke küpsiste eelistusi

E-raamat: Statistical Methods for Recommender Systems

4.27/5 (29 hinnangut Goodreads-ist)
,
  • Formaat: EPUB+DRM
  • Ilmumisaeg: 24-Feb-2016
  • Kirjastus: Cambridge University Press
  • Keel: eng
  • ISBN-13: 9781316564110
Teised raamatud teemal:
  • Formaat - EPUB+DRM
  • Hind: 59,27 €*
  • * hind on lõplik, st. muud allahindlused enam ei rakendu
  • Lisa ostukorvi
  • Lisa soovinimekirja
  • See e-raamat on mõeldud ainult isiklikuks kasutamiseks. E-raamatuid ei saa tagastada.
Statistical Methods for Recommender SystemsSuurem pilt
  • Formaat: EPUB+DRM
  • Ilmumisaeg: 24-Feb-2016
  • Kirjastus: Cambridge University Press
  • Keel: eng
  • ISBN-13: 9781316564110
Teised raamatud teemal:

DRM piirangud

  • Kopeerimine (copy/paste):

    ei ole lubatud

  • Printimine:

    ei ole lubatud

  • Kasutamine:

    Digitaalõiguste kaitse (DRM)
    Kirjastus on väljastanud selle e-raamatu krüpteeritud kujul, mis tähendab, et selle lugemiseks peate installeerima spetsiaalse tarkvara. Samuti peate looma endale  Adobe ID Rohkem infot siin. E-raamatut saab lugeda 1 kasutaja ning alla laadida kuni 6'de seadmesse (kõik autoriseeritud sama Adobe ID-ga).

    Vajalik tarkvara
    Mobiilsetes seadmetes (telefon või tahvelarvuti) lugemiseks peate installeerima selle tasuta rakenduse: PocketBook Reader (iOS / Android)

    PC või Mac seadmes lugemiseks peate installima Adobe Digital Editionsi (Seeon tasuta rakendus spetsiaalselt e-raamatute lugemiseks. Seda ei tohi segamini ajada Adober Reader'iga, mis tõenäoliselt on juba teie arvutisse installeeritud )

    Seda e-raamatut ei saa lugeda Amazon Kindle's. 

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.

Arvustused

'This book provides a comprehensive guide to state-of-the-art statistical techniques that are used to power recommender systems. The text is authoritative and well written, with the authors drawing on their extensive experience of researching, implementing and evaluating real-world recommender systems. The book considers the underlying mathematics of the techniques it describes and, as such, is aimed at a readership with a strong background in statistics and cognate subjects. However, while readers without such a background are likely to find the mathematics somewhat challenging, the prose descriptions are highly readable and enable readers to understand the key principles and ideas which underpin the various approaches. This book should be of interest to those involved with recommender systems as well as to those with a broader interest in machine learning.' Patrick Hill, BCS: The Chartered Institute for IT (www.bcs.org)

Muu info

This book provides an in-depth discussion of challenges encountered in deploying real-life large-scale systems and the state-of-the-art solutions in personalization.
Preface ix
PART I INTRODUCTION
1 Introduction
3(12)
1.1 Overview of Recommender Systems for Web Applications
4(6)
1.2 A Simple Scoring Model: Most-Popular Recommendation
10(4)
Exercises
14(1)
2 Classical Methods
15(24)
2.1 Item Characterization
16(7)
2.2 User Characterization
23(2)
2.3 Feature-Based Methods
25(6)
2.4 Collaborative Filtering
31(5)
2.5 Hybrid Methods
36(1)
2.6 Summary
37(1)
Exercises
38(1)
3 Explore-Exploit for Recommender Problems
39(16)
3.1 Introduction to the Explore-Exploit Trade-off
40(1)
3.2 Multiarmed Bandit Problem
41(7)
3.3 Explore-Exploit in Recommender Systems
48(2)
3.4 Explore-Exploit with Data Sparsity
50(4)
3.5 Summary
54(1)
Exercise
54(1)
4 Evaluation Methods
55(26)
4.1 Traditional Offline Evaluation
56(10)
4.2 Online Bucket Tests
66(4)
4.3 Offline Simulation
70(3)
4.4 Offline Replay
73(4)
4.5 Summary
77(1)
Exercise
78(3)
PART II COMMON PROBLEM SETTINGS
5 Problem Settings and System Architecture
81(13)
5.1 Problem Settings
81(8)
5.2 System Architecture
89(5)
6 Most-Popular Recommendation
94(26)
6.1 Example Application: Yahoo! Today Module
95(1)
6.2 Problem Definition
96(2)
6.3 Bayesian Solution
98(9)
6.4 Non-Bayesian Solutions
107(2)
6.5 Empirical Evaluation
109(8)
6.6 Large Content Pools
117(1)
6.7 Summary
118(1)
Exercises
119(1)
7 Personalization through Feature-Based Regression
120(22)
7.1 Fast Online Bilinear Factor Model
122(4)
7.2 Offline Training
126(5)
7.3 Online Learning
131(3)
7.4 Illustration on Yahoo! Data Sets
134(7)
7.5 Summary
141(1)
Exercise
141(1)
8 Personalization through Factor Models
142(43)
8.1 Regression-Based Latent Factor Model (RLFM)
142(8)
8.2 Fitting Algorithms
150(14)
8.3 Illustration of Cold Start
164(3)
8.4 Large-Scale Recommendation of Time-Sensitive Items
167(5)
8.5 Illustration of Large-Scale Problems
172(10)
8.6 Summary
182(1)
Exercise
182(3)
PART III ADVANCED TOPICS
9 Factorization through Latent Dirichlet Allocation
185(21)
9.1 Introduction
185(1)
9.2 Model
186(5)
9.3 Training and Prediction
191(7)
9.4 Experiments
198(5)
9.5 Related Work
203(1)
9.6 Summary
204(2)
10 Context-Dependent Recommendation
206(31)
10.1 Tensor Factorization Models
207(4)
10.2 Hierarchical Shrinkage
211(7)
10.3 Multifaceted News Article Recommendation
218(15)
10.4 Related-Item Recommendation
233(2)
10.5 Summary
235(2)
11 Multiobjective Optimization
237(26)
11.1 Application Setting
238(1)
11.2 Segmented Approach
239(4)
11.3 Personalized Approach
243(5)
11.4 Approximation Methods
248(2)
11.5 Experiments
250(11)
11.6 Related Work
261(1)
11.7 Summary
262(1)
Endnotes 263(2)
References 265(8)
Index 273
Dr Deepak Agarwal is a big data analyst with more than fifteen years of experience developing and deploying state-of-the-art machine learning and statistical methods for improving the relevance of web applications. He is also experienced in conducting new scientific research to solve notoriously difficult big data problems, especially in the areas of recommender systems and computational advertising. He is a Fellow of the American Statistical Association and associate editor of two top-tier journals in statistics. Dr Bee-Chung Chen is a Senior Staff Engineer and Applied Researcher at LinkedIn. He has been a key designer of the recommendation algorithms that power LinkedIn homepage and mobile feeds, Yahoo! homepage, Yahoo! News and other sites. Dr Chen is a leading technologist with extensive industrial and research experience. His research areas include recommender systems, machine learning and big data analytics.
Lisainfo e-raamatute kohta Lisainfo e-raamatute kohta
Püsilink: https://www.kriso.ee/db/97813165641106e.html
Märksõnad: