Practical Recommender Systems

Book Description

Summary

Online recommender systems help users find movies, jobs, restaurants--even romance! There's an art in combining statistics, demographics, and query terms to achieve results that will delight them. Learn to build a recommender system the right way: it can make or break your application!

Practical Recommender Systems explains how recommender systems work and shows how to create and apply them for your site. After covering the basics, you'll see how to collect user data and produce personalized recommendations. You'll learn how to use the most popular recommendation algorithms and see examples of them in action on sites like Amazon and Netflix. Finally, the book covers scaling problems and other issues you'll encounter as your site grows.

About the technology

Recommender systems are everywhere, helping you find everything from movies to jobs, restaurants to hospitals, even romance. Using behavioral and demographic data, these systems make predictions about what users will be most interested in at a particular time, resulting in high-quality, ordered, personalized suggestions. Recommender systems are practically a necessity for keeping your site content current, useful, and interesting to your visitors.

What's inside>/p>

  • How to collect and understand user behavior

  • Collaborative and content-based filtering

  • Machine learning algorithms

  • Real-world examples in Python

  • About the reader

    Readers need intermediate programming and database skills.

    About the author

    Kim Falk is an experienced data scientist who works daily with machine learning and recommender systems.

    Table of Contents

    1. Copyright
    2. Brief Table of Contents
    3. Table of Contents
    4. Preface
    5. Acknowledgments
    6. About this book
    7. About the author
    8. About the cover illustration
    9. Part 1. Getting ready for recommender systems
      1. Chapter 1. What is a recommender?
        1. 1.1. Real-life recommendations
        2. 1.2. Taxonomy of recommender systems
        3. 1.3. Machine learning and the Netflix Prize
        4. 1.4. The MovieGEEKs website
        5. 1.5. Building a recommender system
        6. Summary
      2. Chapter 2. User behavior and how to collect it
        1. 2.1. How (I think) Netflix gathers evidence while you browse
        2. 2.2. Finding useful user behavior
        3. 2.3. Identifying users
        4. 2.4. Getting visitor data from other sources
        5. 2.5. The collector
        6. 2.6. What users in the system are and how to model them
        7. Summary
      3. Chapter 3. Monitoring the system
        1. 3.1. Why adding a dashboard is a good idea
        2. 3.2. Doing the analytics
        3. 3.3. Personas
        4. 3.4. MovieGEEKs dashboard
        5. Summary
      4. Chapter 4. Ratings and how to calculate them
        1. 4.1. User-item preferences
        2. 4.2. Explicit or implicit ratings
        3. 4.3. Revisiting explicit ratings
        4. 4.4. What are implicit ratings?
        5. 4.5. Calculating implicit ratings
        6. 4.6. How to implement implicit ratings
        7. 4.7. Less frequent items provide more value
        8. Summary
      5. Chapter 5. Non-personalized recommendations
        1. 5.1. What’s a non-personalized recommendation?
        2. 5.2. How to make recommendations when you have no data
        3. 5.3. Implementing the chart and the groundwork for the recommender system component
        4. 5.4. Seeded recommendations
        5. Summary
      6. Chapter 6. The user (and content) who came in from the cold
        1. 6.1. What’s a cold start?
        2. 6.2. Keeping track of visitors
        3. 6.3. Addressing cold-start problems with algorithms
        4. 6.4. Those who doesn’t ask, won’t know
        5. 6.5. Using association rules to start recommending things fast
        6. Summary
    10. Part 2. Recommender algorithms
      1. Chapter 7. Finding similarities among users and among content
        1. 7.1. Why similarity?
        2. 7.2. Essential similarity functions
        3. 7.3. k-means clustering
        4. 7.4. Implementing similarities
        5. Summary
      2. Chapter 8. Collaborative filtering in the neighborhood
        1. 8.1. Collaborative filtering: A history lesson
        2. 8.2. Calculating recommendations
        3. 8.3. Calculating similarities
        4. 8.4. Amazon’s algorithm to precalculate item similarity
        5. 8.5. Ways to select the neighborhood
        6. 8.6. Finding the right neighborhood
        7. 8.7. Ways to calculate predicted ratings
        8. 8.8. Prediction with item-based filtering
        9. 8.9. Cold-start problems
        10. 8.10. A few words on machine learning terms
        11. 8.11. Collaborative filtering on the MovieGEEKs site
        12. 8.12. What’s the difference between association rule recs and collaborative recs?
        13. 8.13. Levers to fiddle with for collaborative filtering
        14. 8.14. Pros and cons of collaborative filtering
        15. Summary
      3. Chapter 9. Evaluating and testing your recommender
        1. 9.1. Business wants lift, cross-sales, up-sales, and conversions
        2. 9.2. Why is it important to evaluate?
        3. 9.3. How to interpret user behavior
        4. 9.4. What to measure
        5. 9.5. Before implementing the recommender...
        6. 9.6. Types of evaluation
        7. 9.7. Offline evaluation
        8. 9.8. Offline experiments
        9. 9.9. Implementing the experiment in MovieGEEKs
        10. 9.10. Evaluating the test set
        11. 9.11. Online evaluation
        12. 9.12. Continuous testing with exploit/explore
        13. Summary
      4. Chapter 10. Content-based filtering
        1. 10.1. Descriptive example
        2. 10.2. Content-based filtering
        3. 10.3. Content analyzer
        4. 10.4. Extracting metadata from descriptions
        5. 10.5. Finding important words with TF-IDF
        6. 10.6. Topic modeling using the LDA
        7. 10.7. Finding similar content
        8. 10.8. Creating the user profile
        9. 10.9. Content-based recommendations in MovieGEEKs
        10. 10.10. Evaluation of the content-based recommender
        11. 10.11. Pros and cons of content-based filtering
        12. Summary
      5. Chapter 11. Finding hidden genres with matrix factorization
        1. 11.1. Sometimes it’s good to reduce the amount of data
        2. 11.2. Example of what you want to solve
        3. 11.3. A whiff of linear algebra
        4. 11.4. Constructing the factorization using SVD
        5. 11.5. Constructing the factorization using Funk SVD
        6. 11.6. Doing recommendations with Funk SVD
        7. 11.7. Funk SVD implementation in MovieGEEKs
        8. 11.8. Explicit vs. implicit data
        9. 11.9. Evaluation
        10. 11.10. Levers to fiddle with for Funk SVD
        11. Summary
      6. Chapter 12. Taking the best of all algorithms: Implementing hybrid recommenders
        1. 12.1. The confused world of hybrids
        2. 12.2. The monolithic
        3. 12.3. Mixed hybrid recommender
        4. 12.4. The ensemble
        5. 12.5. Feature-weighted linear stacking (FWLS)
        6. 12.6. Implementation
        7. Summary
      7. Chapter 13. Ranking and learning to rank
        1. 13.1. Learning to rank an example at Foursquare
        2. 13.2. Re-ranking
        3. 13.3. What’s learning to rank again?
        4. 13.4. Bayesian Personalized Ranking
        5. 13.5. Implementation of BPR
        6. 13.6. Evaluation
        7. 13.7. Levers to fiddle with for BPR
        8. Summary
      8. Chapter 14. Future of recommender systems
        1. 14.1. This book in a few sentences
        2. 14.2. Topics to study next
        3. 14.3. What’s the future of recommender systems?
        4. 14.4. Final thoughts
    11. Index
    12. List of Figures
    13. List of Tables
    14. List of Listings

    Product Information

    • Title: Practical Recommender Systems
    • Author(s): Kim Falk
    • Release date: January 2019
    • Publisher(s): Manning Publications
    • ISBN: 9781617292705