Machine Learning for Hackers
Case Studies and Algorithms to Get You Started
Publisher: O'Reilly Media
Final Release Date: February 2012
Pages: 324

If you’re an experienced programmer interested in crunching data, this book will get you started with machine learning—a toolkit of algorithms that enables computers to train themselves to automate useful tasks. Authors Drew Conway and John Myles White help you understand machine learning and statistics tools through a series of hands-on case studies, instead of a traditional math-heavy presentation.

Each chapter focuses on a specific problem in machine learning, such as classification, prediction, optimization, and recommendation. Using the R programming language, you’ll learn how to analyze sample datasets and write simple machine learning algorithms. Machine Learning for Hackers is ideal for programmers from any background, including business, government, and academic research.

  • Develop a naïve Bayesian classifier to determine if an email is spam, based only on its text
  • Use linear regression to predict the number of page views for the top 1,000 websites
  • Learn optimization techniques by attempting to break a simple letter cipher
  • Compare and contrast U.S. Senators statistically, based on their voting records
  • Build a “whom to follow” recommendation system from Twitter data
Table of Contents
Product Details
About the Author
Colophon
Recommended for You
Customer Reviews

REVIEW SNAPSHOT®

by PowerReviews
oreillyMachine Learning for Hackers
 
3.6

(based on 13 reviews)

Ratings Distribution

  • 5 Stars

     

    (3)

  • 4 Stars

     

    (5)

  • 3 Stars

     

    (3)

  • 2 Stars

     

    (1)

  • 1 Stars

     

    (1)

69%

of respondents would recommend this to a friend.

Pros

  • Helpful examples (8)
  • Easy to understand (6)
  • Well-written (6)

Cons

  • Not comprehensive enough (3)

Best Uses

  • Intermediate (8)
  • Student (3)
    • Reviewer Profile:
    • Developer (7)

Reviewed by 13 customers

Displaying reviews 1-10

Back to top

Previous | Next »

 
5.0

Nice introduction for statisticans

By Dan

from London

About Me Statistician

Verified Reviewer

Pros

  • Accurate
  • Easy to understand
  • Helpful examples
  • Well-written

Cons

    Best Uses

    • Intermediate
    • Novice

    Comments about oreilly Machine Learning for Hackers:

    This was my first book on machine learning and it was the perfect start for me to start programming in R simple procedures. A key concept of the book is to explain important concepts by some examples. It does not teach you "how to ..." but provides a gentle introduction. I still prefer to write my own code instead of using caret thanks to my start with this book. This may be a bit more time consuming but I know what I am doing! I needed other books afterwards but I still recommend this as the first book to start for people with a good background in statistics.

     
    3.0

    A Good Introduction for the Beginner

    By Joe

    from West Liberty, WV

    About Me Bioinformatician, Biometrician, Biostatistician, Educator, Sys Admin

    Verified Buyer

    Pros

    • Easy to understand
    • Helpful examples

    Cons

    • Bw Photos

    Best Uses

    • Novice
    • Student

    Comments about oreilly Machine Learning for Hackers:

    This is a good introduction for the novice who is interested in removing some of the mystery around what machine learning really is. Prior experience with R would be a benefit, though isn't absolutely necessary. The most frustrating flaw of the book is that all images are in black and white, yet the book refers to the colors of data points.

     
    4.0

    Good Introduction with Examples

    By MRC01

    from Earth

    Pros

    • Covers Common Algorithms
    • Helpful examples
    • Practical Examples

    Cons

    • Glosses Over The Math
    • Not comprehensive enough

    Best Uses

    • Intermediate

    Comments about oreilly Machine Learning for Hackers:

    Targeted for people with some programming experience, this book is a good introduction to machine learning. It covers a variety of algorithms and examples.

    (17 of 19 customers found this review helpful)

     
    2.0

    There's just too much missing

    By Rob

    from Seattle, WA

    About Me Developer

    Verified Reviewer

    Pros

    • Easy to understand
    • Well-written

    Cons

    • Not comprehensive enough
    • Too basic

    Best Uses

      Comments about oreilly Machine Learning for Hackers:

      This was a pretty disappointing text.

      I'm reading this as an experienced programmer and hobbyist AI/ML/Statistics guru, and there's just too much that's missing for me to recommend this book. It reads less like "Machine Learning for Hackers" and more like "Statistics for People Who Want to Use R Without Understanding the Fundamentals." I found myself excited at the beginning of the chapter and disappointed by how little actual detail or information was provided beyond "type these commands to get numbers and hope for a good number."

      Chapter 12 ("Model Comparison") is a great example of this. While talking about SVMs, this is a snippet of what's provided:

      "As you can see from looking at Figure 12-6, the rather complicated decision boundary chosen by the sigmoid kernel wraps around as we change the value of gamma. To really get a better intuition for what's happening, we recommend that you experiment with many more values of gamma than the four we've just shown you."

      ... Really? To get a better feel for what's happening, I should just try more values for gamma? There is no mention of what, fundamentally, gamma is. The reader is supposed to just try different values and not worry about any details. This is one example, but it is a good example of how disappointed I was near the end of many chapters.

      I understand this book is targeted at beginners, but the number of times the author glosses over (or cleverly avoids) actually explaining an incredibly fundamental piece of a chapter leaves the reader wondering if the authors genuinely understand the material themselves.

      I'm giving it two stars because it's easy to read, there are decent suggestions as to which R packages are useful for statistical analysis, and because I enjoyed the overfitting examples and graphical depictions early on in the book.

      The book is not terrible, but it is lacking.

      (1 of 1 customers found this review helpful)

       
      3.0

      Basic machine learning theory

      By dahla

      from Ringsted, Denmark

      About Me Developer

      Verified Reviewer

      Pros

      • Practical Examples

      Cons

      • Not for beginners

      Best Uses

        Comments about oreilly Machine Learning for Hackers:

        I've long been fascinated by Artificial Intelligence and wanted to get started without knowing where to begin. This is why I picked up this book, thinking this would be a good starting point.
        Truth be told, this was a good book and gave some insight, but not what I was currently looking for though. So for beginners into AI this is not the starting point.
        What this book did give me though, was a brush-up on statistics, predictions and an introduction to R. Going through the book the author starts building up knowledge on how to use predictions, estimates, clustering and similar techniques in order to make a machine learn to know what to do next based on previous events. The theory is then backed up by practical using the language R.
        The one chapter I liked the most was about building a simple recommendation engine on who to follow on Twitter based on your current profile. That sample got through some graph theory combined with clustering models, all summed up with some graphical elements summing up the points going through the chapter.
        Unfortunately in the end I still felt left in the dark not knowing where to go from here. R seems like a really strong language for performing many types of statistical analysis, but I have yet to see how I should use that in some mainstream application. This is probably due to lack of knowledge on my side, but it just underlines my point about this not being a "beginners" book regarding machine learning.

        To summarize it, the author did present basic statistical models that can be used in order to aid machine learning, all this combined with practical examples. But you need to have a higher baseline and previous knowledge about machine learning and ideas about in to utilize it in order to fully enjoy this book.

         
        4.0

        Machine Learning for Hackers

        By Mary Anne

        from Portland, Oregon

        About Me Data Scientist

        Verified Reviewer

        Pros

        • Helpful examples
        • R Plyr Svm Glmnet

        Cons

          Best Uses

          • Intermediate

          Comments about oreilly Machine Learning for Hackers:

          Machine Learning for Hackers gets you started using R for machine learning. The book does a good job telling you how to install R and where to find help.
          There are lots examples on how to explore data using ggplot2. Other package covered include plyr which they equal to map reduce. tm package which is used in polynomial regression. glmnet and the Lamda function. K-Nearist neighbor algorithm which uses the class package.

          (5 of 8 customers found this review helpful)

           
          1.0

          Broken Code

          By Craig

          from Sacramento, CA

          Verified Reviewer

          Pros

            Cons

            • Too many errors

            Best Uses

              Comments about oreilly Machine Learning for Hackers:

              A book heavily focused on the results of code to illustrate concepts takes on a BIG risk of that code being or becoming broken. The UFO example should refer to Unidentified Faulty Objects. Used the online code to work through a few more steps but still ended up with errors, errors that should not occur when cutting and pasting.

              (2 of 2 customers found this review helpful)

               
              4.0

              Great book

              By Filipe X

              from Recife, Brazil

              About Me Developer, Maker

              Verified Reviewer

              Pros

              • Easy to understand

              Cons

                Best Uses

                • Intermediate
                • Student

                Comments about oreilly Machine Learning for Hackers:

                It goes through the very basics of statistics to build the necessary knowledge to the machine learning algorithms. On the other hand it doesn't explains in depth shown blocks of code, leaving to the reader to understand particularities of the R programming language. The use of R allows the easy processing of data with few lines of code, on the downside its a very different language so it requires some effort to be understood. For beginners in R, its very valuable to lookup and understand used functions to enlighten used algorithms. This book is truly made for hackers as it requires low level statistics and high level of curiosity to play with code, it also uses real word data on its examples making it even more attractive and fun.

                (11 of 11 customers found this review helpful)

                 
                3.0

                Enjoyable but light on useful detail

                By XYZ

                from Cambridge, MA

                About Me Developer

                Verified Reviewer

                Pros

                • Concise
                • Easy to understand
                • Well-written

                Cons

                • Not comprehensive enough
                • Too basic

                Best Uses

                • Intermediate

                Comments about oreilly Machine Learning for Hackers:

                (Disclosure: I received a free review copy of this book.)

                I had high hopes for this book after the first few chapters. The emphasis in the early chapters on cleaning data rings true to anyone who has ever had to deal with a body of real-world data.

                But after that it fell into a repetitive pattern: state a problem, give a nontechnical description of a machine learning algorithm, and explain how to call the appropriate ML library in R. With no math and little description of most algorithms, if you want to do something besides use R's built-in libraries, this book isn't so helpful.

                The writing style is lively and enjoyable, and the authors picked interesting real-world examples. They probably could write a really good book on machine learning, but this one isn't it.

                (0 of 8 customers found this review helpful)

                 
                4.0

                the best book to start mACHINE LEARNING

                By abhi1one

                from india

                Pros

                • Easy to understand
                • Helpful examples

                Cons

                  Best Uses

                    Comments about oreilly Machine Learning for Hackers:

                    this may be the book that helped me to start hacking AI and ML!!!

                    Displaying reviews 1-10

                    Back to top

                    Previous | Next »

                     
                    Buy 2 Get 1 Free Free Shipping Guarantee
                    Buying Options
                    Immediate Access - Go Digital what's this?
                    Ebook:  $42.99
                    Formats:  DAISY, ePub, Mobi, PDF
                    Print & Ebook:  $54.99
                    Print:  $49.99

                    Available in Multiple Languages