Machine Learning for Hackers

Machine Learning for Hackers......Page 2
 Machine Learning for Hackers......Page 3
 How This Book Is Organized......Page 4
 Conventions Used in This Book......Page 7
 Using Code Examples......Page 8
 Safari® Books Online......Page 9
 How to Contact Us......Page 10
 Acknowledgements......Page 11
1. Using R......Page 12
 R for Machine Learning......Page 13
  Downloading and Installing R......Page 15
   Mac OS X......Page 16
  IDEs and Text Editors......Page 18
  Loading and Installing R Packages......Page 19
   Loading libraries and the data......Page 22
   Converting date strings and dealing with malformed data......Page 24
   Organizing location data......Page 25
   Dealing with data outside our scope......Page 26
   Aggregating and organizing the data......Page 27
   Analyzing the data......Page 32
  Further Reading on R......Page 35
 Exploration versus Confirmation......Page 37
 What Is Data?......Page 39
 Inferring the Types of Columns in Your Data......Page 44
 Inferring Meaning......Page 47
 Numeric Summaries......Page 48
 Means, Medians, and Modes......Page 49
 Quantiles......Page 51
 Standard Deviations and Variances......Page 52
 Exploratory Data Visualization......Page 55
 Visualizing the Relationships Between Columns......Page 78
 This or That: Binary Classification......Page 87
 Moving Gently into Conditional Probability......Page 92
 Writing Our First Bayesian Spam Classifier......Page 93
  Defining the Classifier and Testing It with Hard Ham......Page 98
  Testing the Classifier Against All Email Types......Page 102
  Improving the Results......Page 105
 How Do You Sort Something When You Don’t Know the Order?......Page 107
  Priority Features of Email......Page 109
  Functions for Extracting the Feature Set......Page 113
  Creating a Weighting Scheme for Ranking......Page 120
   A log-weighting scheme......Page 122
  Weighting from Email Thread Activity......Page 125
  Training and Testing the Ranker......Page 129
  The Baseline Model......Page 138
  Regression Using Dummy Variables......Page 143
  Linear Regression in a Nutshell......Page 144
 Predicting Web Traffic......Page 153
 Defining Correlation......Page 164
 Nonlinear Relationships Between Columns: Beyond Straight Lines......Page 166
  Introducing Polynomial Regression......Page 169
 Methods for Preventing Overfitting......Page 178
  Preventing Overfitting with Regularization......Page 182
 Text Regression......Page 188
  Logistic Regression to the Rescue......Page 191
 Introduction to Optimization......Page 196
 Ridge Regression......Page 203
 Code Breaking as Optimization......Page 208
 Unsupervised Learning......Page 218
  A Brief Introduction to Distance Metrics and Multidirectional Scaling......Page 228
 How Do US Senators Cluster?......Page 236
  Analyzing US Senator Roll Call Data (101st–111th Congresses)......Page 237
   Exploring senator MDS clustering by Congress......Page 240
 The k-Nearest Neighbors Algorithm......Page 246
 R Package Installation Data......Page 253
 Social Network Analysis......Page 256
  Thinking Graphically......Page 258
 Hacking Twitter Social Graph Data......Page 261
  Working with the Google SocialGraph API......Page 263
 Analyzing Twitter Networks......Page 268
  Local Community Structure......Page 269
  Visualizing the Clustered Twitter Network with Gephi......Page 272
  Building Your Own “Who to Follow” Engine......Page 281
 SVMs: The Support Vector Machine......Page 287
 Comparing Algorithms......Page 300
 Articles......Page 305
Index......Page 307
About the Authors......Page 338
Colophon......Page 339