The Internet began as a private network connecting government, military, and academic researchers. As such, there was little need for secure protocols, encrypted packets, and hardened servers. When the creation of the World Wide Web unexpectedly ushered in the age of the commercial Internet, the network's size and subsequent rapid expansion made it impossible retroactively to apply secure mechanisms. The Internet's architects never coined terms such as spam, phishing, zombies, and spyware, but they are terms and phenomena we now encounter constantly. Programming detectors for such threats has proven difficult. Put simply, there is too much information --- too many protocols, too many layers, too many applications, and too many uses of these applications --- for anyone to make sufficient sense of it all. Ironically, given this wealth of information, there is also too little information about what is important for detecting attacks. Methods of machine learning and data mining can help build better detectors from massive amounts of complex data. Such methods can also help discover the information required to build more secure systems. For some problems in computer security, one can directly apply machine learning and data mining techniques. Other problems, both current and future, require new approaches, methods, and algorithms. This book presents research conducted in academia and industry on methods and applications of machine learning and data mining for problems in computer security and will be of interest to researchers and practitioners, as well students. ‘Dr. Maloof not only did a masterful job of focusing the book on a critical area that was in dire need of research, but he also strategically picked papers that complemented each other in a productive manner. … This book is a must read for anyone interested in how research can improve computer security.’ Dr Eric Cole, Computer Security Expert.