Predicting Malicious Behavior: Tools and Techniques for Ensuring Global Security

Chapter 16

Developing AuBA Applications

The previous chapter focused on AuBA tools such as ThemeMate and AutoAnalyzer. Whether we use an automated assist for the manual method or use the tools to fully automate the process, predictive engines will be the result. I call them predictive engines because, much like a car, they are the engines that provide the power to make the vehicle go. In a predictive application there are a number of features that support the predictive engine. For example, we need a graphical user interface (GUI) that allows us to interact with the predictive engine, a method for entering input data, and a reporting feature that provides output and interpretation. Of course, there are also standard features such as Save, Save As, Print, Delete, and so on.

There is a distinction between AuBA tools and AuBA applications. The tools are used to construct and validate a predictive engine—the heart of all AuBA applications. The application is a finished set of features that is user-friendly and allows us to use all the features. As an example, the AuBA tools described throughout this book are used to develop the predictive engines for a variety of applications. CheckMate is an application that protects networks from external hacking attacks. It contains a number of predictive engines, each serving a different predictive purpose.

Modeling from text accounts for any methodology consists of a typical manual process of reviewing all relevant documents to extract clues, ...