Chapter 14

Predictive Capability in Software: Tools for a New Approach

Conducting an analysis of human behavior, whether the focus is an individual or a group, is very difficult. There are many types of analyses, ranging from the qualitative reading and processing of reports while using intuitive ability to characterize the focus to the very empirical and quantitative approaches that present objective results. An excellent analogy is investing in the stock market, the epitome of human behavior prediction. Methods range from superstition and guessing to advanced proprietary algorithms that invest and sell automatically.

There are a number of reasons why analysis of human behavior from a manual perspective has become so difficult. Some of the primary reasons are:

  • The amount of global information to process in most cases is voluminous and in some cases comes in at a fire-hose rate.
  • Because anyone can post anything on the Internet, there is a significant amount of false or misleading information embedded with accurate information.
  • In contrast, in some cases with covert attackers, the type of information needed for complete analysis is scarce or simply missing.
  • Humans have biases and often form an opinion too early, which moves their analysis toward seeking information to support their biases.
  • The idea of history repeating itself is an antiquated and uninformed concept—future behavior is not a replication of past behavior.
  • Human behavior does, however, occur in response to antecedents ...

Get Predicting Malicious Behavior: Tools and Techniques for Ensuring Global Security now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.