On Cognitive Computing (January 2017)

Since my article entitled “Cognitive Computing and Software Development Automation” appeared in the Cutter Business Technology Journal in October 2016, many people have asked me how do cognitive computing innovations differ from artificial intelligence (AI).   The simple answer is that cognitive computing provides its users with many more capabilities.  For example, use of AI was governed by sets of rules which had to be laid down beforehand in order for the system to operate effectively.  To optimize during operations, these rules had to be updated as performance data had to be gathered and analyzed using a sequential process.  In contrast, cognitive computing optimizes performance automatically by analyzing data continuously via its machine understanding and learning algorithms.  In addition, the technology is able to make better sense of the data by searching of “Internet of Everything” via cloud-based servers using natural language techniques for context sensitive data.  In other words, cognitive computing solves problems by learning and adapting rather than with rules once an initial solution is put into place.  This technology is exactly what you need to perform tasks that build knowledge based on feedback gathered as the software learns what works best under a set of variable conditions.  For example, you can improve the effectiveness and efficiency of your test cases based on the knowledge you learn as you execute your test cases.  For insights in to cognitive computing, see our report on the topic which is available via the products tab on this web site.