In the last few years, predictive analytics has become a hot topic.
Companies are now rapidly researching this analytical method. Executives are scrambling to understand how it can help them achieve their organisational goals. These companies also fear being left behind by the competition.
Yet, there’s still a lot of confusion in the marketplace. The misunderstanding often surrounds – what exactly is Artificial Intelligence (AI)? What’s the difference between predictive analytics, machine learning and AI as a whole?
Well, let’s try and break it down into a more meaningful example.
Artificial intelligence (AI) is defined as the capability of a machine to imitate intelligent human behaviour.
To put some context to AI, machine learning and predictive analytics, I am presenting this concept using the idea of the universe, galaxies and solar systems.
In this example, AI represents the Universe. Galaxies fill up this universe. These galaxies, within this AI universe, represent the various sub-fields of AI. These sub-fields of AI focus on specific problems, approaches, the use of a particular tool, or towards satisfying particular applications.
The central problems (or goals) of AI research include reasoning, knowledge presentation, planning & scheduling, machine learning, natural language processing (communication), Computer Vision & Perception and the ability to be creative. The long term goal for is AI is General intelligence. The various galaxies on the outer layer represent these central problems.
One such galaxy is ‘Machine Learning. Machine learning is a subfield of AI that, according to Arthur Samuel in 1959, gives “computers the ability to learn without being explicitly programmed.” Machine Learning is the subset of AI which deals with the extraction of patterns from data sets. Predictive Analytics, as the name suggests, is one of the methods used in machine learning that lends itself to prediction. It’s used to devise and build complex models and algorithms.
Predictive Analytics represents a solar system within the Machine Learning Galaxy. This solar system is made up of a collection models and algorithms – represented in our example as planets.
In business, predictive models – Linear & Logistic Regression, Tree-based (Decision, Random Forest, Gradient Boosting), Neural Networks, etc. – exploit patterns found in historical and transactional data to identify risks and opportunities. These models capture relationships among many factors. These factors allow assessment of risk or potential associated with a particular set of conditions, which guides decision-making for candidate transactions.
I hope this explanation eliminates some of the confusion surrounding AI, machine learning and predictive analytics.
We would love to hear thoughts, feedback and/or suggestions, so please feel free to drop us a line in the comment section below.