Machine learning allows computers to mimic human behaviour by training them with historical and predicted data. This section will look at some of the most exciting machine learning algorithms, such as Adaptive Resonance Theory, Self-Organizing Maps, and Survival Analysis.

Adaptive Resonance Theory

Stephen Grossberg and Gail Carpenter created the neural network technique known as adaptive resonance theory in 1987. Unsupervised learning technique is in ART at its core. Stephen Grossberg and Gail Carpenter developed the adaptive resonance theory (ART) to explain certain aspects of how the brain processes information. It describes various neural network models that tackle issues like pattern recognition and prediction using supervised and unsupervised learning techniques.

The underlying assumption of the ART model is that, in general, object identification and recognition result from the interaction of "top-down" observer expectations and "bottom-up" sensory data. According to the model, "top-down" expectations are represented by a memory template or prototype. So it is then with the basic features of an object as perceived by the senses. The sensed object will be regarded as a member of the expected class if this discrepancy between sensation and expectation does not go beyond a predetermined limit called the "vigilance parameter." The system thus provides a solution to the "plasticity/stability" problem, or the issue of learning new information without altering previously learned information, also known as incremental learning.

Self-Organizing Maps

The Self-Organizing Maps (SOM), also known as a Kohonen map or Kohonen network, was created in the 1980s. The Kohonen map or network is a computationally effective simplification. It is the 1950s Alan Turing morphogenesis and biological models of neural systems from the 1970s. It is a type of Artificial Neural Network inspired by biological models of 1970s neural systems. 

SOM is an unsupervised machine learning technique. It produces a low-dimensional (typically two-dimensional) representation of a higher-dimensional data set while preserving its topological structure. In addition, it employs an unsupervised learning strategy and a competitive learning algorithm to train its network. SOM is for clustering and mapping (or dimensionality reduction) techniques. It maps multidimensional data onto lower-dimensional data, simplifying the interpretation of complex problems.

Survival analysis

Survival analysis is a branch of statistics that looks at how long it is likely to be until something like a person dies or a machine breaks down. In engineering, this is called reliability theory or reliability analysis. In economics, it is called duration analysis or modelling; in sociology, it is called event history analysis.

In a broader sense, survival analysis entails modelling time-to-event data; in this context, failure or death are "events" in the literature on survival analysis; historically, only one event occurs for each subject, following which the organism or mechanism is dead or broken. The assumption is relaxed in models of repeated or recurring events. Nevertheless, in many fields of the social sciences and medical research, as well as system reliability, the study of regular events is essential.