Inverse sets [1] defined in 2011 and, independently,
in [2] (2012) provide a concise mathematical description of
indexing methods that deal efficiently with vast amounts of data.
On the other hand, traditional, non-indexing machine learning
methods are far more computationally demanding. For instance,
aircraft engine prediction maintenance needs analyzing statistical
records of hundreds of operational cycles of thousands of engines
- a processing, which takes Microsoft’s cloud technologies and
unspecified machine learning time. However, it was found that
inverse sets-based approaches require only a PC or a smartphone
platform providing practically instantaneous learning and
prediction time.