3. The Aorist:
: an inflectional form of a verb
typically denoting simple
occurrence of an action without
reference to its completeness,
duration, or repetition
Source: Merriam-Webster
Dictionary
4. THE AORIST IS GENERALLY AN ACTION
IN THE PAST THAT IS COMPLETED.
5. In an earlier time, many
languages around the world
had what is called the aortic.
The aortic is much like the
preterit in English.
6. Knowledge of the aorist is presented here as a way…
to introduce machine learning(ML).
Although not part of this presentation,
familiarity with the aorist can also assist in learning
aoristic analysis as well Aorist, an ML data management
tool.
7. An example of machine learning:
Audiences stream episodes of
their favorite shows daily,
then one day there is a spike in
viewership of special.
8. This explains the aorist in general:
The audience viewed their favorite shows daily,
and the holiday special was viewed with the largest streaming video
audience.
(aorist)
9. • Much of the work in Machine Learning is cleaning data,
the spike in data would likely be an error.
10. Especially in Armenian,
the aorist is applied based upon the context.
To determine if a pattern in a chart is a random spike,
context is also needed.
11. At this point, a theoretical method in
machine learning can be presented.
What is the likelihood of an event
changes based on context?
12. An application of context can be
found in machine learning in terms of
what is called Bayesian Inference.
Essentially what this is is the
likelihood of an event taking place when
the context of the data changes.
13. Although beyond the scope of this presentation to
provide a detailed quantitative analysis of Bayesian
Inference,
the next slide is the equation as a reference
and to hopefully encourage further learning of this
topic:
14. H is for hypothesis E is for evidence
P(H) is for the prior probability
P(H|E) is for the posterior probability
P(E|H) is the likelihood
P(E) is for the marginal likelihood
P(H|E) =
P(E|H) multiplied by P(H)
P(H)