Pattern Recognition terms a little fuzzy

144 Views Asked by At

I've been trying to learn more about probability and the websites I have visited are not describing the relationship to each other that well (when to use what, for what purpose in conjunction with what). I am from a programming background and need to see a clear connection/definition. I am lost in the world of complicated math language at the moment! Could anyone help me understand the following terms in a more simple way?

  • Bayes Decision Theory This results in a formula (Bayes rule) that can be used to calculate the prior and post probabilities.

  • Prior Probability This is used to calculate what we expect will happen based on sample data sets.

  • Conditional Probability When and where is this needed? (In calculations that is)

  • Posterior Probability After observing events of "today", we can guess tomorrow.

  • Discriminant Function The function that separates two classes for example (a straight line on a $2D$ graph)

  • Gaussian Distribution This describes a bell-like shape on a $2D$ graph, for example over a given dataset. How does one confirm if a class of samples are of Gaussian distribution?

I will update the question with relevant information! Any help clarifying these points would be great!