Introducing 

Prezi AI.

Your new presentation assistant.

Refine, enhance, and tailor your content, source relevant images, and edit visuals quicker than ever before.

Loading…
Transcript

Other possibilities

Chance of a high quality brew

Use of a brew-watch 60.3%

Use a brew-thermometer 52.7%

Make teapot mandatory 52.0%

Idea

Invent the 'milk measure'

Ta dah!!

*

Introducing the milk measure increases our chance of high quality tea to a massive 70.5%.

Result.

Dilemma!! What might we do to intervene?

Next step...

Linking these together

Almost done

Use of teapot

What the network looks like

Water temp.

Yes

No

> 95 C

< 95 C

Brew time

Milk amount

Long enough

Not long enough

Brew quality

High

Medium

Low

Good

Sub-standard

Tea quality

High

Medium

Low

*

So, everything else being equal, the chance of a high quality cup of tea is only 50.3%. Most disappointing.

Step Three...

Conditional probabilities

Learning outcomes

In true Blue Peter fashion:

  • Brewing time (time teabag is left in)
  • Amount of milk
  • Temperature of water
  • Use of teapot

Selecting the most likely state given the parent states

1. Appreciate conditional probability

2. Practice building a simple Bayesian network

Crucial question:

What variables affect the 'perfection' of a cup of tea?

An investigation into

How to make the perfect cup of tea

Learn more about creating dynamic, engaging presentations with Prezi