Skip to content
BENIBIX Learn

Math & Statistics

Numbers, odds and making sense of data · 41 terms

Sort
Compare

Number Systems

3

Ways of writing numbers with different sets of digits, and why computers use them.

The normal way to write numbers, with ten digits from 0 to 9.

A way of writing numbers with only 0 and 1. Computers use it inside.

A way of writing numbers with sixteen digits, 0 to 9 and A to F.

Arithmetic

3

Everyday sums about parts and wholes: percentages, the gap between them, and ratios.

A part of a whole, counted out of 100. 40% means 40 in every 100.

The plain gap between two percentages. Going from 10% to 12% is 2 points.

Two amounts compared, part to part. A ratio of 1:3 means one for every three.

Linear Algebra

4

Lists and grids of numbers, and the sums that join them. This is the math inside every neural network.

A list of numbers in order. AI turns words and photos into these.

A grid of numbers in rows and columns. Neural networks are full of them.

A block of numbers. A photo is one, with height, width and colour.

Times each pair of numbers in two lists. Then add it all up.

Calculus

3

How steep a line or curve is, and how fast a thing changes at one moment. Machine learning runs on it.

How steep a line is. Divide how far it rises by how far it runs.

How fast a thing changes at one moment. It is the slope at a single point.

A list of slopes, one for each input. Together they point the steepest way up.

Probability

7

How likely things are, the shape the chances make, and what a choice is worth on average.

How likely a thing is. 0 means it will not happen, and 1 means it will.

The long-run average of a chancy choice. Each outcome counts by its chance.

Do a game of chance lots of times. The average lands close to what you expect.

Thinking a coin owes you heads after a long run of tails.

A way to rate skill with a number that moves with each win and loss.

Every outcome and its chance, side by side. All the chances add up to 1.

A bell-shaped spread. Most values sit in the middle and few out at the edges.

Statistics

21

Making sense of a set of numbers: its middle, its spread, where it came from, and what it can prove.

Averages

4

Ways to sum up a set of numbers with one typical value.

Add the numbers up and divide by how many there are. Also called the average.

The middle value once the numbers are in order. Half sit below and half above.

The value that shows up most often. It works for words as well as numbers.

The mean of the last few values, done again each day. It smooths out the bumps.

Spread

4

How far the numbers sit from each other and from the middle, and the ones that sit far out.

The average squared gap from the mean. A bigger one means more spread out.

How spread out numbers are. Small means bunched up, big means scattered.

Where a value ranks out of 100. The 90th percentile beats 90% of the rest.

A value far away from all the others. It can be a mistake or a real surprise.

Populations and Samples

3

The whole group you want to know about, the part you measure, and how that part is picked.

The whole group you want to know about. You often measure only part of it.

The small group you really measure. You use it to guess about the whole group.

How you pick who to measure. Random picks give every member the same chance.

Correlation and Regression

5

When things move together, when one truly causes another, and fitting a line through the dots.

How closely two things rise and fall together. It runs from -1 to 1.

One thing truly makes another happen. Change the cause and the effect changes.

A hidden third thing that moves two others at once, so they look linked.

Fitting a line through data. It shows how one thing changes when another does.

After a wild high or low score, the next one tends to be less wild.

Statistical Inference

5

Using a sample to judge whether an effect is real, and how sure you can be.

A check of whether data shows a real effect. Or it could just be chance.

The chance of data this odd if nothing real is going on.

A result that clears a bar set before the test. Chance alone would rarely give it.

A range around a guess that likely holds the true value. A narrow one means you are surer.

The smallest change a test is big enough to notice.