Menu
CoddyTech
START LEARNING
flag En iconEnglishdown icon

Reading Scatter Plots

Part of the Statistics for ML section of Coddy's Machine Learning journey. Lesson 29 of 64.

This chapter ends with one picture of how every column of a table relates to every other:

A colored grid for the 4 measurements of 150 iris flowers, each cell holding a number from -1 to 1. Petal length and petal width share a deep red 0.96; sepal width is pale blue against the others

It starts with your eyes. Each dot of a scatter plot is one sample; read the cloud by its direction, strength and shape:

Six small scatter plots: strong positive, a tight rising band; weak positive, a loose rising cloud; no relationship, a shapeless blob; strong negative and weak negative, falling bands; curved, a U shape

Rising is positive, falling negative. A tight band is strong, a loose cloud weak, a blob none. A bend is curved.

Look before you compute: the number in the next lesson sees only straight bands.

Try it yourself

This lesson doesn't include a code challenge.

quiz iconTest yourself

This lesson includes a short quiz. Start the lesson to answer it and track your progress.

All lessons in Statistics for ML

1Center and Spread

Mean and MedianPercentiles and QuartilesVariance and StdSample or Population?The Most Common ValueRecap: Delivery Times

4Samples and Populations

Population and SampleSample Means VaryThe Sampling DistributionBigger Samples, Less NoiseRecap: Polling a Town

7Trend Lines

ResidualsLeast Squares FitDrawing the FitPredicting with a LineHow Good Is the Fit?Recap: Practice Hours

10Encoding Categories

Categories as NumbersOne-Hot EncodingBinning NumbersRecap: Car Listings

2Distributions

Density HistogramsShapes of DataThe Normal CurveThe 68-95-99.7 RuleZ-ScoresRecap: Battery Life

5Outliers

Box PlotsThe IQR RuleThe Z-Score RuleHandling OutliersRecap: Sensor Glitches

8Comparing Groups

Box Plots by GroupDifference in MeansCould It Be Chance?P-ValuesThe t-TestRecap: A/B Test

11Exploring a Dataset

A First LookEvery Feature at OnceFeatures Against the LabelTwo Best FeaturesRecap: Diabetes Data

3Probability by Simulation

Probability as FrequencyThe Law of Large NumbersSums of DiceExpected ValueConditional ProbabilityRecap: Dice Stall Odds

6Relationships

Reading Scatter PlotsCorrelationCorrelation MatrixCorrelation HeatmapsCorrelation Is Not CauseRecap: Ice Cream and Sun

9Feature Scaling

Why Scale FeaturesMin-Max ScalingStandardizationStandardScalerRecap: Climbing Partners

12Final Challenges

Train DelaysAir QualityWine ChemistryClinic Survey

Practice on your own: Online Python compiler