I pointed one AI agent at ten public datasets — here's what it found

Jul 18, 2026

Worked examples on public, open datasets — no client or private data. This is the shareable version of the kind of analysis I run privately on real business data.

I’ve been building a data agent that does the boring-but-hard part of analysis on its own: it loads a table, profiles every column, tests which relationships are real, models the outcome you care about, and then — the part that actually matters — turns all of that into a few sentences a decision-maker can act on. One question in, one clear answer out.

To show what that looks like, I ran the same pipeline, unchanged across ten well-known public datasets — restaurants, real estate, cars, wine, biology, the economy. Each section below has two parts: the business read (what it means and what to do about it) and, folded underneath, the technical report (every algorithm it ran, the tests it chose, and the scores) for anyone who wants to check its work. The point isn’t any single finding — it’s that the same machine gets to a usable answer on wildly different data.

1. Restaurant tipping — what actually decides the tip

Automated analysis · restaurant tipping

What actually decides how much a table tips

244 dinner & lunch checks · one table · analysed end-to-end by the agent

Bottom line: The size of the check explains the tip almost entirely. Party size nudges it a little; the day of week, the shift, whether guests smoked, and the diner's gender make almost no practical difference. If you want tips to grow, grow the check — not the schedule.

What moves the tip

Check size86%

Party size6%

Day of week5%

Smoker at table2%

Guest gender<1%

Lunch vs dinner~0%

The numbers behind it

$19.79

Average check

$3.00

Average tip

~15%

Effective tip rate

±$0.73

Tip predictable from the bill within this

What this means for the floor

Sell the check, not the shift. Tables above ~$34 tip about $5; tables under ~$11 tip under $2. A single upsell — a starter, a second round, dessert — is worth more to tip income than any night of the week.
Don't chase "good tables." The instinct that certain guests tip better (by gender, by shift, by smoker status) doesn't hold up in the data — those signals are real but tiny. Coaching servers on them wastes attention.
Bigger parties help mostly because they spend more. Party size matters, but largely as a proxy for check size — it isn't a separate lever.

Produced automatically by an AI data agent: the table was loaded, profiled, tested for associations, and modelled, then the drivers were ranked and translated into plain language — the same pipeline, unchanged, run on every dataset. Public open dataset; no private data.

See the technical report — every algorithm, test and score

Technical report · Restaurant tipping

How the agent analysed `tips`

244 rows · 8 operations run · deterministic pipeline

Pipeline executed

Profile — profile: Profiled 7 columns over 244 rows
Association matrix — association_matrix: Pairwise association across 7 columns
Model evaluation — regression_metrics: R²=0.523, RMSE=0.957
Feature importance — permutation_importance: Top feature: total_bill
Key-driver tree — decision_tree_key_drivers: Top driver of 'tip': 'total_bill' (tree R²=0.46)

Association tests (dtype-routed)

Columns	Test chosen	Result
`tip~total_bill`	spearman	strong positive correlation (r=0.679, p=2.5e-34)
`tip~size`	kruskal_wallis	significant difference in tip across size (6 groups, epsilon_squared=0.231, p=8.04e-11)
`tip~day`	kruskal_wallis	significant difference in tip across day (4 groups, epsilon_squared=0.035, p=0.0357)

Model performance

Task: regression · backend: gbt · held-out test rows: 61

R²	0.523
RMSE	0.957
MAE	0.729

Feature importance — permutation_importance

Feature	Score	Share
total_bill	0.858	86%
size	0.064	6%
day	0.051	5%
smoker	0.015	2%
sex	0.00427	0%

Every step above is chosen by transparent rules (statistical test picked from column dtypes; k picked by silhouette; model held out a proper test split). Re-running on the same table yields the same output. Public open dataset.

The folklore of the “good table” mostly evaporates: the lever is check size, not the night or the shift.

2. Titanic — what decided who survived

Automated analysis · Titanic survival

Who survived the Titanic — and what actually decided it

1,309 passengers · one table · classified end-to-end by the agent

Bottom line: Survival wasn't random. Two things decided it: sex first, then the class of ticket. Being female moved the odds more than anything else; travelling first class helped on top of that. Age barely mattered once those two were known.

What decided survival

Sex68%

Passenger class27%

Age5%

The numbers behind it

77%

Survival correctly predicted from 3 facts

Sex

Strongest single factor

0.81

Model's ability to rank risk (AUC, 1.0 = perfect)

Factors that mattered

What to do with it

The headline generalises: when a crisis is governed by a rule (“women and children first”), the data recovers that rule cleanly — you don't need the backstory to see it.
Class was a real, separate advantage, not just a proxy for sex — first-class passengers fared better within each group.
Age added almost nothing once sex and class were known — a reminder that more columns aren't more insight.

Produced automatically by the data agent: the table was loaded, profiled, tested for associations, and modelled, then the drivers were ranked and translated into plain language — the same pipeline, unchanged, run on every dataset. Public open dataset.

See the technical report — every algorithm, test and score

Technical report · Titanic survival

How the agent analysed `titanic`

1309 rows · 7 operations run · deterministic pipeline

Pipeline executed

Profile — profile: Profiled 4 columns over 1309 rows
Association matrix — association_matrix: Pairwise association across 4 columns
Model evaluation — classification_metrics: accuracy=0.771, f1=0.761
Feature importance — permutation_importance: Top feature: sex
Key-driver tree — decision_tree_key_drivers: Top driver of 'survived': 'sex' (tree accuracy=0.79)

Association tests (dtype-routed)

Columns	Test chosen	Result
`survived~sex`	chi_square	significant association between survived and sex (Cramér's V=0.527, p=4.59e-81)
`survived~passengerClass`	chi_square	significant association between survived and passengerClass (Cramér's V=0.313, p=1.72e-28)

Model performance

Task: classification · backend: gbt · held-out test rows: 328

Accuracy	0.771
Precision	0.773
Recall	0.771
F1	0.761
ROC-AUC	0.812

Confusion matrix (rows = actual, cols = predicted)

184	19
56	69

Feature importance — permutation_importance

Feature	Score	Share
sex	0.162	66%
passengerClass	0.076	31%
age	0.00793	3%

A crisis governed by a rule leaves a clean signature in the data; you don’t need the story to recover “women and children first.”

3. California housing — what sets a home’s price

Automated analysis · California housing

What really sets a California home's price

20,640 neighbourhood blocks · one table · modelled end-to-end by the agent

Bottom line: Two forces set the price: where the home is (latitude + longitude) and how wealthy the neighbourhood is. Together they explain the vast majority of the variation. How old the house is barely registers.

What sets the price

Location (lat/long)66%

Neighbourhood income23%

Crowding / occupancy7%

Rooms & house age4%

The numbers behind it

83%

Of price variation explained

Location

Biggest lever, combined

Income

Strongest single factor

~0%

Effect of house age

What to do with it

Location and income are the whole game. For pricing, valuation, or where-to-build decisions, these two beat every property-level detail combined.
Stop over-weighting house features. Age, room counts, and bedroom ratios move price only at the margin — useful for fine-tuning, useless as headline drivers.
The model is trustworthy here (it explains 83% of prices on data it never saw), so it's a fair basis for automated valuation.

See the technical report — every algorithm, test and score

Technical report · California housing

How the agent analysed `california_housing`

20640 rows · 7 operations run · deterministic pipeline

Pipeline executed

Profile — profile: Profiled 9 columns over 20640 rows
Association matrix — association_matrix: Pairwise association across 9 columns
Model evaluation — regression_metrics: R²=0.835, RMSE=0.467
Feature importance — permutation_importance: Top feature: Latitude
Key-driver tree — decision_tree_key_drivers: Top driver of 'MedHouseVal': 'MedInc' (tree R²=0.53)

Association tests (dtype-routed)

Columns	Test chosen	Result
`MedHouseVal~MedInc`	spearman	strong positive correlation (r=0.677, p=0)
`MedHouseVal~HouseAge`	spearman	negligible positive correlation (r=0.075, p=4.84e-27)

Model performance

Task: regression · backend: gbt · held-out test rows: 5160

R²	0.835
RMSE	0.467
MAE	0.313

Feature importance — permutation_importance

Feature	Score	Share
Latitude	0.832	37%
Longitude	0.66	29%
MedInc	0.53	23%
AveOccup	0.15	7%
AveRooms	0.045	2%
HouseAge	0.04	2%
Population	0.00886	0%
AveBedrms	0.00658	0%

Location and neighbourhood income are the whole game; property details move price only at the margin.

4. Diamonds — what a price is made of

Automated analysis · Diamond pricing

What a diamond's price is really made of

53,940 diamonds · one table · modelled end-to-end by the agent

Bottom line: Price is almost entirely about size — carat weight and physical dimensions. Clarity and colour fine-tune it; cut grade barely moves the price at all. The agent predicts price within a few hundred dollars.

What makes the price

Size (carat & dimensions)75%

Clarity11%

Colour6%

Cut grade<1%

The numbers behind it

98%

Of price variation explained

±$287

Typical prediction error

Carat

Dominant driver

Cut

Almost no price effect

What to do with it

Price by weight first. For inventory, appraisal, or buy/sell decisions, size is the anchor — everything else is a modest adjustment.
Cut is oversold as a price driver. It matters for beauty, but the market barely pays for it — worth knowing when negotiating.
Automated pricing is viable: the agent explains 98% of price on unseen stones, tight enough to flag mispriced listings.

See the technical report — every algorithm, test and score

Technical report · Diamond pricing

How the agent analysed `diamonds`

53940 rows · 8 operations run · deterministic pipeline

Pipeline executed

Profile — profile: Profiled 10 columns over 53940 rows
Association matrix — association_matrix: Pairwise association across 10 columns
Model evaluation — regression_metrics: R²=0.980, RMSE=560
Feature importance — permutation_importance: Top feature: carat
Key-driver tree — decision_tree_key_drivers: Top driver of 'price': 'carat' (tree R²=0.88)

Association tests (dtype-routed)

Columns	Test chosen	Result
`price~carat`	spearman	very strong positive correlation (r=0.963, p=0)
`price~cut`	kruskal_wallis	significant difference in price across cut (5 groups, epsilon_squared=0.018, p=1.53e-210)
`price~clarity`	kruskal_wallis	significant difference in price across clarity (8 groups, epsilon_squared=0.050, p=0)

Model performance

Task: regression · backend: gbt · held-out test rows: 13485

R²	0.98
RMSE	560.208
MAE	287.4

Feature importance — permutation_importance

Feature	Score	Share
carat	0.647	47%
y	0.468	34%
clarity	0.156	11%
color	0.087	6%
z	0.017	1%
x	0.00557	0%
cut	0.00248	0%
depth	0.00119	0%

Size dominates; clarity and colour fine-tune; cut grade — the thing shoppers obsess over — barely moves the price.

5. Penguins — identifying species from a ruler

Automated analysis · Penguin species

Telling three penguin species apart from a few measurements

344 penguins · one table · classified end-to-end by the agent

Bottom line: Three species can be identified almost perfectly from a handful of body measurements plus which island a bird was seen on. The agent labels species correctly 98% of the time.

What identifies the species

Which island45%

Bill length35%

Flipper length11%

Bill depth8%

The numbers behind it

98%

Correctly identified

Species separated

2–3

Measurements needed

Natural groups found unsupervised

What to do with it

Cheap measurements beat expensive ones. A ruler and a location note classify a bird as well as any lab work — the practical lesson for any “which type is this?” problem.
Location carries real signal. Where something is observed is a feature, not just metadata.
The pattern is genuine, not imposed: left to find groups on its own, the agent recovered clean clusters that track the species.

See the technical report — every algorithm, test and score

Technical report · Penguin species

How the agent analysed `penguins`

344 rows · 7 operations run · deterministic pipeline

Pipeline executed

Profile — profile: Profiled 8 columns over 344 rows
Association matrix — association_matrix: Pairwise association across 8 columns
Model evaluation — classification_metrics: accuracy=0.977, f1=0.977
Feature importance — permutation_importance: Top feature: island
Key-driver tree — decision_tree_key_drivers: Top driver of 'species': 'flipper_length_mm' (tree accuracy=0.95)
Clustering — kmeans: 2 clusters, silhouette=0.417
Cluster profiling — profile_clusters: Characterised 2 clusters

Model performance

Task: classification · backend: gbt · held-out test rows: 86

Accuracy	0.977
Precision	0.978
Recall	0.977
F1	0.977
ROC-AUC	1

Confusion matrix (rows = actual, cols = predicted)

38	0	0
1	16	0
1	0	30

Feature importance — permutation_importance

Feature	Score	Share
island	0.221	45%
bill_length_mm	0.172	35%
flipper_length_mm	0.053	11%
bill_depth_mm	0.041	8%

Unsupervised clustering

2 clusters, silhouette=0.417 — method kmeans. k chosen automatically by silhouette score.

Cheap measurements beat expensive ones: a few body measures and a location note classify a bird 98% of the time.

6. Wine — separating varieties by chemistry

Automated analysis · Wine cultivars

Separating three wine varieties by chemistry alone

178 wines · one table · classified end-to-end by the agent

Bottom line: Three grape varieties are perfectly separable from their chemistry — zero mistakes. A few compounds (flavanoids, proline, colour intensity) do all the work.

What tells the varieties apart

Flavanoids39%

Proline29%

Colour intensity22%

Alcohol & hue7%

The numbers behind it

100%

Correctly classified

Errors on unseen wines

Markers carry the signal

Natural groups found unsupervised

What to do with it

A short test panel is enough. You don't need to measure everything — three compounds authenticate variety with no errors.
Strong fit for QA and authenticity checks: a fast chemical signature that flags a mislabelled or off-spec batch.
The chemistry is genuinely distinct: unsupervised clustering found the same three groups without being told they exist.

See the technical report — every algorithm, test and score

Technical report · Wine cultivars

How the agent analysed `wine`

178 rows · 7 operations run · deterministic pipeline

Pipeline executed

Profile — profile: Profiled 14 columns over 178 rows
Association matrix — association_matrix: Pairwise association across 14 columns
Model evaluation — classification_metrics: accuracy=1.000, f1=1.000
Feature importance — permutation_importance: Top feature: flavanoids
Key-driver tree — decision_tree_key_drivers: Top driver of 'wine_class': 'proline' (tree accuracy=0.93)
Clustering — kmeans: 3 clusters, silhouette=0.311
Cluster profiling — profile_clusters: Characterised 3 clusters

Model performance

Task: classification · backend: gbt · held-out test rows: 45

Accuracy	1
Precision	1
Recall	1
F1	1
ROC-AUC	1

Confusion matrix (rows = actual, cols = predicted)

15	0	0
0	18	0
0	0	12

Feature importance — permutation_importance

Feature	Score	Share
flavanoids	0.262	39%
proline	0.198	29%
color_intensity	0.151	22%
alcohol	0.024	4%
hue	0.02	3%
malic_acid	0.011	2%
magnesium	0.011	2%

Unsupervised clustering

3 clusters, silhouette=0.311 — method kmeans. k chosen automatically by silhouette score.

Three compounds tell three grape varieties apart with zero errors — a short test panel is enough.

7. Cars — and knowing when NOT to trust a model

Automated analysis · Car fuel economy

What drives fuel economy — and why the agent won't overclaim on 32 cars

32 cars · one table · analysed end-to-end by the agent

Bottom line: Weight and engine power drive fuel economy — heavier, more powerful cars burn more. The relationships are strong and clear. But with only 32 cars, the agent flags that there's too little data to build a trustworthy predictive model rather than pretending otherwise.

What relates to MPG (correlation strength)

Engine power (hp)very strong

Weightvery strong

Cylindersstrong

Rear-axle ratiomoderate

The numbers behind it

Cars — a small sample

Weight + power

The real levers

Honest

Model flagged as unreliable

r ≈ 0.9

Strength of the top relationships

What to do with it

The levers are real: to improve fuel economy, cut weight and power — the data is unambiguous on direction.
The restraint is the point. Most tools would still hand you a confident prediction; this one tells you 32 rows can't support one. Knowing when not to trust a model is worth more than a false number.
Next step if it mattered: gather more vehicles before relying on a predictive model — the relationships suggest it would pay off.

See the technical report — every algorithm, test and score

Technical report · Car fuel economy

How the agent analysed `mtcars`

32 rows · 8 operations run · deterministic pipeline

Pipeline executed

Profile — profile: Profiled 11 columns over 32 rows
Association matrix — association_matrix: Pairwise association across 11 columns
Model evaluation — regression_metrics: R²=-0.619, RMSE=5.83
Feature importance — permutation_importance: Top feature: disp
Key-driver tree — decision_tree_key_drivers: Top driver of 'mpg': 'wt' (tree R²=0.96)

Association tests (dtype-routed)

Columns	Test chosen	Result
`mpg~wt`	pearson	very strong negative correlation (r=-0.868, p=1.29e-10)
`mpg~hp`	spearman	very strong negative correlation (r=-0.895, p=5.09e-12)
`mpg~cyl`	kruskal_wallis	significant difference in mpg across cyl (3 groups, epsilon_squared=0.831, p=2.57e-06)

Model performance

Task: regression · backend: gbt · held-out test rows: 8

R²	-0.619
RMSE	5.83
MAE	5.308

The levers (weight, power) are obvious, but the agent refuses to build a confident model on 32 rows. The restraint is the feature.

8. Breast-cancer biopsies — flagging malignancy

Automated analysis · Breast-cancer diagnosis

Flagging malignant tumours from cell-shape measurements

569 biopsies · one table · classified end-to-end by the agent

Bottom line: From measurements of cell size and shape, the agent separates malignant from benign tumours with about 98% accuracy. The strongest signals are the worst-case irregularities in a sample — the most abnormal cells give it away.

What flags a malignant tumour

Worst concave points17%

Worst texture14%

Worst perimeter14%

Size variation (area error)9%

The numbers behind it

98%

Overall accuracy

51 / 53

Malignant cases caught

False alarm out of 90 benign

0.99

Risk-ranking quality (AUC)

What to do with it

Strong decision-support, not a replacement. Catching 51 of 53 malignant cases makes this a serious triage aid — but the 2 misses are why a clinician stays in the loop.
Irregularity is the tell. The “worst” (most extreme) measurements in a sample carry more signal than the averages — focus attention there.
Consistent and auditable: the same sample always yields the same call, with the reasoning inspectable — important in a clinical setting.

See the technical report — every algorithm, test and score

Technical report · Breast-cancer diagnosis

How the agent analysed `breast_cancer`

569 rows · 5 operations run · deterministic pipeline

Pipeline executed

Profile — profile: Profiled 31 columns over 569 rows
Association matrix — association_matrix: Pairwise association across 31 columns
Model evaluation — classification_metrics: accuracy=0.979, f1=0.979
Feature importance — permutation_importance: Top feature: worst concave points
Key-driver tree — decision_tree_key_drivers: Top driver of 'diagnosis': 'worst radius' (tree accuracy=0.94)

Model performance

Task: classification · backend: gbt · held-out test rows: 143

Accuracy	0.979
Precision	0.979
Recall	0.979
F1	0.979
ROC-AUC	0.993

Confusion matrix (rows = actual, cols = predicted)

89	1
2	51

Feature importance — permutation_importance

Feature	Score	Share
worst concave points	0.033	17%
worst texture	0.027	14%
worst perimeter	0.027	14%
area error	0.018	9%
worst area	0.017	9%
mean texture	0.012	6%
mean concave points	0.012	6%
mean area	0.00839	4%
worst concavity	0.00769	4%
symmetry error	0.00699	4%
worst smoothness	0.00559	3%
perimeter error	0.0049	3%

About 98% accuracy from cell-shape measurements — strong decision-support, with the misses that keep a clinician in the loop.

9. Air quality — seeing bad-air days coming

Automated analysis · Urban air quality

What drives ground-level ozone — and how to see bad-air days coming

153 days · one table · modelled end-to-end by the agent

Bottom line: Ground-level ozone is mostly a weather story: it climbs with heat and sunlight and falls when it's windy. Temperature is the single biggest driver. The agent predicts ozone from the weather with about 79% accuracy.

What drives ozone

Temperature54%

Wind26%

Sunlight (solar)11%

Time of month9%

The numbers behind it

79%

Of ozone variation explained

Heat

Biggest driver

Wind

Biggest reliever

Hot + still + sunny

The high-ozone recipe

What to do with it

Bad-air days are forecastable. Hot, calm, sunny conditions are the recipe — a simple early-warning trigger for public-health alerts.
Wind is the natural cleaner. Still air is when ozone accumulates; factor that into when to schedule sensitive activity.
Weather explains most of it (79%) — the rest points to sources worth investigating beyond the weather.

See the technical report — every algorithm, test and score

Technical report · Urban air quality

How the agent analysed `airquality`

153 rows · 8 operations run · deterministic pipeline

Pipeline executed

Profile — profile: Profiled 6 columns over 153 rows
Association matrix — association_matrix: Pairwise association across 6 columns
Model evaluation — regression_metrics: R²=0.794, RMSE=14.7
Feature importance — permutation_importance: Top feature: Temp
Key-driver tree — decision_tree_key_drivers: Top driver of 'Ozone': 'Temp' (tree R²=0.73)

Association tests (dtype-routed)

Columns	Test chosen	Result
`Ozone~Temp`	spearman	very strong positive correlation (r=0.774, p=2.25e-24)
`Ozone~Wind`	spearman	strong negative correlation (r=-0.590, p=3.13e-12)
`Ozone~Solar.R`	spearman	moderate positive correlation (r=0.348, p=0.000181)

Model performance

Task: regression · backend: gbt · held-out test rows: 29

R²	0.794
RMSE	14.67
MAE	10.878

Feature importance — permutation_importance

Feature	Score	Share
Temp	0.577	54%
Wind	0.278	26%
Solar.R	0.115	11%
Day	0.069	7%
Month	0.023	2%

Ground-level ozone is mostly a weather story: hot, still and sunny is the high-ozone recipe.

10. The US economy — 48 years of unemployment

Automated analysis · US economy

48 years of US unemployment — the shifts that actually matter

574 monthly readings, 1967–2015 · one table · time-series analysis by the agent

Bottom line: Over five decades, unemployment didn't drift smoothly — it moved in regimes, with 11 sharp structural breaks that line up with known recessions (1975, 1982, 1990, 2001, 2008…). Across the span it rose about 34%, and the near-term forecast points modestly down.

What the series shows

Long-run rise across the period+34%

Structural breaks detected11

Seasonal swing within each yearyearly

Near-term directiondown

The numbers behind it

+34%

Rise across the period

Structural breaks (regime shifts)

2008

Largest recent break

Down

Direction of the 12-month forecast

What to do with it

Plan around regimes, not averages. A single long-run average hides the breaks; the useful unit is the period between shocks, and the agent finds those boundaries automatically.
The breaks aren't noise — they map onto real recessions, which is a good sanity check that the method is finding structure, not artefacts.
Forecasts come with honest error bands. The near-term call is a modest decline, but the widening range is the point: it says how much to trust it.

See the technical report — every algorithm, test and score

Technical report · US economy

How the agent analysed `economics`

574 rows · 6 operations run · deterministic pipeline

Pipeline executed

Profile — profile: Profiled 6 columns over 574 rows
Association matrix — association_matrix: Pairwise association across 5 columns
Seasonal decomposition — seasonal_decompose: Decomposed unemploy (additive, period=12)
Forecast — arima: 12-step forecast of unemploy (ARIMA(1, 1, 1))
Change-point detection — ruptures_pelt_rbf: 11 changepoint(s) in unemploy; first at 1970-06-01T00:00:00
Period comparison — two_window_comparison: unemploy up +33.9% across the split (significant, p=1.77e-21)

The series moves in regimes, not a smooth drift; the agent finds the 11 breaks — and they line up with real recessions.

That’s the through-line across all ten: the honest answer is usually simpler than the one people expect, it points at a decision you can act on, and — when the data can’t support a claim — the agent says so instead of inventing one. The value isn’t a cleverer model. It’s getting to that one sentence fast, on any table you hand it.

I’m a data scientist — if you’ve got a pile of data and a decision hiding in it, get in touch.

I pointed one AI agent at ten public datasets — here's what it found

1. Restaurant tipping — what actually decides the tip

What actually decides how much a table tips

How the agent analysed tips

2. Titanic — what decided who survived

Who survived the Titanic — and what actually decided it

How the agent analysed titanic

3. California housing — what sets a home’s price

What really sets a California home's price

How the agent analysed california_housing

4. Diamonds — what a price is made of

What a diamond's price is really made of

How the agent analysed diamonds

5. Penguins — identifying species from a ruler

Telling three penguin species apart from a few measurements

How the agent analysed penguins

6. Wine — separating varieties by chemistry

Separating three wine varieties by chemistry alone

How the agent analysed wine

7. Cars — and knowing when NOT to trust a model

What drives fuel economy — and why the agent won't overclaim on 32 cars

How the agent analysed mtcars

8. Breast-cancer biopsies — flagging malignancy

Flagging malignant tumours from cell-shape measurements

How the agent analysed breast_cancer

9. Air quality — seeing bad-air days coming

What drives ground-level ozone — and how to see bad-air days coming

How the agent analysed airquality

10. The US economy — 48 years of unemployment

48 years of US unemployment — the shifts that actually matter

How the agent analysed economics

How the agent analysed `tips`

How the agent analysed `titanic`

How the agent analysed `california_housing`

How the agent analysed `diamonds`

How the agent analysed `penguins`

How the agent analysed `wine`

How the agent analysed `mtcars`

How the agent analysed `breast_cancer`

How the agent analysed `airquality`

How the agent analysed `economics`