Three classic classifiers on the Iris dataset (150 flowers, 4 measurements, 3 species) — how they decide, and what each is good at.

Support Vector Machine (SVM)

Python

from sklearn import datasets
from sklearn.model_selection import train_test_split
from sklearn.svm import SVC
from sklearn.metrics import accuracy_score

iris = datasets.load_iris()
X = iris.data[:, :2]              # only 2 features (sepal length & width) so it can be plotted in 2-D
y = iris.target
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=0)

model = SVC(kernel="rbf", C=1.0, gamma="auto").fit(X_train, y_train)
print(f"Accuracy: {accuracy_score(y_test, model.predict(X_test)):.2f}")   # Accuracy: 0.80

Parameter	Effect
`kernel`	`linear` (straight boundary), `rbf` (curved), `poly`
`C`	Low = wider margin, tolerates mistakes; high = fits training data tightly (risk of overfitting)
`gamma`	For `rbf`: how far one sample's influence reaches — high gamma makes wiggly boundaries

Decision Tree

Python

from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeClassifier, export_text
from sklearn.metrics import accuracy_score

iris = load_iris()
X_train, X_test, y_train, y_test = train_test_split(iris.data, iris.target, test_size=0.3, random_state=0)
model = DecisionTreeClassifier(criterion="entropy", max_depth=3, random_state=0).fit(X_train, y_train)
print(f"Accuracy: {accuracy_score(y_test, model.predict(X_test)):.2f}")   # Accuracy: 0.98
print(export_text(model, feature_names=iris.feature_names).splitlines()[0])   # |--- petal width (cm) <= 0.75

Random Forest

Python

from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score

iris = load_iris()
X_train, X_test, y_train, y_test = train_test_split(iris.data, iris.target, test_size=0.3, random_state=0)
model = RandomForestClassifier(n_estimators=100, max_depth=3, random_state=0).fit(X_train, y_train)
print(f"Accuracy: {accuracy_score(y_test, model.predict(X_test)):.2f}")   # Accuracy: 0.98

ranked = sorted(zip(model.feature_importances_, iris.feature_names), reverse=True)
print([name for _, name in ranked[:2]])          # ['petal length (cm)', 'petal width (cm)']

	SVM	Decision tree	Random forest
Idea	Widest-margin boundary	Sequence of feature questions	Many trees voting
Interpretable	Low	High (you can read the tree)	Medium (feature importances)
Scaling features	Needed	Not needed	Not needed
Overfitting risk	Tune `C`, `gamma`	High — limit depth	Low