diff --git a/sklearn/semi_supervised/_self_training.py b/sklearn/semi_supervised/_self_training.py
index 7b9bbe5e389fd..fee014a3aa838 100644
--- a/sklearn/semi_supervised/_self_training.py
+++ b/sklearn/semi_supervised/_self_training.py
@@ -91,6 +91,12 @@ class SelfTrainingClassifier(MetaEstimatorMixin, BaseEstimator):
 
         .. versionadded:: 0.24
 
+    feature_names_in_ : ndarray of shape (`n_features_in_`,)
+        Names of features seen during :term:`fit`. Defined only when `X`
+        has feature names that are all strings.
+
+        .. versionadded:: 1.0
+
     n_iter_ : int
         The number of rounds of self-training, that is the number of times the
         base estimator is fitted on relabeled variants of the training set.
@@ -285,6 +291,7 @@ def predict(self, X):
             Array with predicted labels.
         """
         check_is_fitted(self)
+        self._check_feature_names(X, reset=False)
         return self.base_estimator_.predict(X)
 
     def predict_proba(self, X):
@@ -301,6 +308,7 @@ def predict_proba(self, X):
             Array with prediction probabilities.
         """
         check_is_fitted(self)
+        self._check_feature_names(X, reset=False)
         return self.base_estimator_.predict_proba(X)
 
     @if_delegate_has_method(delegate="base_estimator")
@@ -318,6 +326,7 @@ def decision_function(self, X):
             Result of the decision function of the `base_estimator`.
         """
         check_is_fitted(self)
+        self._check_feature_names(X, reset=False)
         return self.base_estimator_.decision_function(X)
 
     @if_delegate_has_method(delegate="base_estimator")
@@ -335,6 +344,7 @@ def predict_log_proba(self, X):
             Array with log prediction probabilities.
         """
         check_is_fitted(self)
+        self._check_feature_names(X, reset=False)
         return self.base_estimator_.predict_log_proba(X)
 
     @if_delegate_has_method(delegate="base_estimator")
@@ -355,4 +365,5 @@ def score(self, X, y):
             Result of calling score on the `base_estimator`.
         """
         check_is_fitted(self)
+        self._check_feature_names(X, reset=False)
         return self.base_estimator_.score(X, y)
diff --git a/sklearn/tests/test_common.py b/sklearn/tests/test_common.py
index 2ba4b59e283a1..ea32a6de37126 100644
--- a/sklearn/tests/test_common.py
+++ b/sklearn/tests/test_common.py
@@ -329,7 +329,6 @@ def test_check_n_features_in_after_fitting(estimator):
     "kernel_approximation",
     "model_selection",
     "multioutput",
-    "semi_supervised",
 }
 
 _estimators_to_test = list(