diff --git a/doc/whats_new/v0.11.rst b/doc/whats_new/v0.11.rst
index 666be580c..8b3b54d73 100644
--- a/doc/whats_new/v0.11.rst
+++ b/doc/whats_new/v0.11.rst
@@ -14,6 +14,10 @@ Bug fixes
   they are plugged into an Euclidean distance computation.
   :pr:`1014` by :user:`Guillaume Lemaitre <glemaitre>`.
 
+- Raise an informative error message when all support vectors are tagged as noise in
+  :class:`~imblearn.over_sampling.SVMSMOTE`.
+  :pr:`1016` by :user:`Guillaume Lemaitre <glemaitre>`.
+
 - Fix a bug in :class:`~imblearn.over_sampling.SMOTENC` where the median of standard
   deviation of the continuous features was only computed on the minority class. Now,
   we are computing this statistic for each class that is up-sampled.
diff --git a/imblearn/over_sampling/_smote/filter.py b/imblearn/over_sampling/_smote/filter.py
index 4966b211b..b02fc7be2 100644
--- a/imblearn/over_sampling/_smote/filter.py
+++ b/imblearn/over_sampling/_smote/filter.py
@@ -495,6 +495,11 @@ def _fit_resample(self, X, y):
             support_vector = _safe_indexing(
                 support_vector, np.flatnonzero(np.logical_not(noise_bool))
             )
+            if support_vector.shape[0] == 0:
+                raise ValueError(
+                    "All support vectors are considered as noise. SVM-SMOTE is not "
+                    "adapted to your dataset. Try another SMOTE variant."
+                )
             danger_bool = self._in_danger_noise(
                 self.nn_m_, support_vector, class_sample, y, kind="danger"
             )
diff --git a/imblearn/over_sampling/_smote/tests/test_svm_smote.py b/imblearn/over_sampling/_smote/tests/test_svm_smote.py
index cf753b275..49e01f6b9 100644
--- a/imblearn/over_sampling/_smote/tests/test_svm_smote.py
+++ b/imblearn/over_sampling/_smote/tests/test_svm_smote.py
@@ -1,5 +1,6 @@
 import numpy as np
 import pytest
+from sklearn.datasets import make_classification
 from sklearn.linear_model import LogisticRegression
 from sklearn.neighbors import NearestNeighbors
 from sklearn.svm import SVC
@@ -61,3 +62,27 @@ def test_svm_smote_not_svm(data):
     err_msg = "`svm_estimator` is required to exposed a `support_` fitted attribute."
     with pytest.raises(RuntimeError, match=err_msg):
         SVMSMOTE(svm_estimator=LogisticRegression()).fit_resample(*data)
+
+
+def test_svm_smote_all_noise(data):
+    """Check that we raise a proper error message when all support vectors are
+    detected as noise and there is nothing that we can do.
+
+    Non-regression test for:
+    https://github.com/scikit-learn-contrib/imbalanced-learn/issues/742
+    """
+    X, y = make_classification(
+        n_classes=3,
+        class_sep=0.001,
+        weights=[0.004, 0.451, 0.545],
+        n_informative=3,
+        n_redundant=0,
+        flip_y=0,
+        n_features=3,
+        n_clusters_per_class=2,
+        n_samples=1000,
+        random_state=10,
+    )
+
+    with pytest.raises(ValueError, match="SVM-SMOTE is not adapted to your dataset"):
+        SVMSMOTE(k_neighbors=4, random_state=42).fit_resample(X, y)