diff --git a/hypertune/scoring.py b/hypertune/scoring.py
new file mode 100644
index 0000000..1569318
--- /dev/null
+++ b/hypertune/scoring.py
@@ -0,0 +1,20 @@
+import nltk
+from sklearn.metrics.pairwise import cosine_similarity
+from sentence_transformers import SentenceTransformer
+
+def perplexity_score(text):
+    # Implement perplexity calculation
+    pass
+
+def semantic_coherence(text):
+    model = SentenceTransformer('all-MiniLM-L6-v2')
+    sentences = nltk.sent_tokenize(text)
+    embeddings = model.encode(sentences)
+    scores = [cosine_similarity([embeddings[i]], [embeddings[i+1]])[0][0] for i in range(len(embeddings)-1)]
+    return sum(scores) / len(scores)
+
+def factual_accuracy(text):
+    # Implement fact-checking logic
+    pass
+
+# Add more scoring functions