Update eda.py

eda.py

@@ -21,6 +21,14 @@ def process_review(review):
     review = re.sub(r'[^a-z\s]', '', review) # Remove non-alphabetical characters
     return review
 
+def find_review_column(data):
+    """Attempt to identify the review column from the dataset."""
+    potential_columns = ['review', 'text', 'comment', 'message', 'description']
+    for col in potential_columns:
+        if col in data.columns:
+            return col
+    return None
+
 def display_eda(data):
     # Derive the 'sentiment' column from 'rating' if it doesn't exist
     if 'sentiment' not in data.columns:
@@ -42,11 +50,16 @@ def display_eda(data):
 
     # Word cloud for each sentiment
     st.subheader("Word Clouds for Sentiments")
+    review_column = find_review_column(data)
+    if not review_column:
+        st.error("Couldn't find a column with reviews. Please check the dataset.")
+        return
+    
     sentiments = data['sentiment'].unique()
     for sentiment in sentiments:
         st.write(f"Word Cloud for {sentiment}")
         subset = data[data['sentiment'] == sentiment]
-        text = " ".join(process_review(review) for review in subset['review'])
+        text = " ".join(process_review(review) for review in subset[review_column])
         wordcloud = WordCloud(max_words=100, background_color="white").generate(text)
         plt.figure()
         plt.imshow(wordcloud, interpolation="bilinear")