Developed Streamlit app for uploading CSV files, performing sentiment analysis, and displaying recommendations

Author: revanth1718

Web_app/__pycache__/utils.cpython-311.pyc

@@ -0,0 +1,44 @@
+import streamlit as st
+import pandas as pd
+from utils import analyze_reviews, recommend_movies
+
+st.title("IMDb Movie Review Analysis and Recommendation System")
+
+st.sidebar.header("Upload your CSV")
+uploaded_file = st.sidebar.file_uploader("Choose a CSV file", type="csv")
+
+def load_data(file):
+    try:
+        return pd.read_csv(file, encoding='utf-8')
+    except UnicodeDecodeError:
+        try:
+            return pd.read_csv(file, encoding='latin1')
+        except UnicodeDecodeError:
+            st.error("File encoding not supported. Please upload a CSV file with UTF-8 or Latin1 encoding.")
+            return None
+
+if uploaded_file is not None:
+    reviews_df = load_data(uploaded_file)
+
+    if reviews_df is not None:
+        st.write("Data Preview:")
+        st.write(reviews_df.head())
+
+        st.write("Column Names:")
+        st.write(reviews_df.columns.tolist())
+
+        if 'review' in reviews_df.columns:
+            st.write("Sentiment Analysis:")
+            sentiment_df, analyzed_df = analyze_reviews(reviews_df)
+            st.write(sentiment_df)
+
+            st.write("Analyzed DataFrame with Sentiments:")
+            st.write(analyzed_df.head())
+
+            st.write("Movie Recommendations:")
+            recommendations = recommend_movies(analyzed_df)
+            st.write(recommendations)
+        else:
+            st.error("The uploaded CSV file does not contain a 'review' column.")
+else:
+    st.write("Please upload a CSV file to proceed.")

Web_app/app.py

@@ -0,0 +1,8 @@
+Installations:
+pip install streamlit beautifulsoup4 requests pandas scikit-learn
+
+Run the Streamlit app:
+streamlit run app/app.py
+
+
+

Web_app/instructions.txt

@@ -0,0 +1,33 @@
+import pandas as pd
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.svm import SVC
+from sklearn.metrics import classification_report
+
+def analyze_reviews(df):
+    vectorizer = TfidfVectorizer()
+    X = vectorizer.fit_transform(df['review'])
+    
+    # Dummy sentiment labels for demonstration (replace with actual sentiment labels)
+    y = [1 if i % 2 == 0 else 0 for i in range(len(df))]  
+    
+    model = SVC()
+    model.fit(X, y)
+    y_pred = model.predict(X)
+
+    report = classification_report(y, y_pred, output_dict=True)
+    df['sentiment'] = y_pred
+
+
+    print("Analyzed DataFrame with Sentiments:")
+    print(df.head())
+
+    return pd.DataFrame(report).transpose(), df
+
+def recommend_movies(df):
+    positive_reviews = df[df['sentiment'] == 1]
+    
+    
+    print("DataFrame with Positive Reviews:")
+    print(positive_reviews.head())
+    
+    return positive_reviews.head(5)