App

app.py

@@ -8,19 +8,41 @@ st.image("Netflix.png")
 movies_list = pickle.load(open("content_dict.pkl",'br'))
 movies = pd.DataFrame(movies_list)
 
-def recommend(movie):
-    index = movies[movies['title'] == movie].index[0]
-    distances = sorted(list(enumerate(similarity[index])), reverse=True, key=lambda x: x[1])
-    recommended_movie_names = []
-    for i in distances[1:6]:
-        recommended_movie_names.append(movies.iloc[i[0]].title)
+similarity= pickle.load(open('cosine_similarity.pkl','rb'))
+
+def recommend(title, cosine_sim=similarity, data=movies):
+    recommended_content=[]
+    # Get the index of the input title in the programme_list
+    programme_list = data['title'].to_list()
+    index = programme_list.index(title)
+
+
+    # Create a list of tuples containing the similarity score and index 
+    # between the input title and all other programs in the dataset
+    sim_scores = list(enumerate(cosine_sim[index]))
+
+
+    # Sort the list of tuples by similarity score in descending order
+    sim_scores = sorted(sim_scores, key=lambda x: x[1], reverse=True)[1:11]
+
+
+    # Get the recommended movie titles and their similarity scores
+    recommend_index = [i[0] for i in sim_scores]
+    rec_movie = data['title'].iloc[recommend_index]
+    rec_score = [round(i[1], 4) for i in sim_scores]
+
+
+    # Create a pandas DataFrame to display the recommendations
+    rec_table = pd.DataFrame(list(zip(rec_movie, rec_score)), columns=['Recommendation', 'Similarity_score(0-1)'])
+    # recommended_content.append(rec_table['Recommendation'].values)
+
+
+    return rec_table['Recommendation'].values
 
-    return recommended_movie_names
 
 # Displaying title
 st.title(" Movie Recommender System ")
 
-similarity = pickle.load(open('cosine_similarity.pkl','rb'))
 
 movie_list = movies['title'].values
 selected_movie = st.selectbox(