ASS 14(recommender system)

 import pandas as pd 

from sklearn.feature_extraction.text import TfidfVectorizer 

from sklearn.metrics.pairwise import cosine_similarity 

 

movies=pd.read_csv('movies.csv') 

movies.dropna(inplace=True) 

movies = movies.head(1000)  # we can adjust this number if RAM allows 

 

from sklearn.feature_extraction.text import TfidfVectorizer 

from sklearn.metrics.pairwise import cosine_similarity 

 

# Clean genre strings 

movies['genres'] = movies['genres'].str.replace('|', ' ', regex=False) 

# TF-IDF transformation 

tfidf = TfidfVectorizer() 

tfidf_matrix = tfidf.fit_transform(movies['genres']) 

 

cosine_sim = cosine_similarity(tfidf_matrix) 

indices = pd.Series(movies.index, index=movies['title']).drop_duplicates() 

 

def recommend_movie(title, num_recommendations=5): 

    if title not in indices: 

        return "Movie not found in dataset." 

    idx = indices[title] 

    sim_scores = list(enumerate(cosine_sim[idx])) 

    sim_scores = sorted(sim_scores, key=lambda x: x[1], reverse=True) 

    sim_scores = sim_scores[1:num_recommendations+1] 

    movie_indices = [i[0] for i in sim_scores] 

    return movies['title'].iloc[movie_indices].tolist() 

print("      Recommendations for 'Toy Story (1995)':") 

recommendations = recommend_movie('Toy Story (1995)', 5) 

print(recommendations) 

for i, movie in enumerate(recommendations, 1): 

    print(f"{i}. {movie}")