Movie Recommendation System

The video tutorial explains how to build an item-based collaborative filtering recommendation system for movies. It starts with creating a dummy rating matrix, standardizing user ratings, and calculating movie similarities using cosine similarity. The tutorial then demonstrates how to make movie recommendations based on user input and improve accuracy by adjusting similarity scores. Finally, it summarizes the steps to finalize the recommendation system.

What is the primary approach used in this project for recommending movies?

• Hybrid filtering

• User-based collaborative filtering

• Item-based collaborative filtering

• Content-based filtering

Why is it not logical to replace missing ratings with zero?

• It makes calculations slower

• It increases the dataset size

• It complicates the matrix

• It assumes the user disliked the movie

What is the purpose of standardizing user ratings?

• To remove non-values

• To simplify the matrix

• To correct for user bias in ratings

• To increase the dataset size

Which metric is used to calculate similarity between movies in this project?

• Cosine similarity

• Euclidean distance

• Jaccard index

• Manhattan distance

Why is it necessary to transpose the matrix before calculating similarity?

• To reduce computation time

• To focus on user similarities

• To calculate similarity based on items

• To simplify the matrix

What inputs are required from the user to make movie recommendations?

• User's age and location

• User's watch history

• Movie name and user's rating

• User's favorite genre

What issue was identified when testing the recommendation system with a user who disliked a movie?

• The system recommended the same movie

• The system recommended unrelated genres

• The system did not recommend any movies

• The system recommended only action movies

How is the threshold for user preference determined in the refined recommendation system?

• By the average rating

• By the median rating

• By the highest rating

• By a fixed value of 2.5 stars

What adjustment is made to the similarity score to handle user preferences more accurately?

• Subtracting the mean rating

• Dividing by the total number of ratings

• Multiplying by the rating minus 2.5

• Adding a constant value

What is the final step in refining the recommendation system for multiple movie ratings?

• Filtering out low-rated movies

• Sorting movies alphabetically

• Row-wise summation of similarity scores

• Calculating the average score