International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 11 Issue: 07 | July 2024
www.irjet.net
p-ISSN: 2395-0072
Comprehensive Comparative Study of Movie Recommendation Algorithms for Optimal Effectiveness Sanif Kandel1, Bijay Gautam2, Shrawan Thakur3, Manoj Shrestha4 1,2Research Scholar, Department of Computing, Softwarica College of IT and E-Commerce, Kathmandu, Nepal.
3,4Professor, Department of Computing, Softwarica College of IT and E-Commerce, Kathmandu, Nepal. ---------------------------------------------------------------------***--------------------------------------------------------------------photographs are available to more individuals. Netflix, Abstract - In the current landscape of technological
Amazon Prime, and Disney+ offer a massive library of movies to watch at home or on the move, changing the way people watch movies. Smartphones and high-speed internet make streaming movies on several devices easier. With this revolution in movie viewing, audiences may pick what to watch, when, and how. They can discover new releases, genres, and old favorites at their own speed. Since more individuals can view movies, they may select how.
advancement, Online platforms now dominate the moviewatching experience in the current technological world. This evolution has presented viewers with unique issues, such as the overwhelming number of movie selections and the difficulty of finding individualized content recommendations. The movie industry has yet to fully utilize recommendation algorithms, unlike food delivery and fashion e-commerce. This study examines movie recommendation algorithms, their efficacy, and user involvement to close this gap. Google Scholar, GitHub, and APIs were used to acquire data for this research. The study examined collaborative filtering, content-based recommendations, and hybrid machine learning methods. RMSE and MAE were used to evaluate each algorithm's accuracy and performance. Algorithm development ethics were extensively assessed to ensure industry best practices, user privacy, and data security. Agile and SEMMA methods ensured flexibility and reactivity during development. This research produced a powerful movie recommendation system that uses hybrid algorithms to provide individualized content suggestions. This method solves the problem of too many movies and improves the movie-watching experience, increasing user pleasure and engagement. The study shows how algorithmic advances in movie streaming might change user experiences and improves streaming user pleasure and engagement by carefully exploring and implementing algorithms. Algorithmic advances have broader consequences, highlighting the potential for data-driven solutions to promote innovation across sectors.
But as the huge number of movies gets easier to find, movie sites and streaming services face a new problem. Users can feel overwhelmed by the number of options, which can make it hard to choose the right movie for their mood or tastes. Users may become irritated if they aren't given clear instructions on how to find movies that relate to their interests. Movie recommendation services have turned to intricate algorithms to combat this issue. In order to learn about platform usage, user preferences, and user connections, these algorithms employ cutting-edge technology like machine learning and data analytics [2]. By figuring out what each user likes, the recommendation algorithms can put together lists of movies that match each person's hobbies. The main idea behind algorithms that suggest movies is that they can guess what users want and give them what they want. Content-based algorithms consider the quality of movies while making recommendations, while collaborative filtering techniques analyze the behavior of similarly situated people to make predictions. Hybrid models use all of these methods together to make a complete and accurate system for recommending movies. Companies like Netflix, Amazon Prime, and Hulu have spent a lot of time and money creating and enhancing the quality of their recommendation algorithms for the benefit of its users.
Key Words: Movie Recommendation, Prediction, Hybrid Recommendation, Retention, Ratings, Reviews
1. INTRODUCTION Movies have entertained and told stories for almost 100 years. From silent films to today's blockbusters, movies are a worldwide craze. People rushed to theaters in the early days of film to experience its wonder. Movies in a dark theater were a worldwide favorite. However, as technology has improved and the internet has grown, movie watching has altered drastically. Streaming platforms and online movie databases have started a new movie watching era. Movies can now be watched on demand. Movie-watching is becoming more flexible as theater queues and showtimes disappear [1]. Today, more
© 2024, IRJET
|
Impact Factor value: 8.226
|
ISO 9001:2008 Certified Journal
|
Page 873