Регистрация / Вход
Прислать материал

Collaborative Filtering Recommendation Systems: A Review

Фамилия
Efa
Имя
Bodena Terfa
Отчество
Номинация
Информационные технологии
Институт
Институт информационных технологий и автоматизированных систем управления (ИТАСУ)
Кафедра
Автоматизации проектирования и дизайна
Академическая группа
Innovative Software Systems: Design, Development and Applications
Научный руководитель
PhD, prof. Anton Aristov
Название тезиса
Collaborative Filtering Recommendation Systems: A Review
Тезис

Nowadays, people are able to get easy access to and spread their activities on the web because of the expansion of the Internet and the availability of smartphones. However, it has made much more difficult to explore useful information from all the available online information. The overwhelming amount of data requires approaches for efficient information filtering. Recommender system is one of the software tools and techniques dealing with this problem. Currently, there are various recommendation systems available in academia and industry to provide users with personalized services. Among them, the most widely used technique in E-commerce is collaborative filtering. Collaborative filtering approaches provide online users with opportunities to discover new information about items based on personal interests. Collaborative filtering is a technique of making automatic predictions about the interest of a user by collecting preferences or taste information from multiple users. Unfortunately, collaborative filtering systems have technical issues such as sparsity, scalability, optimization and recommendation quality. Researchers have successfully applied collaborative filtering algorithms to solve sparsity and scalability issues. In this review, we thoroughly assessed various research papers about collaborative filtering systems and found that; still performance optimization is needed to make efficient recommendation with large datasets. This paper suggests hybrid approaches to deal with the collaborative filtering systems performance optimization with large datasets in order to make highest quality recommendation.