Дослідження методів штучного інтелекту для розробки рекомендаційної системи в сфері аніме

Abstract

The work examines the intersection of anime as a cultural phenomenon and the evolution of recommendation systems enhanced by artificial intelligence. It specifically highlights collaborative and content-based filtering methods, emphasizing their role in tailoring recommendations to diverse viewer preferences. The challenges such as the "cold start" problem and the potential for a "filter bubble" are discussed, alongside the benefits of hybrid systems that integrate both approaches for more accurate and diverse recommendations. The work underscores the necessity for continuous technological advancement to adapt to the dynamic preferences of the anime audience.