MOVIE RECOMMENDER SYSTEM USING SENTIMENT ANALYSIS
Keywords:
Recommendation Systems, Sentiment Analysis, Ensemble Learning, KNN Algorithm, Collaborative Filtering, Content-Based Filtering, Hybrid Systems, User Experience.Abstract
In today's digital landscape, users are
constantly bombarded with countless
options on platforms such as streaming
services, e-commerce sites, and social
media. This sheer volume of content can
lead to decision fatigue, making it difficult
for users to discover content that truly
resonates with their interests. To alleviate
this challenge, Recommendation Systems
(RS) have become pivotal in delivering
personalized suggestions, thereby
improving user satisfaction and
engagement. Conventional
recommendation strategies typically fall
into two primary categories:
Content-Based Filtering (CBF): Suggests
items similar to those a user has previously
interacted with, using attributes like genre,
director, or cast. While effective in
personalization, CBF can suffer from
limited diversity—often referred to as the
"over-specialization" issue. Collaborative
Filtering (CF): Relies on analysing user
interactions, such as ratings and reviews, to
identify patterns and recommend items
enjoyed by similar users. However, CF
faces challenges like the cold-start problem,
where recommendations are hindered due
to insufficient data on new users or items.
