Movie analytic using CNN and VADER

Abdul Latif, Aina Syazzween Suraya (2023) Movie analytic using CNN and VADER. Project Report. Melaka, Malaysia, Universiti Teknikal Malaysia Melaka. (Submitted)

	Text (24 Pages) Movie analytic using CNN and VADER.pdf - Submitted Version Download (889kB)
	Text (Full Text) Movie analytic using CNN and VADER.pdf - Submitted Version Restricted to Registered users only Download (4MB)

Abstract

The Movie Analytic Using CNN And VADER system incorporates three fundamental modules which are movie recommendation, genre prediction, and review analysis. Its core objectives involve the development of a sophisticated movie analytics platform, utilizing cutting-edge techniques such as Convolutional Neural Networks (CNN), the VADER lexicon, and content-based filtering. Within the movie recommendation module, content-based filtering drives personalized movie suggestions based on user preferences and viewing history, enhancing the overall viewing experience. For genre prediction, four distinct CNN models, namely AlexNet, Keras Sequential, LeNet, and VGGNet 16, are employed, with AlexNet leading with a 91% genre prediction accuracy, offering automated genre tagging for streamlined content management. The movie review analysis module, leveraging the VADER lexicon, impressively achieves 98.67% accuracy in assessing sentiment, providing valuable insights into audience reactions and opinions. This system serves as a comprehensive toolset for movie analysis, benefiting a wide array of stakeholders including business organizations for data-driven insights, production companies for content tagging, and local residents seeking personalized movie recommendations. Ongoing refinement and adaptation to evolving user preferences and technological advancements are pivotal for continuously elevating its performance in the dynamic field of movie analytics.

Actions (login required)

View Item