Pengembangan Website Rekomendasi Film Dan Tv Series Berbasis Mood Menggunakan Metode Scrum

Abstract

The rapid growth of the digital entertainment industry has increased user demand for content recommendation platforms that are more personal and relevant. However, most existing streaming platforms rely on watch history and ratings as recommendation parameters, without considering users' current emotional conditions. This research develops FLIX, a mood-based film and TV series recommendation web platform built using React 19, Vite, Tailwind CSS, Express.js, and PostgreSQL via Supabase, integrating TMDB API, GIPHY API, and Gemini AI API. The development process adopts the Scrum framework, carried out through three Sprints covering eleven product backlog items prioritized using the MoSCoW method. Data collection was conducted through literature study and interviews with five respondents. System design includes Use Case Diagram, navigation structure, and Entity Relationship Diagram. System testing was performed using the Black Box Testing method with Postman assistance, and the results show that all developed features function well in accordance with user needs. This research proves that the Scrum method is effective in developing the FLIX platform as a functional mood-based film and TV series recommendation system accessible via the web.