Prediksi Rating Film IMDb Menggunakan Decision Tree

Abstract

Industri Film bukan hanya industri atau pusat hiburan semata melainkan menjadi pusat bisnis global. Popularitas atau kesuksesan film box office selalu menjadi perhatian di seluruh dunia. Data kesuksesan atau popularitas film saat ini tersedia secara online. IMDb merupakan satu dari sekian situs daring penyedia informasi yang berkaitan dengan film, acara televisi, yang meliputi sinopsis, daftar pemain, ulasan penilaian, dan tentunya pemberian rating film. Keberhasilan film dapat ditandai dengan perolehan rating yang tinggi. Prediksi rating film menjadi topik menarik untuk menilai keberhasilan film baik yang telah diproduksi maupun yang belum diproduksi. Pada penelitian ini, dilakukan prediksi nilai rating film menggunakan metode decision tree. Hasil dari penelitian ini diperoleh kesimpulan bahwa atribut popularitas film dan nilai vote user pada laman IMDb berpengaruh terhadap nilai rating film. Nilai akurasi penggunaan model decision tree pada data training, validasi dan testing bertuturt – turut adalah 0,7529, 0,7237 dan 0,7079. AbstractThe film industry is not just an industry or entertainment but also a global business center. The popularity or success of box office movies has always been a concern around the world. Data on the success or popularity of a movie is currently available online. IMDb is one of the many online sites that provide information related to movies, television shows, which include synopsis, cast lists, rating reviews, and of course movie rating assignments. Prediction of movie ratings is an interesting topic to assess the success of films that have been produced and those that have not been produced. Prediction of movie ratings values can be modeled through machine learning using the decision tree model. From this research, it can be concluded that the popularity of the film and the value of user votes on the IMDb page have an effect on the film rating value. The accuracy values of using the descision tree model in training data, validation and testing are respectively 0.7529, 0.7237 and 0.7079.