Movie Genres Classification System using AI Model

As multimedia information has grown at an unprecedented rate in the digital age, streamlined ways of managing and retrieving such information, including classifying movies according to their genre has become a crucial aspect in the personalized and content recommendation as well as managing information. Genres affect the performance of the traditional Deep Learning (DL) models because these models do not support sequential dependencies and thus work better with narrative text. In this paper, these limitations are overcome with the help of a Recurrent Neural Network (RNN)-based model which is able to capture the sequential nature of plot summaries in an effective manner. The data is preprocessed within the system in case of textual data: in terms of tokenization, removal of stop-words, and lemmatization. FastText embeddings of words are used to embed words in high dimensional semantic spaces, even out-of-vocabulary words. A Bidirectional RNN (Bi-RNN) makes sure that it grasps context in both directions, whereas an attention mechanism places the most informative areas of the text. The proposed Bi-RNN with attention-based architecture greatly enhances classification of movie genre. The proposed model prove that it has the capability to scale and create effective contribution to a multi genre movie classification task.