@article{ART003317647},
author={Jong-Hyun Kim},
title={A Korean Movie Genre Prediction Model Using Multi-Representation Learning},
journal={Journal of The Korea Society of Computer and Information},
issn={1598-849X},
year={2026},
volume={31},
number={3},
pages={139-147}
TY - JOUR
AU - Jong-Hyun Kim
TI - A Korean Movie Genre Prediction Model Using Multi-Representation Learning
JO - Journal of The Korea Society of Computer and Information
PY - 2026
VL - 31
IS - 3
PB - The Korean Society Of Computer And Information
SP - 139
EP - 147
SN - 1598-849X
AB - In this study, we propose a multi-representation learning model that integrates heterogeneous information—movie poster images, audience reviews, and textual phrases extracted from posters—to predict the genres of Korean films. While visual features are learned from posters using a CNN, relying solely on posters may fail to fully capture genre context, especially when discrepancies exist between the director’s intended genre and the audience’s perceived genre. To address this limitation, we additionally construct a Word2Vec-based review analysis model and an LSTM-based text genre prediction model using OCR-extracted poster phrases. By integrating the genre probabilities from the three models, the proposed approach achieves approximately 75% overall stable performance, effectively compensating for genre-specific errors compared to the single poster-based CNN model (approximately 78%). The proposed multimodal framework enhances the interpretability and reliability of genre prediction, and future work will focus on expanding genre categories and refining the weighting of OCR-based textual features to further improve performance.
KW - Movie genre prediction;Multi-representation learning;Convolutional Neural Network;Natural Language Processing
DO -
UR -
ER -
Jong-Hyun Kim. (2026). A Korean Movie Genre Prediction Model Using Multi-Representation Learning. Journal of The Korea Society of Computer and Information, 31(3), 139-147.
Jong-Hyun Kim. 2026, "A Korean Movie Genre Prediction Model Using Multi-Representation Learning", Journal of The Korea Society of Computer and Information, vol.31, no.3 pp.139-147.
Jong-Hyun Kim "A Korean Movie Genre Prediction Model Using Multi-Representation Learning" Journal of The Korea Society of Computer and Information 31.3 pp.139-147 (2026) : 139.
Jong-Hyun Kim. A Korean Movie Genre Prediction Model Using Multi-Representation Learning. 2026; 31(3), 139-147.
Jong-Hyun Kim. "A Korean Movie Genre Prediction Model Using Multi-Representation Learning" Journal of The Korea Society of Computer and Information 31, no.3 (2026) : 139-147.
Jong-Hyun Kim. A Korean Movie Genre Prediction Model Using Multi-Representation Learning. Journal of The Korea Society of Computer and Information, 31(3), 139-147.
Jong-Hyun Kim. A Korean Movie Genre Prediction Model Using Multi-Representation Learning. Journal of The Korea Society of Computer and Information. 2026; 31(3) 139-147.
Jong-Hyun Kim. A Korean Movie Genre Prediction Model Using Multi-Representation Learning. 2026; 31(3), 139-147.
Jong-Hyun Kim. "A Korean Movie Genre Prediction Model Using Multi-Representation Learning" Journal of The Korea Society of Computer and Information 31, no.3 (2026) : 139-147.
