During movie production many review sessions take place in which many text notes on the film are generated. These notes are manually distributed among the studio departments by the production team. This is a time-consuming and error prone process and many notes are not addressed because they do not reach the appropriate departments. In this paper, a multi-label text classification system is presented to automatically distribute these notes. By combining a DistilBERT transformer model and a Logistic Regression classifier, a mean accuracy of 0.68 is achieved when assigning the review notes to the appropriate departments. The resulting deep learning approach is compared against a standard classifier showing an improvement in the results. The outcome shows that it is possible to automatically distribute movie review notes to the interested departments in an automatic way, to improve the movie production process.
GARCÉS Diego;
SANTOS Matilde;
FERNANDEZ LLORCA David;
2023-11-16
Springer Nature
JRC133693
2367-3370 (online),
https://link.springer.com/chapter/10.1007/978-3-031-42529-5_1,
https://publications.jrc.ec.europa.eu/repository/handle/JRC133693,
10.1007/978-3-031-42529-5_1 (online),
This document is only visible at the Commission level.
You are not authorized to publish or distribute it outside the European Commission.
This is a public document. You can share this publication.
Additional supporting files
| File name | Description | File type | |
|---|
