Analyzing Movie Script Similarity: A Weighted Multilayer Network Approach

Majda Lafhel (1) , Hocine Cherifi (2) , Mohammed El Hassouni (1) , Benjamin Renoust (3, 4, 5)

1 UM5 - Université Mohammed V de Rabat [Agdal]
2 ICB - Laboratoire Interdisciplinaire Carnot de Bourgogne
3 Institute for Datability Science
4 JFLI - Japanese French Laboratory for Informatics
5 NII - National Institute of Informatics

Measuring similarity between multilayer networks is a challenging task that involves capturing various layers and relationships using distance measures. Existing techniques have limitations in comparing layers with the same number of nodes and ignoring inter-relationships. In this research, we propose a novel approach for measuring similarity between multilayer networks while considering inter-relationships and weighted networks of various sizes. Our approach is applied to multilayer movie networks consisting of layers of different entities (char- acter, keyword, and location) and inter-relationships between them. Our proposed method captures both intra-layer and inter-layer relationships, providing a comprehensive overview of the multilayer network. This approach has potential applications in analyzing movie story structures and social network analysis. Our results demonstrate that our approach improves the accuracy of similarity measurement between multilayer networks.

Mots clés: Movie Similarity Multilayer network Network Distance; Movie Similarity; Multilayer network; Network Distance

Domaines: Informatique [cs]

Majda Lafhel, Hocine Cherifi, Mohammed El Hassouni, Benjamin Renoust. Analyzing Movie Script Similarity: A Weighted Multilayer Network Approach. French Regional Conference on Complex Systems, CSS France, May 2024, Montpellier, France. ⟨hal-04553586⟩