TY - JOUR
T1 - The Same Thing - Only Different
T2 - 6th Workshop on Narrative Extraction From Texts, Text2Story 2023
AU - Liu, Chang
AU - Shmilovici, Armin
AU - Last, Mark
N1 - Funding Information:
This research was partially supported by the Israeli Council for Higher Education (CHE) via the Data Science Research Center, Ben-Gurion University of the Negev, Israel.
Publisher Copyright:
© 2023 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
PY - 2023/1/1
Y1 - 2023/1/1
N2 - Story types depict the development of movie stories in terms of the protagonist's character traits and the motivations that drive him in facing his challenges. We define a novel task of story type classification of movies and propose a lightweight machine learning solution. A crowdsourcing experiment was performed to label 45 movies for their perceived story types. We extract movie features that indicate different aspects of the movie characters and apply Decision Tree and Naive Bayes classification algorithms. Although the labeled dataset is relatively small, the story type classification accuracy is significantly above the baseline with the F1 measure in the range of [0.63-0.77]. The preliminary results suggest that simple movie features can be used by machine learning algorithms to detect the abstract concepts of story types.
AB - Story types depict the development of movie stories in terms of the protagonist's character traits and the motivations that drive him in facing his challenges. We define a novel task of story type classification of movies and propose a lightweight machine learning solution. A crowdsourcing experiment was performed to label 45 movies for their perceived story types. We extract movie features that indicate different aspects of the movie characters and apply Decision Tree and Naive Bayes classification algorithms. Although the labeled dataset is relatively small, the story type classification accuracy is significantly above the baseline with the F1 measure in the range of [0.63-0.77]. The preliminary results suggest that simple movie features can be used by machine learning algorithms to detect the abstract concepts of story types.
KW - Computational Narrative Understanding
KW - Story Analytics
KW - Story Type Classification
UR - http://www.scopus.com/inward/record.url?scp=85154545303&partnerID=8YFLogxK
M3 - Conference article
AN - SCOPUS:85154545303
SN - 1613-0073
VL - 3370
SP - 93
EP - 101
JO - CEUR Workshop Proceedings
JF - CEUR Workshop Proceedings
Y2 - 2 April 2023
ER -