A Study on Deriving Game Scenario Similarity Based on the Convergence of Textual Meaning and Event Structure

Q: What does this research mean for the field?

A Late Fusion framework combining textual semantic features and SVO-based event structures achieves near-perfect accuracy (ROC-AUC 0.9966) in identifying game scenario similarities and compressing source candidates. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

Kim Yun-Ho; Hye-Young Kim

doi:10.7583/jkgs.2026.26.4.25

August 31, 2026Journal of Korea Game Society

A Study on Deriving Game Scenario Similarity Based on the Convergence of Textual Meaning and Event Structure

View at Publisher

Ask AI

Bookmark

Share

Key Points

To establish a late fusion framework for evaluating game scenario similarity by combining textual semantic features with Subject-Verb-Object (SVO)-based event structures.
Extracted text channel features utilizing SBERT, Smith-Waterman alignment, Dynamic Time Warping (DTW), LMSS, and coverage metrics.
Modeled story narratives into an SVO event Tri-Graph and processed it using a Graph Isomorphism Network (GIN) encoder.
Benchmarked the framework using an English visual novel controlled transformation dataset.
Achieved an ROC-AUC of 0.9966 on the controlled transformation benchmark dataset.
Attained a Top-1 accuracy of 1.0 in identifying matching narrative scenarios.
Demonstrated effective screening capability for filtering similar story cases and narrowing candidate source texts.

Implication

Machine learning evaluation demonstrates near-perfect scenario matching across controlled visual novel variants, indicating robust automated screening for narrative similarity.

Abstract

No abstract yet. The journal hasn't published one we can show.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Yun-Ho et al. (2026) studied this question.

synapsesocial.com/papers/6a99355b636c6408cfa7d7e6 https://doi.org/10.7583/jkgs.2026.26.4.25

View at Publisher

Ask AI

Bookmark

Share

A Study on Deriving Game Scenario Similarity Based on the Convergence of Textual Meaning and Event Structure

Key Points

Implication

Abstract

Discussion

Cite This Study

A Study on Deriving Game Scenario Similarity Based on the Convergence of Textual Meaning and Event Structure

Key Points

Implication

Abstract

Discussion

Cite This Study

Keep reading

More related papers

More from Journal of Korea Game Society

Keep reading

More related papers

More from Journal of Korea Game Society