PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
February 9, 20261 citationsOpen Access

Data-Driven Reconstruction of the Singapore Stone Using Numerical Imputation

Data-Driven Reconstruction of the Singapore Stone: A Numerical Imputation Method of Epigraphic Restoration

View Full Paper
Ask AI
Bookmark
Share

Authors

TZTehreem ZahraFCFrancesco Perono CacciafocoMZMuhammad Tayyab Zamir

Discussion

Loading...

Member takes

Overview

A numerical imputation method improves epigraphic restoration of the Singapore Stone, suggesting innovative techniques for ancient texts.

Key Points

  • This research aims to develop a data-driven method for reconstructing incomplete inscriptions, specifically the Singapore Stone.
  • Modeling the inscription as a system of categorical symbols with original spacing.
  • Training a model using observed symbols to learn local transition statistics.
  • Generating probabilistic hypotheses for missing characters while maintaining transparency with experts.
  • Achieved mean top-one error of 53.3% in masked-character recovery.
  • Performance was notably worse than traditional baseline methods.
  • The procedure enhances interaction and auditing capabilities for experts.

Cite This Study

Zahra et al. (2026) studied this question.

synapsesocial.com/papers/69897a06f0ec2af6756e8313https://doi.org/10.3390/info17020170
View Full Paper
Ask AI
Bookmark
Share