Journal of Open Humanities Data

7 papers14 directions
The Journal of Open Humanities Data serves as a specialized venue for the publication of datasets and computational methods that bridge the humanities and artificial intelligence. Within the machine learning community, the journal is recognized for its focus on the development of robust natural language processing (NLP) techniques for orthographically unstable historical corpora and the application of linguistic priors to improve optical character recognition (OCR). It provides a platform for research into automated archival pipelines and the computational reconstruction of historical lacunae, facilitating the scaling of historical research through algorithmic approaches. The venue’s contributions to AI research include the release of specialized datasets such as CuneiML for cuneiform studies and multimodal datasets for situated tasks. Research published here often addresses technical challenges like deep learning for long-tailed distributions and neutralizing spurious contextual correlations in computer vision. Prolific authors such as Ari Vesalainen, Filip Ginter, and Jenna Kanerva contribute to a body of work that includes the creation of historical migration datasets and the development of gazetteers for ancient urban landscapes. By focusing on the representation of complex historical events and the preservation of information in bird's-eye-view representations, the journal supports the development of AI models capable of handling the unique complexities of longitudinal and historical data.

Papers by Year

2025 1
2024 3
2023 2
2022 1

Top Research Directions

Research directions most represented by papers from this venue.

Computational Restoration of Gaps in the Historical Record 1
Modeling Conversation Through Functional Dialogue Acts 1
Perspective Rectification and Planar Mapping via Homographies 1
Robust Information Extraction from Heterogeneous Document Layouts 1
Automated Transcription and Translation of Historical Archives 1
Geometric Rectification of Distorted Document Images 1
Overcoming Dataset Licensing Restrictions in AI Research 1
Vertical Feature Collapse in Spatial Projections 1
Modeling Text as Syntactic Dependency Graphs 1
Mitigating Trajectory Fragmentation under Occlusion 1
Overcoming the Manual Linguistic Annotation Bottleneck 1
Resolving Overlapping Speech in Multi-Speaker Audio 1
Normalizing Orthographic Variation in Historical Texts 1
Translationese Artifacts in Multilingual Language Model Evaluation 1

Top Authors

AuthorPapers
Enrique Jiménez1
Filip Ginter1

Impactful Papers

Papers from this venue that shaped research clusters.

Influential

Papers

TitleYear
Creating a Historical Migration Dataset from Finnish Church Records, 1800–19202025
A Survey of Body Part Construction Metaphors in the Neo-Assyrian Letter Corpus2024
Transliterated Cuneiform Tablets of the Electronic Babylonian Library Platform2024
When Text and Speech are Not Enough: A Multimodal Dataset of Collaboration in a Situated Task2024
CuneiML: A Cuneiform Dataset for Machine Learning2023
The Mesopotamian Ancient Place-names Almanac (MAPA): A Gazetteer of the Uruk Urbanscape in the Age of Empires2023
Representing Persons and Objects in Complex Historical Events using the Event Model F2022