EAA 2023: Session #408

High-quality research data forms the basis of scientific research. To guarantee and objectively evaluate the quality of research data, some research projects and institutions are working towards the development of measures and applicable control mechanisms. Quality control is essential and should ideally be carried out at various stages in the data life cycle. Without it, research data cannot be re-used beyond the project's original scope and cannot be FAIR or LOUD.
However, appropriate metrics for ensuring high-quality data and their integration into the research process is still poorly discussed beyond single projects or across institutions. In general, only formal and technical aspects of the data and metadata (e.g. the resolution of images or naming conventions) are integrated directly into the processing pipeline in an automated way. While virtually no metrics for evaluating the content or scientific validity of data currently exist.
This session addresses quality measures of data in archaeology and neighboring disciplines from multiple perspectives, inviting the wider archaeological community to explore key questions such as:
- How can we better understand and measure the data and metadata quality of archaeological data?
- How do we reflect differences in records of scientific data and interpretive or deductive reasoning?
- Which methods ensure and improve data quality?
- How can we implement quality checks in automated data collection workflows?
- How can we optimize archaeological data and metadata for re-use?
We also invite papers that stem from specific case studies or projects that address the above-mentioned questions in a practical manner. For instance, the experience of large scale research projects working with big data, the utilization of rapid inventory of large quantities of data, the use of new methods in the processing pipeline like machine learning and statistics, or the presentation of best practices from established guidelines.

FAIR, quality control, machine learning, mass recording, large scale projects, BIG data

Regular session

  Abstract (PDF)

Fabian Riebschläger¹

David Novák²
Katja Roesler³
Frederic Auth⁴

Quality Assurance and Big Data: Developing a machine learning model for the identification of archaeological landscape features

Accuracy of 3D point clouds generated by photogrammetry by comparison with the laser scanning

Imagining artefacts: organising data for ML-based artefact analysis

Recording the ritual - A digital workflow to record large quantities of pottery

Application of Machine Learning and Big Data Analytics Techniques in Archaeological Research

A Machine Learning Approach for the Dimensional Reconstruction of the Laminar Artefacts of the MPPNB sites in the Southern Levant

Improving interdisciplinary data acquisition methodologies by algorithmically piecing together archaeological 3D puzzles