Contract researcher in data processing for traceology

Université de Technologie de Compiègne is seeking to fill a researcher position in data processing for the traceology of stone tools at the Roberval Laboratory, Mechanical Engineering Department (IM).

This recruitment is part of the WOLF research project and is funded by the Sorbonne University Alliance (ASU).

Missions

Actively contribute to research work for the WOLF project

Ensure the development of a methodology for characterising and classifying wear marks on prehistoric stone tools.

Activities

• Research significant metrological parameters that can be used to distinguish the topographical signature associated with different configurations of stone tool use

• Develop a model for automatic recognition and classification of surface alterations on stone tools using artificial intelligence algorithms

• Validate algorithms through blind testing

• Write reports, articles and tutorials.

• Promote the work through communications.

• Participate in project meetings and present results.

Project background and summary

The WOLF project – Wear analysis of prehistoric tools and classification of their uses using multiscale topography measurements – is funded by the Sorbonne University Alliance as part of the Emergence call for projects.

This project brings together the Roberval laboratory at UTC and the Natural History of Prehistoric Humanities laboratory HNHP – UMR 7194 at the National Museum of Natural History.

This research project focuses on the analysis of wear on prehistoric tools and the classification of their uses using multiscale topography measurements.

Tool wear analysis applied to prehistory mainly uses qualitative methods to describe the surface alterations of tools made of stone, bone or other kinds of raw materials in order to determine tools functions and understand past human behaviors. The repeatability and thus reliability of these methods are however questioned.

It is proposed to develop a method of multiscale characterization of the topography of the altered surfaces using a corpus of existing experimental tools made from different rocks (flint, felsitic meta-rhyolith and limnosilicite), which were used to saw or scrape different materials (bone, antler, wood and reed). This corpus will be used to identify post-treatment measurement strategies in order to propose a classification of uses and find the best strategies for describing heterogeneous surfaces. Classical statistical approaches will be compared to deep learning approaches.

ADDITIONAL INFORMATION

Type of contract and expected recruitment dates

Fixed-term contract – expected duration of 18 months – to be filled in January 2026

Gross monthly salary

According to experience and funding

Hours worked

1,607 hours/year

Working environment and context

The work will be carried out at the Roberval laboratory, mechanical engineering department.

The successful candidate will join the WOLF project team, report to the project managers, maintain regular dialogue with them and work closely with the relevant contacts. A laptop will be provided for the duration of the contract.

Occasional travel is to be expected as part of the project.

We are seeking a candidate with:

knowledge about artificial intelligence algorithms and their applications and should be willing to apply it to traceology

or

skills in metrology, surface measurement in particular and statistics for surface data treatment and be willing to learn about artificial intelligence algorithms.

Proficiency in programming languages such as Python and/or R is essential.

Experience with MountainMaps and Statistica software is appreciated.

Diploma: PhD

Field: Mechanics or archaeology.