Job offer

The TopOL (Top of the Lake) project is a research project funded by the French National Research Agency (ANR) (AAPG 2025, CE23 - Artificial Intelligence and Data Science), with a duration of 48 months. Its objective is to develop intuitive tools for exploration, querying, and analysis of large heterogeneous data lakes, targeting non-technical users, in particular investigative journalists.

The project relies on rich entity-relationship representations, modeled as property graphs, constructed from heterogeneous data sources (text, documents, tables, semi-structured data). The new interactive exploration approaches developed in this project will include similarity search and automatic generation of exploration pipelines over entity-relation graphs.

Laboratoire d'Informatique Fondamentale d'Orléans is a joint laboratory of Université d'Orléans and INSA Centre Val de Loire, comprising around 100 active researchers and faculty members, including 52 permanent staff members. The laboratory is located on two campuses, in Orléans and Bourges.

LIFO conducts both theoretical and applied research activities, organized around three complementary research areas: Cybersecurity, Artificial Intelligence & Data Science, and Algorithms & Modelling. These research areas are supported by five thematic teams.

The position will be hosted at LIFO (Université d'Orléans), within the PAMDA team of the Artificial Intelligence & Data Science research area. The recruited researcher will work on site.

The recruited researcher will work in close collaboration with the local project leaders, Mirian Halfeld Ferrari and Patrick Marcel, as well as with project partners at the University of São Paulo (Brazil), and will contribute to the supervision of Master-level internships.

Title of the postdoctoral project: Interactive Exploration and Discovery Generation over Graph Data Lakes

The recruited researcher will contribute to the design and implementation of interactive exploration mechanisms for entity-relationship graphs, with the following objectives:

• Design similarity operators applied to these graphs (over entities (nodes), relationships, or paths), leveraging attributes (textual, temporal, geographical, numerical) or graph topology.
• Design insight collectors capable of extracting patterns, comparisons, trends, anomalies, etc., either significant or surprising, in these entity-relationship graphs.
• Design approaches for the automatic generation of exploration pipelines combining the collectors defined, for instance using deep reinforcement learning, that was already successfully applied to interactive data exploration.

The expected contributions cover both theoretical aspects (definition of operator syntax and semantics) and practical aspects (implementation, experimentation, evaluation in interactive exploration scenarios, and integration into the project platform)., Application files must include a CV, a cover letter, and ideally academic references and the PhD defense report.

Research Field Computer science » Computer systems

Education Level PhD or equivalent

Skills/Qualifications

Candidates must hold a PhD in computer science, and have strong expertise in graph databases, query languages, graph analytics, and interactive data exploration.

Good programming skills (Python, Java, Scala, or equivalent) are expected, as well as the ability to conduct experimental research and a good command of English.

Languages ENGLISH

Level Good

Research Field Computer science » Computer systems, Applications will be evaluated based on the theoretical and technical skills relevant to the project, the quality of the application, and the candidate's ability to work both independently and collaboratively.