EXplainable and parsimonious Preference models to get the most out of Inconsistent DAtabases

Nowadays, data is increasingly becoming a commodity of tremendous value for many real-world domains. Due to the heterogeneity and the imprecision of sources, gathered information is often incomplete and inconsistent. For example, social network users can easily post their observations over some events, but unfortunately such posts are frequently far from being precise or correct. Additionally, different users may report on the same event from a different point of view, consequently creating inconsistent information in the network. Then, potential consumers need to decide how to use and exploit these data. For sensitive domains such as air transportation and other safety-critical activities, many important efforts have been made to recording reports of abnormal events (e.g. ASRS and ECCAIRS databases). Clearly, computational techniques would be desired to assist experts in exploring, analyzing and dealing with such kinds of possibly inconsistent raw data. In this respect, the project EXPIDA (EXplainable and parsimonious Preference models to get the most out of Inconsistent DAtabases) aims to develop a series of principled and powerful methods to better analyze and particularly to explain the actions that we can take over uncertain and inconsistent data to get the most out of these data. More explicitly, the scientific guiding principle of EXPIDA aims to provide database users with a rich family of interactive explainable methods to effectively handle imperfect data, to assist the analysis of aviation incident data, and thus to better support the research on civil air flight safety. The outcomes of the project will include different rich explainable inconsistency-tolerant methods, as well as their algorithms and implementations together with the datasets generated from the ASRS database and other domains to validate EXPIDA results.


Scientific Responsible for CRIL :
Participants :
Partners :
Duration :
2023-2027