Daniel Neagu: Abstract

Due to recent advances in data storage and sharing for further data modelling, and also due to more accessible, user-friendly data mining tools used to generate large amounts of models from the data collections, there is an increasing need for flexible and consistent processing of data and model representations. In order to better manage such data collections and related models, new functionalities aiming to make high quality data and models reusable sources of information are requested. We introduce our proposal for the data and models governance framework supported with examples from the predictive toxicology domain. The aim of this work is to formalise certain processes in order to ensure better data and model management, and consequently increase confidence in the use of the data and models in predictive toxicology and chemical risk assessment. Our framework can be extended to other domains where data and models are shared and reused by larger research, regulatory and industry communities.

Organised by BCS SGAI The Specialist Group on Artificial Intelligence http://www.bcs-sgai.org