Toward a Big Data Knowledge-Base Management System for Precision Livestock Farming
Fabrice Nolack Fote, Amine Roukh, Saïd Mahmoudi, Sidi Ahmed Mahmoudi, Olivier Debauche
Abstract
Nowadays, we are in the era of advanced technologies where tremendous amount of data is produced by multiple sources such as sensors, devices, social media, user experiences, etc. Furthermore, this raw data has a low value, and major part is not really useful or important for business. One way to give an added value to this stored data is to extract useful knowledge from it, for the ending-system or the end-users by a process commonly called knowledge Discovery in Database (KDD). Smart Farming uses a large amount of connected technologies producing also a huge amount of data in order to maximize productions by reducing: human efforts, environment impact and wasting natural resources. In this paper, We develop a new data analytic architecture dedicated to Precision Livestock Farming (PLF) to improve in particular the livestock animals production, animals’ welfare, and farming processes. We present a new data processing architecture for a knowledge-base management system (KBMS) allowing to ease decision support and monitoring operations that can help farmers and stakeholders to better exploit data and have a long-term view of the evolution of the knowledge it contains. Our main contribution in the present paper is a new architecture specifically developed for the precision livestock farming integrating a periodical data reevaluation which address the problematic of data conservation and the decrease in data value over time.