An Architecture of a Data Lake for the Sharing, Agricultural Knowledge in Burkina Faso

---

Résumé

With the advent of big data today, agricultural research institutes have stored enough data in databases at every level. This data contains a lot of knowledge that is often not well known by researchers and on the other hand, it may contain hidden semantic relationships. Therefore, it is important to find an adequate storage system for these data in order to discover new useful information that can be shared between the different actors in the field to improve agricultural productivity in Burkina. The concept that best responds to this storage and analysis problem is the data lake. Indeed, data lakes offer the possibility of storing a large volume of data in any format and data structure. They also provide services for data access and analysis. Our work revolves around the integration and storage of agricultural data (structured, semi-structured and unstructured). With the plurality of agricultural storage systems, the heterogeneity of the actors in the field and the disparity of the tools leads us to think that the data lake is a solution. The purpose of this essay is to investigate data lakes and their requirements while also putting out a broad framework for an agricultural data lake in Burkina Faso.

Mots-clés

data lakes, agricultural data, data architecture, data sharing, agricultural knowledge, heterogeneous data storage, unstructured and semi-structured data, agricultural research institutions, Burkina Faso