International Journal of Computational Linguistics Research Orlando Belo, Paulo Rodrigues, Rui Barros, Helena Correia 7 4 2016 http://www.dline.info/jcl/fulltext/v7n4/jclv7n4_1.pdf Today dashboards are a clear factor of differentiation for decision-makers, providing critical business indicators in useful time. They are very important data visualization mechanisms, simple to interpret, easy to deal, and fast on showing pertinent data. Dashboards provide means for reducing time of information analysis, disposing business indicators in quite understandable and attractive interface platforms – not the ability to evolve, to adapt by themselves to new user needs, tendencies or preferences. In order to provide such abilities, we designed and implemented a personalization system for data visualization platforms. All system’s services, ranging from data gathering to data profiling, were developed using a community of autonomous agents with the ability to act together using a private data cloud as a cooperative environment. In this paper we present how the system was conceived, its functional architecture and services, and how agents were implemented, especially the ones related to the personalization of business dashboards.