New tendencies in linear prediction of events

Carmen ROTUNĂ, Antonio COHAL, Ionuț SANDU, Mihail DUMITRACHE
National Institute for Research and Development in Informatics ˗ ICI Bucharest,,,

Abstract: The continuous development in the field of data science has greatly transformed big data analysis. Data is the foundation of innovation, but their value comes from the information that data experts can collect and then interpret. The volume of data is constantly increasing nowadays, thus businesses could benefit from analyzing existing data to make valuable predictions about the future and to develop a coherent business plan. Time series analysis enables companies to analyze data in order to extract meaningful characteristics and generate useful timely predictions. Mainly, time-series data consists of sequences of chronologically stored observations and are generated by recording, business metrics, monitoring sensors, observing network traffic, etc. In this study, we set out to implement a forecast model for the .ro Registry and, therefore, chose the Prophet FB, because it offers an open source software tool that supports the business area and has been successfully tested in different scenarios. The results showed that Prophet can generate accurate forecasts which can be used to optimize Registry services.

Keywords: time-series, quantitative forecasting, linear prediction, FB Prophet, data model.

View full text

Carmen ROTUNĂ, Antonio COHAL, Ionuț SANDU, Mihail DUMITRACHE, New tendencies in linear prediction of events, Romanian Journal of Information Technology and Automatic Control, ISSN 1220-1758, vol. 29(3), pp. 19-30, 2019.