1st Open Call
Station-based weather observations serve as a crucial input to different analysis and forecast models. To use these data in a reasonable way, outliers have to be identified. There are two types of outliers: extremes, which represent rare and intense events, and erroneous values. The challenge proposed by UBIMET is to derive an algorithm capable of detecting incorrect values while retaining extremes.
AMIGO solution incorporates the best of their expertise into one disruptive algorithm, SOON (Station Observation Outliers fiNder). SOON is a highly advanced classificator based on the team’s experience in meteorological data, statistical analysis, and machine learning. It boosts state-of-the-art methods with highly innovative tools, resulting in the perfect solution for UBIMET challenge.
AMIGO is the first Italian consultancy SME in the field of Climate Services. AMIGO operates worldwide, carrying on projects both at a national and an international level, and collaborates with academic institutions, such as the Imperial College London, as well as non-academic organizations (European Commission, United Nations World Food Program). Working for, and in close collaboration with, large international organizations allowed AMIGO to grow, increase yearly turnover, the staff, build a solid network of clients and a strong project portfolio in relation to climate services especially for the insurance sector and agribusiness.