IBM and The Weather Company will be collaborating to develop a new predictive model called Deep Thunder to help better predict the weather using machine learning.
In October of 2015, IBM acquired The Weather Company's B2B data business for over $2 billion and now it seems we are seeing what its analytics can accomplish using that data. On Wednesday, The Weather Company announced that it will be launching its new custom forecaster Deep Thunder, which will be hyper-local and will use historical weather data to help train machine learning models. This will allow it to help businesses better predict the real-world effects that bad weather could have on their organisation.
In March of last year, IBM and The Weather Company formed a pact that they would work together on IBM's Internet of Things (IoT) unit. The two companies decided to use The Weather Company's sensors and combine them with IBM's analytics in order to power the newly formed unit.
The Weather Company is already able to analyse over 100 terabytes of third-party data daily and its regional model are currently being used by businesses around the world to get accurate guidance on the weather and weather related events. The new models that will form Deep Thunder were all designed by IBM and were created with business in mind. They especially excel at hyper local forecasts at a 0.2 to 1.2 mile resolution.
The head of science & forecast operations for The Weather Company, Mary Glackin explained the partnership between the two companies further saying: “The Weather Company has relentlessly focused on mapping the atmosphere, while IBM Research has pioneered the development of techniques to capture very small scale features to boost accuracy at the hyper local level for critical decision making."
The machine learning-based weather impact models that will be used by Deep Thunder will allow it to predict even the slightest variations in temperature.
IBM and The Weather Company have certainly made the most out of their pact to work together on IoT and will be interesting to see how well their new forecaster is able to predict the weather.
Image Credit: JuliusKielaitis / Shutterstock