Russian Federation Sberbank Tender

Sberbank has created a new model of machine learning Auto ML announced Deputy Chairman of the Executive Board of Sberbank Anatoly Popov at the Russian Investment Forum.Auto ML is an algorithm that can create other models. And in their turn they solve application tasks for example forecasting client creditworthiness while applying for a loan or helping to separate law-abiding clients from law-breakers.In January 2019 the first pilot was held Auto ML algorithms were used to create several baseline models first versions of Next BestAction class targeting of sales campaigns.The algorithm allows one of the most important steps of creating machine learning models used in business to be optimised notably creation of the baseline model first version of the model that will be developed with human interaction. The quality of baseline models created by Auto ML can be compared to the quality of the manually-developed models. At the same time the algorithm works 10-15 times faster than a human.The received results prove the possibility of using automatic modelling technology to rapidly form basic models to process data and use it to launch sales campaigns for Sberbanks corporate and investment business.