Accurate forecasting course
The course “Predictive Big Data analytics as support in financial sector decision making” was created by the NCCT researchers Klavdia Bochenina, Anna Kalyuzhnaya and Denis Nasonov in close collaboration with Bank Saint Petersburg’s own experts. This allowed for the course to be adapted to real-life problems and be presented in such a way that not only programmers but also employees of other departments could benefit from it.
“This is not the first collaboration that we’ve had with the Bank’s specialists. We’ve created a variety of new products based on state-of-the-art machine learning algorithms and complex financial systems models. This year, for instance, we decided to focus on branch network optimization and recommendation algorithms. Collaboration between scientists and financial experts led us to realize we all need to speak the language of data processing and analytics. And that, in its turn, was our main idea for this professional development course,” comments Klavdia Bochenina, curator of the course, Head of Research at the NCCT laboratory.
The main feature of the course is that students get involved in creating and testing predictive models that are used in today’s banking. It will be the first step towards a new generation of cognitive decision making support systems, meaning that the Bank’s employees will be able to learn to solve complex problems in the financial industry with the use of machine learning and cognitive technologies.
Knowledge that works
According to the Bank’s employees, there were quite a lot of people willing to take part in the course. Full access, however, was granted to a limited number of participants, which is why out of over 100 employees who attended theoretical seminars in the Bank only 44 got the chance to participate in ITMO University’s practical sessions.
The course curriculum covered Big Data mining, predictive analytic models for financial sector data, infrastructure and technical support of the financial sector Big Data analysis, predictive modeling of the individual credit default, analysis and predictive modeling of client characteristics based on corporate and open data.
“It’s a good course, easy-to-understand and very practice-oriented. Most of all I liked the time-series data part. I’ve already started applying what I’ve learned, this time-series part equipped me with the basics of analytics. As far as I know, there were many applicants, but, unfortunately, my colleagues from other departments only got to the theoretical part. I hope it’s not the last time the program is held so that everyone will have a chance to try their hand at practical tasks,” shares Timur Tarpischev, Chief Specialist at Bank Saint Petersburg.
Results of the program
One of the challenges was adapting the course to the interests and skills of the employees.
“ITMO University has created many professional development programs with numerous partners, but this is the first cognitive technologies project. We were faced with a challenge to make the course universally useful and accessible for people with diverse backgrounds. We even changed some aspects along the way to improve the course. I hope, we’ve accomplished our goal,” says Alexey Dukhanov from the NCCT.
Ksenia Khoinatska, Chief Specialist of the Staff Development and Internal Communications Department at Bank Saint Petersburg, states that it was a successful program despite all the challenges. The project contributed to the Bank’s Digital Academy, a platform created to improve the employees’ digital skills. Since both partners were pleased with the results, the cooperation will continue and include new frameworks of collaboration.
Translated by Catherine Zavodova
