Practical seminar on predictive analytics solutions in business was held in Graduate School of Management SPbU together with Trinity Ltd and IBM.

 

Speakers of the seminar presented their views on current trends in the field of predictive analytics, and discussed the prospects of innovative technologies and innovative business models development that have a high market potential.

  

Also, at the seminar such issues as the use of predictive technologies in various fields of business; use of predictive analytics for sales forecasting, customer segmentation, analysis of consumer baskets, identifying fraudulent transactions and much more were discussed.

  

Vladimir Bykov, sales director at Trinity, opened the seminar with a presentation on trends in predictive analytics and its role in an effective business. Vladimir presented an overview of industries where predictive analysis is demanded and has been used successfully. Among them is the financial sector, where the forecasting methods help to assess the solvency of borrowers, identify fraudulent schemes in the insurance, lending, online payments, sales and marketing, where forecasting is used to retain customers, as well as to render possible customer operations in retail, telecom. Vladimir also told about the companies successfully using completely new business models, such as Uber (mobile application, which allows consumers to submit a trip request for a taxi or a private driver), INTOUCH and its auto insurance program, airbnb, e-bay, and others.

  

Maxim Goncharov, an IBM expert in predictive analytics, spoke about one of the solutions for predictive analytics - IBM company SPSS Modeler.

 

Head of Trinity predictive analytics department, Evgeniya Evdokimova presented results of specific projects in banking and retail, performed by the Trinity company.

 

Vladimir Gorovoj, senior lecturer at Graduate School of Management SPbU made a presentation “Machine learning methods for business application”.

 

In his presentation he gave examples of the use of predictive techniques in areas such as product recommendations, customer classification, predicting the probability of users churn in telecommunications, machine learning in metallurgy, as well as spoke about the successful projects of the Yandex company.

 

Evgeny Vinogradov, head of data warehouse development department, Yandex.Money explained how machine learning techniques are used to monitor tens of thousands of counterparties receiving online payment through the service. Analyzing text and visual content of websites, special algorithms automatically check it for compliance with the international payment systems requirements. In conjunction with other monitoring tools this gives the best results.

  

Maxim Kononenko, B2B director at Dom.ru Business, told how the telecommunications operator Dom.ru Business Interzet uses the BI-system, which analyzes the number of indicators (subscriber base, rates, packages, business indicators, sales through different channels) for a certain period. According to the analysis system builds a forecast with daily plans showing which indicators and how need to be changed in order to achieve the desired result.

 

The workshop was an effective platform for professional communication. During the lively discussion, participants discussed the prospects of predictive analytics usage in particular companies and industries.