Information technology rapidly comes into the business. We can’t imagine the decision-making process without the data processing systems of high quality.
Our company is focused on methods of data mining, machine learning technology, data analysis, as well as on building predictive models and providing related consulting services.
We work with projects in different areas - telecommunications, finance, retail, clinical and social research.
Fields of activity
Our company is focused on building of analytical systems for businesses, object analysis and modeling. We also offer our clients support in the field of modeling of client’s behavior, building loyalty programs, forecasting, optimization and risk management.
We work in different directions:
Consulting and outsourcing — we help our clients to do their best in solving a problem related to the big data analysis, statistical analysis, classification etc. If necessary, we will carry out the audit of capabilities of the existing system analysts and suggest methods of improvement.
Training — we conduct individual or group courses, on which our stuff will teach all you need for data analysis.
Programming — we provide turnkey soft development and support.
Data quality assessment
Customer’s life cycle analysis
Outflow of clients analysis
Customers database analysis
Classification of clients
Data processing for clinical research
Data processing for social research
Our services include:
Audit capabilities – the study of the company in order to identify the available and potential sources of data and systems, the consolidation of which in short, medium and long-term horizons, creates conditions for increasing productivity depending on the objectives.
Data analysis – identifying hidden patterns and determinants of the behavior of consumers or other stakeholders of the enterprise; identify triggers for decision-making; selection of optimal cuts segmentation and profiling; Construction and description of predictive models using Data Mining techniques.
Development of management principles – the layout of the audit, analysis and modeling results in a system: description of the spectrum of reports and predictive models with a list of required data and their update regulations; offer rules and principles of decision-making, including the rules for monitoring their efficiency criteria.
Offer initiatives – on the basis of the revealed regularities offer commercial or other initiatives aimed at achieving the objectives; support proposals with calculations of economic impact and description of the calculation methodology.
Adaptation – the active transfer of knowledge, the building of procedures and other administrative documents in the company’s adopted form, the training of personnel.
— Scoring systems. All types of scoring – application, behavioral, collection and fraud. By analyzing information on previously issued loans and on their return, the bank is able to build a model, based on which, it is possible to evaluate the degree of reliability of clients with high accuracy.
— Detection of credit and card fraud. By analyzing past transactions that subsequently proved fraudulent, the bank identifies stereotypes of such fraud.
— Customer segmentation. Breaking customers into different categories, the banks provide their marketing policy more focused and efficiently, offering a variety of services to needed group of customers.
— Analysis of the records of the client’s calls and services usage. Purpose of this analysis is to define categories of customers with similar stereotypes of services usage and to develop attractive sets of services and tariffs for each category of clients.
— Identification of customer loyalty. By analyzing client’s behavior we determine the potential risk, e.g. the customers who may leave. Knowledge of this group of customers allows us to provide a strategy of targeted marketing for the customer’s retention.
— Optimization of product mix. Formation of the optimal range, in order to profit maximization, support desired profit over a long period of time is very important for enterprises that want to be competitive.
— Loyalty programs. Using the methods of data analysis we can define consumption patterns of customers and identify target groups. That gives the opportunity to build an effective loyalty program taking into account the characteristics of each group of customers and offer them the most interesting products.
— Demand stimulating. Market basket analysis allows finding the best way to offer a combination of products that cause the greatest interest, in order to organize cross-sales and discount schemes.
Clinical and social research:
Statistical support in clinical and social research with a detailed commentary. We provide assistance in the formulation of goals and objectives, the choice of statistical methods, and conclusions. The results of the analysis are provided in charts, graphs, calculation criteria, advice on any question.
Raiffeisen Bank Aval
— Behavioral Car Loans scoring model. The GINI index is 85% avarage on training, testing and out of time testing sets.
— Application Cash Loans scoring model with getting into account data from client's mobile account. The GINI index is 50% average on training and testing sets.
— Customer segmentation for marketing campaigns based on probability of credit card's activation and usage. The Lift coefficient is 20%.
— Application scoring, Reject Inference modeling with using Fuzzy Augmentation method. The GINI index is 39% on training and testing sets.
— Classification of client's database on probability of payment the debt. Payments flow forecasting. Other predictive modeling.
— High load classification and prediction system for online advertising auction. Prediction CTR, ITR. Pricing model based on click probability.
— Automated bidding, pacing and boosting algorithms.
— Game recommendation system, deposit recommendation system.
— Churn, LTV prediction.