WhatsApp +7(499) 113-2062 info@adg.ru Okhotnyi Ryad str.2 Moscow, Russia
Folow us on social

Data Mining Services

We can help you connect data of all your business processes with critical external sources (public health status, utility data, financial indexes, economic indicators, government policy, climate and geo indicators etc.) to forecast your future results. And the question always arises - how to apply them in each specific case?

c
Data collection
Data collection

Logs files, CSV/XML/text/Excel files, databases, web services, business applications, etc.

Data cleaning
Data cleaning

Sorting, preprocessing, linking, organizing and consolidating data using preparation methods

Modeling
Modeling

Selection of factors, machine learning, forecasting, complex calculations

Visialization
Visualization

Visual interpretation of results from multivariate analysis via charts and graphs dashboards

Data Collection

Data collection

After formulating the problem, the company’s specialists begin a preliminary study of the data necessary to solve the problem.

On the part of the customer, participation may be required to clarify, for example, the meaning of the data being investigated or dive into the specifics.

Tools & methods

Languages: Python, R, shell, PHP, SQL
Methods: API, ODBC
Tools: Power Pivot, Excel, MSSQL, MySQL

Datasets

Customer Databases or File Archives

Public Datasets or Open Data APIs

Data cleaning

At this stage, the company’s specialists prepare data for further analysis. For this, the whole range of data preparation methods is used; in each case, specialists choose the most suitable methods.

Tools & methods

Languages: Python, R, shell, PHP, SQL
Methods: API, ODBC
Tools: Power Pivot, Excel, MSSQL, MySQL

Datasets

Customer Databases or File Archives

Public Datasets or Open Data APIs

f
c

Modeling

The main stage is data analysis. This is a completely technical process that the company’s specialists carry out both using their own algorithms and using software from world leaders in analysis and hypothesis testing.

Tools

Languages: Python, R, shell, PHP, SQL
Software: Statistica TIBCO, IBM SPSS/Watson

Datasets

Customer Databases or File Archives

The list of methods used for the analysis and search for Hidden Knowledge

Neural networks;

Grouping and exploratory analysis;

Frequency tables and contingency tables;

Analysis of Multiple Response;

Nonparametric statistics;

Methods of Power Analysis;

General linear models (GLM);

General Regression Models (GRM);

Factor analysis;

Generalized linear models (GLZ);

General Partial Least Squares (PLS) models;

Methods of variance and mixed ANOVA / ANCOVA models;

Survival analysis;

General nonlinear assessment;

Time Series Analysis / Forecasting;

Structural Equation Modeling (SEPATH);

Methods of cluster analysis;

Principal component analysis and classification;

Canonical correlation analysis;

Reliability and positional analysis;

Analysis of correspondences;

Discriminant analysis;

General Models of Discriminant Analysis (GDA);

Etc.

Visualization

The company’s specialists are engaged in the interpretation and visualization of the acquired knowledge and patterns. The knowledge found is presented in a convenient and understandable form using business intelligence tools.

Tools & methods

Tools: Power BI, Tableau, Google Data Studio, IBM SPSS, Power Pivot, Excel, etc.

Datasets

Customer Databases or File Archives

Public Datasets or Open Data APIs

v