In many domains of enterprise and scientific data analysis the lack of structural knowledge encourages the use of machine intelligence methods, which surpass the structural flexibility of classical statistical methods. Since this approach, however, generally is paid by high mathematical uncertainty, it is indispensable to properly select and adapt the method to it’s respective application. This requires a close cooperation between enterprise analytics and data science.

Nemoa Deep Belief Model

Nemoa is a data analysis framework for collaborative data science and enterprise application. The key goal of the project is to provide a long-term data analysis framework, which seemingly integrates into existing enterprise data environments and thereby supports collaborative data science. To achieve this goal nemoa orchestrates established frameworks like TensorFlow® and SQLAlchemy and dynamically extends their capabilities by community driven algorithms for probabilistic graphical modeling [PGM], machine learning [ML] and structured data-analysis [SDA].


Nemoa is open source and based on the Python programming language. It provides:

  • A transparent DW architecture for the seamless integration of existing SQL databases, flat data from laboratory measurement devices or data generators.
  • A versatile and fast data modeling and data analysis framework.