Welcome to the documentation of the C3 Artificial Intelligence Services

C3 Artificial Intelligence services form the upper level of core services in the NEANIAS ecosystem and provide facilities that may reach up to the end user offering features of a typical Machine Learning (ML) workflow lifecycle, and that are also intended for composition of higher level services.

Providers and contributors

The C3 services are provided by:

  • SZTAKI
  • INAF
  • ALTEC
  • UNIMIB

C3 Services

A web-based development environment for scientists to design, develop and evaluate machine learning solutions. After development, training of large models on big data can be done in a distributed environment, refer to C3.3 and C3.4.

Serving trained machine learning models as a web service.

The training of deep neural networks on big data takes significant time, even on GPU-enabled workstations. To increase efficiency, a distributed computation cluster should be used, where users can define models to train and collect results.

C3.4 supports a variety of machine learning models such as Support Vector Machines, Decision Trees and ensemble models (Random Forests and Gradient Boosting) for both classification and regression, clustering algorithms, and other support functions that are useful in real-world, large scale, machine learning pipelines. The set of basic ML algorithms and models can also be extended by means of creative parallel computing approaches (e.g. parallel implementations of DBSCAN algorithm on top of Spark, an algorithm which is not natively provided by MLib, are known and available).