• preprocessing,
  • classification
  • and automatic
  • summarization
  • of emails

Melusine is a high-level Python library for emails classification and feature extraction, written in Python and capable of running on top of Scikit-Learn, Keras or Tensorflow. Melusine is designed for the preprocessing, classification and automatic summarization of emails written in french.


  • Email preprocessing Toolbox
  • Custom email preprocessing Pipeline
  • NLP Toolbox
  • Phraser and Tokenizer Pipeline
  • Embeddings training
  • Metadata Pipeline
  • Summarization by Keywords extraction
  • JSON Config file valuable to preprocessing that can be customized
  • Custom name file to discard or flag according to name relevance
  • Classification Toolbox
  • Convolutional networks and Recurrent networks
  • Runs seamlessly on CPU and GPU
  • Semantic detector for sentiment score prediction and topic search

easy preprocessing

Provide a TransformerScheduler class to build your own transformer and integrate it into a scikit-learn Pipeline


high adaptability

Freedom to train his own classification and automatic summarization model according to the constraints your problem