This is a simple sanic applications that use 3 major NLP python libraries:

• NLTK
• MITIE
• Freeling

To install freeling please use the official documentation at github, it could be trickie

To verify installation execute

python tests.py

Requires python 3.5

To run server

python app.py

We expose for major services:

• /summarize: Naive frequency summarization
• /ner: Name Entity Recognition
• /syntax: Syntax tree
• /summarize-posttagger: A more clever frequency summarization

All methods require a x-www-form-urlencoded param called article, and the responses should return an structure similiar to:

{
  "status": <text>, # Verbose description of transaction status
  "article": <text>, # Text to be summarized
  "summary": <text>, # Text summarized
  "frequency": [
    {
      "word": <text>, # Non stop word
      "frequency": <float> # Relative frequency
    }
  ],
  "topics": [
    <text> # Topic 
  ]
}