Asent

Asent is a rule-based sentiment model, building on spaCy, allowing the user to take advantage of rules based on part-of-speech tags, names and much more. The goal with asent is to have a fast and transparent sentiment pipeline, which is easily customizable.

Contents

The documentation consists of three parts:

• Getting started contains the installation instructions, guides, and tutorials on how to use asent.

• Lexicon resources contains the documentation on the resources used for each specific language.

• API references contains the documentation of each function and public class.

Getting started

• Installation
• Getting started
• Customizing Asent
• Analysis of a speech
• Languages

News

• News and Changelog

FAQ

• Frequently asked questions
  • Citing asent
  • How do I test the code and run the test suite?
  • How is the documentation generated?
  • How do I add a new language?
  • Couldn’t you just train a transformer for this?

API references

• Visualizers
  • asent.visualize
    • visualize()
    • visualize_sentence_prediction()
• Getters
  • asent.getters
    • allcap_differential_getter()
    • make_doc_polarity_getter()
    • make_intensifier_getter()
    • make_is_contrastive_conj_getter()
    • make_is_negated_getter()
    • make_is_negation_getter()
    • make_span_polarity_getter()
    • make_token_polarity_getter()
    • make_txt_getter()
    • make_valance_getter()
• Components
  • asent.component
    • Asent

GitHub

• GitHub Repository

Indices and search

• Index

• Search Page