The search analysis library contains multiple tools to help you analyze ranking and explain ranking differences from search systems.


Keywords
search_analysis
License
MIT
Install
pip install search-analysis==0.1.23

Documentation

Search Analysis

Documentation Status

The search analysis library contains multiple tools to help you analyze your search engine approach.

  • Free software: MIT license
  • Documentation: Coming soon

Features

Use-Case: Analyze Single Approach

  • Get Metrics: Precision, Recall, F-Score
  • Find queries that perform especially bad/good
  • Analyze false positives, false negatives and true positives
  • For every query get score, field value, highlighting

Use-Case Compare Two Approaches

  • Visualization of metrics side by side
  • Find queries with biggest difference
  • Analyze false positives, false negatives, true positives and calculate disjoint sets
  • Get scores visualized side by side

(For now it's only possible to work with Elasticsearch)

Help

For questions you can contact us via E-Mail or through our website (https://www.pragmalingu.de/).

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.