Skip to content
This repository was archived by the owner on Nov 2, 2025. It is now read-only.

Test files for the browser

Chris Brogly edited this page Mar 8, 2019 · 4 revisions

Unit Tests

These are currently unavailable. Some exist for the clickbait detector machine learning features but these fell behind early in development due to limited resources. The best way to test the program right now is with manual testing.

Manual Steps

The easiest way to try all of the manual test documents for each of the three detectors is to navigate to "...\litrl_code\manual_browser_test_files" in Litrl Browser where you can browse all of these test files in a file explorer page.

Clickbait detector

A manual test file for the clickbait detector can be viewed in the browser by navigating to ... "litrl_code\manual_browser_test_files\clickbait_tests.html" - this contains sample headlines for clickbait and non-clickbait texts. There is also another file called "clickbait_tests_for_improvements.html" which contains misclassified texts. This file can be used to test if changes to the clickbait detector improve real-world accuracy. Misclassifications tend to happen more frequently with non-clickbait headlines as the detector is not perfect.

Satire detector

Manual test files for the satire detector can be viewed in the browser by navigating to ... "litrl_code\manual_browser_test_files\satire_tests" - this contains sample headlines for satire and non-satire texts. Note that the detector does misclassify some of these as it is not perfect. There are a selection of these texts labelled "false_positive" in the filename.

Falsification detector

Manual test files for the falsifications detector can be viewed in the browser by navigating to ... "litrl_code\manual_browser_test_files\falsification_tests" - this contains sample headlines for falsified and legitimate texts. Note that the detector does misclassify some of these as it is not perfect. There are a selection of these texts labelled "false_positive" in the filename.

Clone this wiki locally