Why does the "Element Text Should Be" ML assertion step sometimes succeed even if the value of the execution text is not correct?

In the “Element Text Should Be” steps of ML assertion, if the difference between the expected text and the actual text is slight, as in the following example, Autify sometimes may not be able to determine the text difference and the test succeeds.

• Expected text: hogehoge 123456
• Actual text: hogehoge 123457

If the target text is short or the difference between the expected and actual text is large, Autify will correctly determine the difference, and the test will fail.

• Target text is short

  • Expected text: hogehoge 1
  • Actual text: hogehoge 2

• The difference between the expected and the actual text is large

  • Expected text: hogehoge 12345
  • Actual text: hogehoge 67890

If the difference between the expected and the actual text value is so small, and the ML assertion is successful, please use the assertion instead of the ML assertion.

Related articles

• Machine Learning(ML) Assertion