I am trying to extract some movie entity from sentence but it failed to extract properly (in RASA X)for example: if i am giving this as movie name WA BISAT ELRIH (ARB) then its fetching " WA BISAT ELRIH (ARB" like this unable to fetch “)” this even i tried to train multiple sample like this but still no luck and if i am trying same thing in (rasa interactive)
Sorry, for the late reply. Just to clarify, it is not working on
rasa shell as well as in Rasa X? What Rasa version are you using and how does your pipeline look like?
even if i am passing those movies in button format its unable to take whole thing as movie its taking only ARB with wrong intent value
Can you give an example? How does your button look like?
I just checked the code. During preprocessing of the text any punctuation characters are removed. So,
WA BISAT ELRIH (ARB) becomes
WA BISAT ELRIH ARB . Let’s say the entity
WA BISAT ELRIH (ARB) starts at position 5 in the text and ends at position 25. Due to the fact that we removed the closing bracket
) during preprocessing the end position of the entity does not match anymore/is not present anymore. Thus, the entity is misaligned. To avoid such problems, you should avoid punctuation characters in your entities, especially at the end of your entities. If an entity is misaligned, it will be ignored, so you will have one training example less. Other than that training/the bot works as usual.
Can you suggest me how to deal with this with an example , as i want to extract the same with brackets and all
Unfortunately, you will not be able to include the closing bracket. For now, you should remove the brackets from your examples, if you want them to be processed. I cannot think of an easy fix for it. Feel free to open a GitHub issue and we will give it more thoughts and hopefully find a solution for this. Thanks.
Thanks for the update