Defining pattern rules for spaCy Matcher

[@polarbear986]

Very good lesson. I like that you explain clearly and slowly.

[@AppliedLanguageTechnology]

Glad you liked it!

[@walid2635]

Nice Video

[@aricircle8040]

The lessons are so great, thanks for easy, good and fast explanations and information that are impossible to find anywhere else!!! It would be also great to know about other methods of Token class more, as cluster id or similarity, I was wondering whether it is possible to use similarity for creating an adjective classification model for finding errors in the adjective order, e.g. “red beautiful wooden Chinese old spoon”

[@AppliedLanguageTechnology]

Hey! Thanks for your feedback! I'm planning to do more extensive videos on spaCy Token/Span/Doc objects and their properties, but I won't have time until next autumn. I will cover the similarity() method in the tutorials that I will publish this week; not sure if they would work for your use case though!

[@shefalishrivastava1189]

God bless these people.

[@AppliedLanguageTechnology]

I take that as a compliment. :-)

[@ildaraskarov8755]

Great video, thank you. I have a question, I am looking for phrases of the specific patterns, is there a way to determine whether founded phrases belong to subject or object in the sentence? Thank you.

[@AppliedLanguageTechnology]

Hi Ildar! Yes, but you will have to use the DependencyMatcher - I have a video here: kzbin.info/www/bejne/nYbFfnxrZ8d8g7s

[@pipyaku8276]

Great video, thanks a lot! Would be possible to share '.ipynb' file?

[@AppliedLanguageTechnology]

Hey! You can find all notebooks here: github.com/Applied-Language-Technology/notebooks

[@Masoumehf]

Hi what about if we want to find a specific word in the document and return the whole paragraph including it?

[@AppliedLanguageTechnology]

Hi Sarnia! Just define a pattern rule that matches the word (or its lemma, if you also wish to retrieve all conjugated forms). Any hits that are returned consist of spaCy Span objects, and you can use the "doc" attribute of the Span object to retrieve the Doc that contains the Span. Paragraphs are a bit trickier - maybe you can look for sequences of two line changes (e.g. ) or any sequence of characters that are used to mark paragraph changes in your document. Hope this helps! :-)