Text analysis with Stanza - Stanford NLP Python Package
Вставка
- Опубліковано 21 бер 2023
- Stanza is a very powerful, highly accurate, and efficient Python package for text analysis. In this video, I explain how you can use it to tokenize texts, lemmatize words, recognize names entities (NER), analyze the sentiment of a sentence, and represent the dependency and constituency structure of the sentences.
-- Documentation --
stanfordnlp.github.io/stanza/
-- Support Pythonology --
www.buymeacoffee.com/pythonology - Наука та технологія
This is one of the most helpful videos I've ever seen! Thank you so much
This just thought me more in less than 20 minutes than three weeks of computer linguistics at uni. Thank you!
Very helpful , thank you. I found your video while I was searching for more/better information about the NER annotation (format of .tsv file) than Stanford’s documentation and still seeking as of now 😅
Ahh. Finally another video. After so long. Interesting video
Hi Irfan, thanks for the comment, so good to see the videos are useful. Yea I've been busy but I'll try to post now videos. :)
I came back to this video because i knew it existed.
I have switched to a full time job now where im serving as an analyst for customer operations and my first taks is to create a customer-agent chat sentiment analyser model.
Came back to see the sentiment pary especially.
Its partially useful for me now and just wanted to understand if it can analyse the whole chat entirely and not individual sentences.
That way i can just gather the chat data and apply the logic and can easily classify chats based on irate and satisfied customers.
Also does it have the capacity to identify and rate nuances like sarcasm.
Would wait for you to comment.
Thanks a tonn
Hi again, Irfan. Sarcasm is very difficult to detect, so i would not count on it. For the rest, the easiest approach is to use a library like Vader or TextBlob for sentiment analysis. I have some tutorials but you can also check this article on datacamp:
www.datacamp.com/tutorial/text-analytics-beginners-nltk
@@Pythonology thank you very much good sir
how to detect the subject, the predicate, and the object of a sentence? I think the dependency and constituency structure may be useful, but i have no idea.
Im using spello to auto correct the spelling mistakes while entering data.
hello am working on project that use stanza to do depandancy parsing but in Arabic language ,the results of parsing a stentence somtimes changes and gives me false results how can i deal with this please
Really interesting video. I wonder whether can we also analyze (count certain words) in pdf file (not english language) using this package?
Thank you
It covers many languages. So it should be possible