Skip to main content
10 events
when toggle format what by license comment
May 17, 2023 at 17:25 vote accept Z Bokaee
May 17, 2023 at 17:25 comment added Z Bokaee Honestly, no one could answer my question in such a smooth and clear way. I get the point to a great extend. Thanks a million!
May 14, 2023 at 14:14 comment added Luca Anzalone Thanks for the extra information. I have updated my answer, have a look.
May 14, 2023 at 14:13 history edited Luca Anzalone CC BY-SA 4.0
added 1232 characters in body
May 14, 2023 at 14:07 history edited Luca Anzalone CC BY-SA 4.0
added 1232 characters in body
May 14, 2023 at 7:15 comment added Z Bokaee I appreciate your time. I've updated my question with relevant information. I would be thankful for any help.
May 12, 2023 at 17:59 comment added Luca Anzalone If you're able to compute F1 and F2 then you just need to setup a classifier where each sentiment is a class, there is no need to train a model that predicts them. Can you update your question with the papers you mentioned? Knowing them may help, others too
May 12, 2023 at 14:28 comment added Z Bokaee The problem is that I read lots of articles on classification problem like sentiment analysis and sarcasm detection and in most of them authors think about different features and try to teach their models using those features. However, I have no idea how I should predict the classification labels for new sentences. I would appreciate it if you can help me.
May 12, 2023 at 14:18 comment added Z Bokaee thank you for your response. So you believe It’s better to use n-gram or vocabulary lookup, that’s a good idea. But what if I want to use those features only? Let’s say I add a function to my program to calculate F1 and F2 for any new input. How should I tell the model that the output of the function is equal to F1 and F2. ?
May 10, 2023 at 20:30 history answered Luca Anzalone CC BY-SA 4.0