Hello Rasa community !
I need your help please ! i’ve developped a chatbot based on rasa and now i’m looking to add sentiment analysis , i 'm using frensh as a language of the chatbot , my idea is just detecting from my nlu data positive and negative intention and have a report of sentiment analysis ! I’m blocked help me pleaseeeeeeeeeeeeee!
Hi @m2cci-sba! It sounds like you will want to create a custom NLU component. There is a blog post about how to do this that actually covers a sentiment analysis component example here.
actually i didn’t get how to use these codes in my chatbot. can you please tell me how to use these codes.i m new to rasa.plz help me if u got my problem.
Hi @soumya-SR! Did you read through the blog post? If so, what did you not understand? If you are new to Rasa, I reccomend watching the Rasa Masterclass
i have seen rasa matserclass . i just wanted to ask how can we implement these code.actually when i added sentiment analysis code in my chatbot.bot threw an error due to pipeline i added to resolve this i removed supervised embedding pipeline but after that code bot was trained smoothly but replies were very different.it always answered default action text. how can i solve this issue.
one thing i also want to ask what is the expected output of the bot after added sentiment analysis. sorry to ask you such type of silly question 
and also thanks for your reply
Hi @soumya-SR. There is an example in the blog post about what the pipeline should look like and what the output will be after you added the component
not OP, but thx for the link <3
Hello Rasa community!
Integrating sentiment analysis into your Rasa-based chatbot for French involves several key steps. First, ensure your NLU training data includes diverse examples of positive and negative intents expressed in French, specifically tailored for sentiment analysis reviews. Augment this data with varying sentiment expressions to enhance model comprehension. Next, choose a sentiment analysis model compatible with French, such as those available through libraries like Flair, NLTK, or spaCy. Configure your Rasa pipeline to incorporate this sentiment analysis component after the NLU pipeline to analyze user inputs effectively. Implement a mechanism to generate sentiment analysis reports based on user interactions, storing sentiment scores alongside user messages and intents for comprehensive insights into sentiment analysis reviews. Finally, rigorously test and iterate on your sentiment analysis implementation to ensure accuracy and reliability. By following these steps, you can enrich your chatbot’s capabilities, offering users a more nuanced and responsive interaction experience in French.