hello …
could you pls help me with the creation of the flow for a chat bot that will be connected to a csv dataset so I can ask questions regarding the data … e.g. give me the total number of posts per network (when the network will be an existing column in the dataset)
many many thanks in advance,
Valentina
Could you provide more detail about your data structure and the results you’re trying to produce? If they’re fairly fixed it seems like a chatbot is overkill. I know its the current rage, but possibly a “normal” workflow would be sufficient.
I have an Excel file with some analytics data per post and per date. I need to create a bot with which I can ask some questions related to the datasets … eg. how many posts we have for Jun 2023 … I know that this is a simple question but this is just an example … If I manage to create a bot that can be connected to the specific dataset and can read and analyze the data then I will be able also to ask more specific questions …
ty for the help
@serefidou
And this needs to be done in KNIME? Who is using this bot? Is it deployed somewhere?
br
yes the bot will be created in knime … so everyone with the link of the bot can make questions to it that will be related to the dataset … 
So it is deployed via KNIME hub?
Then you need a vectorstore of your document(s) and an LLM (OpenAI or OpenSource) to answer your prompts
Although this example is not directly usable for your case, it will give you an idea about what you’re facing.
I would still advocate a simpler starting approach like a column selector to filter your data set and then the Statistics node to give you some basic statistics.
OK thank you for the help
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.