hello, everyone. I’m a novice to KNIME extension development. I wonder if there is any way that allows users to answer some questions then automatically generate nodes on the interactive panel. Many users don’t have any data engineering background, they don’t know how to build a workflow from scratch especially when they first use KNIME they have to pick up nodes from a huge pool, so my purpose is to design a guidance application that can be integrated to KNIME platform to help the users generate the outline nodes, so they can simply adjust some arguments on this generated nodes and continue their data analysis work.
Or if there exist some extensions that do the similar job I mentioned, could you list them in the below comments? I have no idea how to develop this extension for KNIME. I only know you can develop a new node for KNIME.
Thank you for your time and comments!
Hi @Yanfgeng, (welcome to the forum!).
Is this not to some extent what the “Workflow Coach” panel is attempting to achieve?
With that open, when you select a Node on your workflow, it gives suggestions about what the next Node might be based on things that other members of the community have done. I certainly found that quite useful in providing initial guidance, and it flags up Nodes that you might not otherwise consider.
I guess you are looking at this more from the point of view of producing some kind of “template workflow” based on the type of thing that the user is trying to achieve, rather than guidance on the “next node”… as they go along?
Hi @Yanfeng
yes this is absolutely possible! We started this some time ago and published it under same name of “Integrated Deployment”
Using this a KNIME Workflow can build a KNIME Workflow. The workflow can be controlled via the KNIME Webportal, to let the user make decisions and configurations. In the end you can reuse the workflow.
You can here find a starting page linking you more material: KNIME Integrated Deployment | KNIME
And a just released additional blogpost: Monitoring the Production Workflow: Model Monitoring in a Data Science Context | KNIME
Cheers, Iris
Thank you, takbb. Your reply really helps me a lot. I didn’t notice that KNIME has this useful feature before. Combined with KNIME HUB templates use, KNIME has provided pretty much guidance for novices.
Thank your for your reply. Iris. I’ve explored the integrated deployment feature this morning, it seems like we need to build and deploy the complete workflow first, then let the user make decisions about which path should be executed. But what I mean is the guidance should be in the building workflow stage, instaed of someone has built a complete and nice workflow that ready for use.
Anyway, I’m really glad that this community is so friendly and helpful. I was amazed that someone could response me even during the weekends. Thanks for you guys time again.
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.