I was reading an articel the other about how GenAI could improve the world of data analytics. Although the ability to generate the models and the "How To". I'm not sure if this is something that can really help us with the interpretation of the data results. Am I wrong here?
I think we should start with a few known models, tests and outcomes. Give it the same data we had back then and ask it to re-create the work, and re-envision the results and let us know it's conclusions ... and then judge what we have. No use trying to second guess what it could do until we establish a baseline of just how accurate can it be in the first place.
No I think you're spot on. I suspect GenAI will definitely help us create better models and target our audience data pool in a faster and easier way. Ut I'm also leery of how it would accurately interpret the results. I'm open to other views too though.
I'm not sure about this. The steps to modelling can be very different for specific geographical regions, cultures and languages. Will the AI apply proven models and techniques known as valid for the American market to data samples taken from the EU and Eastern European countries? If nos do we trust the results at all?
I'm in favor of anything that makes my life easier! Better to correct the model and reinterpret results than to start from scratch on both sides and put months into building a new model and project than a few weeks to correct one.
