Making use of our discoveries
What should we do? We have a k-NN telling us that single women are not buying additional items and we have Naïve Bayes being no help at all. We could do some more classification models, but let's assume we feel good enough about our analysis and want to go to production with this model. How should we do that? A key issue to consider is that the model is based on some static data in one of our database tables that is not updated via the normal transactions of the company. This means we really don't need to retrain the model frequently. Another problem we have is that we need to figure out the gender and marital status of the people ordering our bikes. Perhaps we are asking the wrong question. Instead of asking how ...
Get Mastering .NET Machine Learning now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
