What kinds of problems is machine learning and AI best suited to solve? To help define your analytics approach, use this 2×2 matrix to identify the kind of problem you’ve got, from a data perspective, and then select approaches based on that problem type.
Machine learning solves four different kinds of problems:
– Supervised continuous – we know what outcome we’re after, and the data is numeric
– Unsupervised continuous – we don’t know what outcome we’re after, and the data is numeric
– Supervised categorical – we know what outcome we’re after, and the data is non-numeric
– Unsupervised categorical – we don’t know what outcome we’re after, and the data is non-numeric
What kinds of problems might we solve in marketing with these?
– Supervised continuous – what channels drive conversions on our website?
– Unsupervised continuous – which SEO data points should we pay attention to?
– Supervised categorical – what’s the sentiment of this pile of Instagram posts?
– Unsupervised categorical – what are the most common themes in this collection of articles?
Copyright and Permissions
Our ideas are available under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International license. This means you are free to use these ideas for non-commercial purposes (such as journalistic endeavors, open-source academic research, etc.) where no transaction or financial benefit is involved; you must attribute your usage appropriately. Under the terms of the license, you may not make derivative works without permission.
This work by Trust Insights, Inc. is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Based on a work at https://www.trustinsights.ai/.
Permissions beyond the scope of this license may be available at https://www.trustinsights.ai/.