Define AI - Artificial Intelligence
Definition of AI - an overblown, overused acronym and the latest large, shiny object that gets invoked so often it’s lost clarity.
Let’s better define AI - G2 defines AI as “computer systems performing tasks with similar, equal, or superior intelligence to that of a human (e.g. decision-making, object classification and detection, speech recognition and translation)”.
This definition introduces one of the MAJOR complexities of AI in that it requires this comparison to human capabilities that really isn’t all that interesting or valuable. There are some things that computers so SO MUCH BETTER than humans such as
storing and retrieving lots of information,
basic and very complex math,
repetitive tasks.
Computers don’t get bored or make more mistakes because they’re tired.
More helpful in my opinion than getting worked up about whether you’re deploying AI is starting with some fundamental questions -
What would you like to learn from your data? What questions would yield the greatest return to your business or project? In eCommerce (where we do a lot of work), these key questions often seem to involve which products to put into stock because we know they’ll sell, where to place these products and then how to price them to maximize profits.
Do you have a good clean data set that you can use to do some analytics that might ultimately lead to some predictive analytics for your business? We often see that companies don’t have clean, well organized data and the first step is to create a data lake in the cloud.
Are there outside data sets that you could merge with your internal data to enrich the data to look for additional insights? For example, we’ve found that getting insights on which or your SKUs are selling on Amazon is a huge guide to what may sell on your own site.
So get started small with one question you’d like to understand, test the cleanliness of your data and expect things to go poorly. But getting started learning and experimenting is critical to eventually making it all the way to actual AI.
Terms are used quite loosely in the AI/ML/Analytics space - which is kind of ironic since this is a field that prefers, even requires precision. Definitions matter and a great website to learn some key definitions in the field is https://www.g2.com/articles/artificial-intelligence-terms.
Let me know how I can be helpful at d at nimblegravity.com
