{"id":23259,"date":"2022-02-22T16:19:45","date_gmt":"2022-02-22T21:19:45","guid":{"rendered":"https:\/\/www.3pillarglobal.com\/?p=23259"},"modified":"2025-05-14T17:28:41","modified_gmt":"2025-05-14T17:28:41","slug":"data-models-gone-bad-can-spoil-your-predictions","status":"publish","type":"blog","link":"https:\/\/www.3pillarglobal.com\/insights\/blog\/data-models-gone-bad-can-spoil-your-predictions\/","title":{"rendered":"Data Models Gone Bad Can Spoil Your Predictions"},"content":{"rendered":"\n

Artificial Intelligence and Machine Learning Require Ongoing Tuning<\/h2>\nWe recently discussed how important it is to choose the correct set of data<\/a> from which to drive predictions and inferences. It\u2019s also important that the data not be skewed in any particular direction\u2014 or else it will generate skewed predictions. In this post, we discuss the number-crunching processes downstream from data acquisition.\n

\nA data model represents what was learned by a machine learning algorithm. As data flows downstream through your various business processes, the data models supporting your algorithms that direct your systems or people can fall out of alignment. This occurs when the premises and assumptions surrounding the data models change enough to erode the accuracy of the analytics. Or\u2026the data models take the algorithms out of meta-stable equilibrium and into turbulent predictive behavior.\n

\nIt\u2019s not a good thing when data models have gone bad, spoiling your predictions and inferences!\n

\n

Artificial Intelligence and Machine Learning: Not Set and Forget Tools<\/h2>\nUnderstanding why this happens requires realizing that machine learning algorithms and artificial intelligence are not \u201cset and forget\u201d tools. Like any other software system, they require tuning to maintain optimal performance. Also, it is important to understand the boundaries of the data values going into the algorithms so you aren\u2019t pushing them out of their \u201cflight envelope.\u201d\n

\nTuning includes the normal maintenance engineering tasks you would apply to any software deployment\u2014upgrades, patches, and adaptations\u2014in reaction to infrastructure changes. As you maintain, be sure to validate the algorithms are still \u201cin tune\u201d and that data feeds adhere to the original assumptions and conditions.\n

\nThe process is similar to what Formula One race car drivers go through. They know the techniques and wisdom (data model) gathered about the nuances of one track do not apply to another track. They also contend with races that require a change to the car type\u2014Indy, NASCAR, or an off-road vehicle; all these things impact their driving technique. So as premises and assumptions change from race-to-race and from car-to-car, drivers adapt their internal algorithms that direct them on how to best navigate each track.\n

\n

Factors Causing Data Models to Spin-out<\/h2>\nHere\u2019s a rundown of various factors that can cause data models to derail:\n

\n