In predictive analytics and machine learning, the concept drift means that the statistical properties of the target variable, which the model is trying to predict, change over time in unforeseen ways. This causes problems because the predictions become less accurate as time passes.
The term concept refers to the quantity to be predicted. More generally, it can also refer to other phenomena of interest besides the target concept, such as an input, but, in the context of concept drift, the term commonly refers to the target variable.
Read more about Concept Drift: Examples, Possible Remedies, Software, Projects, Meetings, Mailing List, Bibliographic References
Famous quotes containing the words concept and/or drift:
“Modern man, if he dared to be articulate about his concept of heaven, would describe a vision which would look like the biggest department store in the world, showing new things and gadgets, and himself having plenty of money with which to buy them. He would wander around open-mouthed in this heaven of gadgets and commodities, provided only that there were ever more and newer things to buy, and perhaps that his neighbors were just a little less privileged than he.”
—Erich Fromm (1900–1980)
“To drift with every passion till my soul
Is a stringed lute on which all winds can play,
Is it for this that I have given away
Mine ancient wisdom, and austere control?
Methinks my life is a twice-written scroll
Scrawled over on some boyish holiday”
—Oscar Wilde (1854–1900)