{"id":41765,"date":"2017-11-06T00:00:00","date_gmt":"2017-11-06T00:00:00","guid":{"rendered":"https:\/\/www.techopedia.com\/experts\/why-is-machine-bias-a-problem-in-machine-learning\/"},"modified":"2022-05-05T02:20:23","modified_gmt":"2022-05-05T02:20:23","slug":"why-is-machine-bias-a-problem-in-machine-learning","status":"publish","type":"q-a","link":"https:\/\/www.techopedia.com\/why-is-machine-bias-a-problem-in-machine-learning\/7\/33002","title":{"rendered":"Why is machine bias a problem in machine learning?"},"content":{"rendered":"
This question can be answered in two different ways. First, why is machine bias problem, as in, why does it exist in machine learning<\/a> processes?<\/p>\n Machine learning, though sophisticated and complex, is to an extent limited based on the data sets that it uses. The construction of the data sets involves inherent bias. Just like in the media, where omissions and deliberate choices of inclusion may show a particular bias, in machine learning, the data sets that are used must be examined to determine what kind of bias exists.<\/p>\n <\/p>\n