Reporting bias — искажение, возникающее, когда частота информации в доступных records не отражает ее реальную частоту. Поэтому модели, обученные на reported data, могут усваивать не то, что реально происходит, а то, что люди склонны документировать или обсуждать.
The fact that the frequency with which people write about actions, outcomes, or properties is not a reflection of their real-world frequencies or the degree to which a property is characteristic of a class of individuals. Reporting bias can influence the composition of data that machine learning systems learn from. For example, in books, the word laughed is more prevalent than breathed. A machine learning model that estimates the relative frequency of laughing and breathing from a book corpus would probably determine that laughing is more common than breathing. See Fairness: Types of bias in Machine Learning Crash Course for more information.