Describes a disproportionate weight, e.g. in the training of AI, in favour of or against information contained in the data. Conversely, this can result in disadvantages or perpetuate unfair biases, which is particularly critical when the data is used as the basis for making decisions with (in)direct impact on daily life.