Such as in all machine learning tasks, in the field of automatic text simplification, also a data collection (also called corpus) is required to evaluate and/or train automated text simplification systems. A corpus for text simplification is a monolingual data collection that always consists of a collection of pairs, where one side of the pair consists of a complex text in one language and the other side consists of at least one simplified version of these texts (called simplified texts) in the same language. Ideally, both sides of a simplification pair contain the same content but differ in their complexity level.