Estimating the Difficulty of a Foreign Document

This work shows to calculate the difficulty of a foreign document with respect to one's native language. For example, a portuguese document is easier to read by a spanish speaker than by an english speaker. Our model considers such differences by examining both:

• How easy are the words in the document.
  We estimate this recording how many times a word appears on the internet. For this we leverage the power of search engines.
• The presence of cognate words with respect to the reader's native language.
  As an example, the word 'Haus' in German is a cognate for the word 'house' in English and can be easily understood by an english speaking person.