After these steps, the text is represented by a finite
list of words, where all words with the same root are
seen as one term and the stop words are removed. The
list is ordered according to the appearance of the word
with a weightage relative to the number of all valued
words (words without stop words). The outcome of this
algorithm is a ranked list of valued terms, which
represents the written idea. In a set of combinations,
these valued terms will then be used to query several
data sources (see Figure 2, S3).