Advanced Document Description, a Sequential Approach.

The article presents a dissertation that addresses the problems of the extraction, selection and exploitation of word sequences, with a particular focus on the applicability to document collections of any type and written in any language. The dissertation's main contribution is the definition of a formula and an efficient algorithm to address the problem of computing the probability of occurrence of a discontinued sequence of items. A direct evaluation of word sequences is done through the comparison of their expected and observed frequency. The evaluation presented is unsupervised and does not depend on the intended use of the phrases.