SOBEK is able to identify relevant concepts in a text from the frequency analysis of textual material in these terms. SOBEK uses a process known as text mining, defined as a method of extracting relevant information in unstructured or semi-structured data. Having its origins in data mining, text mining has become increasingly popular, mainly due to the growth of the Internet and developments in the area of computational linguistics. The technique differs from data mining by the fact that the later searches for patterns of information in formal database records, while text mining employs unstructured data sources.

Sobek has been developed at the Graduate Program of Computers in Education, at the Federal University of Rio Grande do Sul (UFRGS), Brazil. It has been used in several educational settings, providing support to pedagogical applications in elementary school, high school, higher education and distance learning.

Sobek is distributed freely and may be downloaded here.