Abordagem de Inteligência de Fontes Abertas com Mapas Auto-Organizáveis De Kohonen e Processamento de Linguagem Natural para Execução Automática de Dorks

In the initial phase of the pentest, named Open Source Intelligence (OSINT), Google Hacking (GH) is executed, a practice that uses strings composed by search operators called Dorks. The objective of this work was to develop and apply an Open Source Intelligence approach with Kohonen Self-Organizing Maps and Natural Language Processing in the automatic execution of Dorks to improve the performance of Google Hacking practice. The proposed approach was developed in 10 phases, including the application of Natural Language Processing in the pre-processing of Dorks and Self-Organizing Maps in the generation of 53 groupings of Dorks. Keywords: OSINT, Open Source Intelligence, Natural Language Processing, Kohonen Self Organizing-Maps, Automatic Pentest. 1.