Abstract

The accident investigation process in oil industry is a critical activity. In accident investigation, the volume of information collected and analyzed, makes analysis of its causes a challenging task. Making sense of its large volume of information is also a challenging task because of the diverse interpretations made by experts and auditors. With the advances of new technologies, data of these systems have become increasingly huge. The main objective is to propose and evaluate information extraction techniques in occupational health control process, particularly, for automatic detection of accidents from unstructured texts. Our proposal divides the problem in subtasks such as text analysis, recognition and classification of failed occupational health control, resolving accidents.