Artificial Intelligence in the generation of conceptual models from text

In the development of digital systems, the input for many activities is in natural language. Problems, stakeholders, goals, requirements, specifications, processes, are largely described in textual documents. Such documents can be of considerable size. Achievements in automatic natural language processing (NLP), one of the traditional areas of artificial intelligence, have suggested since the second half of the last century that linguistic tools can be used to support the creation of conceptual models from texts. In this context, the seminar will illustrate the evolution of systems to support this activity, in the light of the dichotomy symbolic vs. statistical (sub-symbolic) systems, referring to the experience gained in research projects related to different application areas