• problem-solving
• problem-solving ontology
• text mining

Problem solving is an essential skill for engineers, however until now no scholar has tried to precisely define the elements that constitute the domain of knowledge of problem solving.
This thesis has the intent to define all the entities that constitute the problem solving domain through a quantitative approach. The research is based on the analysis of a dataset composed of 21’000 scientific articles concerning problem solving. The first step consisted in designing algorithms for the extraction of all the problem solving entities (methods, tools, approaches, etc.) that constitute the problem solving domain. Thereafter text mining algorithms were also used to determine the intensity of the relationships between each of the entities belonging to the problem solving domain.
Finally, was developed a graphic representation of the ontology, shown as a network composed of 142 nodes and 556 arcs. The nodes represent all the problem solving methodologies, approaches, tools etc. extrapolated through text mining algorithms and the arcs represent the intensity of connections between nodes. This problem solving ontology is intuitive and outlines all the topics an engineer should know to solve simple and complex engineering dilemmas.