Knowledge Engineering is at the crossroads of Industrial Engineering (Data, Information, Knowledge theme of the GDR MACS) and Artificial Intelligence (Knowledge Representation and Reasoning Modeling theme of the GDR IA).
This scientific discipline is based on knowledge models that are used to prepare for decisionmaking, whatever the field of application.
The objective of this scientific discipline is to provide researchers with the resources necessary to maintain (or increase) their skills in the tools and methods for extracting, modelling and exploiting knowledge.
Different scientific approaches emerge from this objective:
- The creation or enrichment of knowledge models goes through a phase of knowledge extraction which can be carried out either by interviewing experts or directly from databases (massive or not). During the knowledge exploitation phase, a system of recommendations can also be very beneficial to the user when making decisions.
- Several distinct types of knowledge are formalized in Knowledge Engineering: mathematical functions, compatibility, optionality, for example. This formalization requires the coordinated use of several methods of formalization and exploitation of knowledge. In addition, knowledge models may require several occurrences of the same element (activities, components, etc.). These multiple occurrences require formalization and exploitation of specific knowledge.
- When exploiting knowledge, users sometimes needs to know which capitalized knowledge has been mobilized and exploited in order to prepare their decision-making. They can thus easily question some of these choices that have led them either to a non-acceptable solution or to a non-solution.
- The user interface has a non-negligible impact on decision-making. An outlook on human and social sciences is therefore a definite advantage for conducting research.