Formal methods for knowledge extraction and reuse from heterogeneous sources for semantic interoperability of distributed architectures

Abstract

The tendency in industry, manufacturing, and agriculture nowadays goes towards adopting the Industry 4.0 practices. Additionally, Internet of Things (IoT) has seen a huge increase in its usage over the last decade, and companies are eager to profit from the advantages it has offers. Between these tendencies, the usage of data as a means to increase productivity, or similarly, to minimize loss in production is found. In those lines, Formal Concept Analysis (FCA) is a clusterization method whose output is based on patterns of concepts (sets of objects and attributes). Some extensions such as Relational Concept Analysis have arisen to tackle the use case in which there are relations between seemingly different objects, which is something FCA cannot do. However, the area of automatically using the conceptual data resulted from these methods is still immature in the sense of formalization and usage. In this Ph.D., the goal is to work in expanding the boundaries of knowledge regarding the existing algorithms, mainly looking for optimizations, and extending their current capabilities.