View article

Data-driven models and applications in data mining need to deal with uncertainty. Human decisions are made under uncertainty and risk, and machine learning models also need to cope with them. Uncertainty appears under different flavors. They include imprecision, vagueness, randomness. This special issue collects papers related to applications of data mining involving uncertainty and uncertainty management for data mining applications. The first paper by Gao and Inuiguchi is about fuzzy LP problems. The authors propose the use of a new type of necessity measure that permits to express the trade-off between the robustness level and the satisfaction level of the constraints. The paper provides an algorithm for solving this type of problem. The second paper by Inuiguchi and Innan is about multiple criteria decision analysis. The authors focus on imprecision on pairwise comparison matrices. The third paper is …