Abstract
Many healthcare-oriented recipe recommendation methods are proposed in previous research. In some previous research, optimal recommendation methods have been studied. In other some previous research, approximate recommendation methods with random sampling have been studied. However, in these previous studies, the optimal solution and the approximate solution were not compared. In previous research, an optimal
recommendation method considering nutrient intake history or adjustment of ingredient amounts has also not
been studied. In this research, new healthcare-oriented recipe recommendation methods considering nutrient
intake history and adjustment of ingredient amounts are studied. An optimal method, and approximate methods
with random sampling are proposed. The optimal solutions and approximate solutions are compared by some
computational examples. Adaptive recipe recommendations considering nutrient intake history and family
composition by the proposed optimal method are confirmed. The characteristics of the proposed approximate
methods are also confirmed. This research is basic research, and future extended research is required.
