We introduce a hybrid multi-modal logic \(\varLambda \) for context-dependent preferences. An individual context-dependent preference is a preference that an agent a has from the point of view of a given context or perspective. The preferences of a may vary from one context to another. But since they are related to the same agent, it is necessary to find common points that unite them. For this purpose, for each agent a, we first consider contexts in which a forms some narrow preferences associated only with a given perspective. Then we define broader contexts in which a combines preferences obtained in narrow ones and gets some complete picture. We also establish an ordering on the set of contexts that allows us to determine that one context is more informative for a than another in the preference choice task. If a considers some contexts to be indistinguishable, then a’s preferences in these contexts coincide. In addition, we define conditions under which a changes his preferences. As a result, we obtain a single composition of a’s preferences across contexts. Next, we aggregate context-dependent preferences of individual agents into different group preferences. In particular, such an aggregation results in priority merge if priorities of agents in a group are taken into account. For groups of agents, we have the same unified composition of preferences as for individual agents. In this paper, we obtain a sound and complete axiomatization of the logic \(\varLambda \) of context-dependent individual and group preferences.
