Learning action sequences for decision-making in home automation systems

Knowing the behavior habits of the individuals can contribute to decision making in human-centered environments. This work presents a learning model within the IntelliDomo project, a system which is able to learn the individuals´ habits and make decisions in order to automatically generate production rules that anticipate the users´ frequent and periodic activities. The learning layer incorporates new features such as the detection of action sequences, since users´ habits can be better defined if they are related to chained actions, creating action-action relationships.