MINING HUMAN ACTIVITY PATTERNS FROM SMART HOME BIG DATA FOR HEALTH CARE APPLICATIONS.

Abstract

In recent years, there's associate ever-increasing migration of individuals to urban areas. Health care
service is one among the foremost difficult aspects that greatly affected by the vast influx of people to town centers.
Consequently, cities round the world square measure finance heavily in digital transformation in a shot to produce
healthier eco systems for individuals. In such a change, innumerable homes square measure being equipped with good
devices (e.g., good meters, sensors), that generate huge volumes of fine-grained and fact knowledge that maybe analyzed
to support good town services. During this project, we tend to propose a model that utilizes good home huge knowledge
as a method of learning and discovering human action patterns for health care applications. We tend to propose the
employment of frequent pattern mining, cluster analysis, and prediction to live and analyze energy usage changes
sparked by occupant’s behaviour. Since people’s habits square measure principally known by everyday routines,
discovering these routines permits us to acknowledge abnormal activities that will indicate people’s difficulties in taking
look after themselves, like not making ready food or not employing a shower/bath. This project addresses the
requirement to investigate temporal energy consumption patterns at the appliance level, that is directly associated with
human activities.