Indian Journal of Science and Technology

Abstract

Background/Objectives: In recent times, urban areas are turned to be a smart dewing spot. In smart cities, most things are done automatically using smart devices such as sensors and smart meters. These smart devices produce large volumes of fine-grained data and stored into servers. To make a study on how to train and detect human health behaviour patterns for healthcare applications from smart meter data. Methods/Statistical Analysis: To analyze and detect human activity patterns, we build an Efficient Mining technique using frequent pattern mining and clustering techniques. Here, we consider the UK-Dale dataset incorporates time series information of intensity utilization gathered somewhere in the range of 2012 and 2015. To training and detecting human pattern we build an Efficient Mining technique. Findings: The identification of human behaviour patterns for appliance usage using this technique is better than existing techniques with accuracy for short and long term predictions. Applications/Improvements: We also extend our work to analyze temporal energy consumption patterns at the appliance level, which is directly related to human activities.

Keywords: Behavioural Analysis, Big Data, Data Clustering, Frequent Pattern Mining, Smart Cities