Tracking changes in user activity from unlabelled smart home sensor data using unsupervised learning methods

Abstract

This paper investigates the utility of unsupervised machine learning and data visualisation for tracking changes in user activity over time. This is done through analysing unlabelled data generated from passive and ambient smart home sensors, such as motion sensors, which are considered less intrusive than video cameras or wearables. The challenge in using unlabelled passive and ambient sensors data for activity recognition is to find practical methods that can provide meaningful information to support timely interventions based on changing user needs, without the overhead of having to label the data over long periods of time. The paper addresses this challenge to discover patterns in unlabelled sensor data using kernel density estimation (KDE) for pre-processing the data, together with t-distributed stochastic neighbour embedding and uniform manifold approximation and projection for visualising changes. The methodology is developed and tested on the Aruba CASAS smart home dataset and focusses on discovering and tracking changes in kitchen-based activities. The traditional approach of using sliding windows to segment the data requires a priori knowledge of the temporal characteristics of activities being identified. In this paper, we show how an adaptive approach for segmentation, KDE, is a suitable alternative for identifying temporal clusters of sensor events from unlabelled data that can represent an activity. The ability to visualise different recurring patterns of activity and changes to these over time is illustrated by mapping the data for separate days of the week. The paper then demonstrates how this can be used to track patterns over longer timeframes which could be used to help highlight differences in the user's day-today behaviour. By presenting the data in a format that can be visually reviewed for temporal changes in activity over varying periods of time from unlabelled sensor data, opens up the opportunity for carers to then initiate further enquiry if variations to previous patterns are noted. This is seen as an accessible first step to enable carers to initiate informed discussions with the service user to understand what may be causing these changes and suggest appropriate interventions if the change is found to be detrimental to their wellbeing.

Data-driven deep density estimation

“…Many data analysis problems, reaching from population analysis to computer vision [6,28], require estimating continuous models from discrete samples. Formally, this is the density estimation problem, where, given a sample fx i g $ pðxÞ, we would like to estimate the probability density function (PDF) p(x).…”

Ambient Assisted Living: A Review of Technologies, Methodologies and Future Perspectives for Healthy Aging of Population

“…An additional important class of approaches, which has been applied in limited cases for AAL systems, are those that are based on unsupervised learning techniques. They can be applied for the clustering of large amounts of unlabelled data, grouping user behaviors, and detecting high-level variations without any prior knowledge or training phase [117,118]. This study has identified several clusters of options for data processing in AAL, depending on the complexity of the target.…”

Can Machine Learning be Moral?

“…However, removing human agency from the training phase also removes any possibility for ethical accountability and oversight from the critical part of machine learning. In response we see a series of human-in-the-loop solutions that have tried to combine the engineering benefits of these approaches with the accountability of human agency (Gupta et al 2020, Sloane et al 2020). These include fairness analytics tools such as AIFairness360, the What-If tool, and explainability tools such as LIME.…”