Self-Adaptation of Activity Recognition Systems to New Sensors

TL;DR

This paper introduces a method for activity recognition systems to adapt to new sensors dynamically using semi-supervised learning, enabling them to leverage additional data sources with minimal user input.

Contribution

It presents a novel opportunistic activity recognition approach that handles changing sensor inputs through unsupervised clustering and semi-supervised label inference.

Findings

Evaluated over 3000 sensor combinations across three experiments.

Demonstrated potential benefits of adaptive sensor integration.

Showed effectiveness of semi-supervised clustering in activity recognition.

Abstract

Traditional activity recognition systems work on the basis of training, taking a fixed set of sensors into account. In this article, we focus on the question how pattern recognition can leverage new information sources without any, or with minimal user input. Thus, we present an approach for opportunistic activity recognition, where ubiquitous sensors lead to dynamically changing input spaces. Our method is a variation of well-established principles of machine learning, relying on unsupervised clustering to discover structure in data and inferring cluster labels from a small number of labeled dates in a semi-supervised manner. Elaborating the challenges, evaluations of over 3000 sensor combinations from three multi-user experiments are presented in detail and show the potential benefit of our approach.