Nuno Nunes

Performance evaluation in non-intrusive load monitoring: Datasets, metrics, and tools-A review

Non‐intrusive load monitoring (also known as NILM or energy disaggregation) is the process of estimating the energy consumption of individual appliances from electric power measurements taken at a limited number of locations in the electric distribution of a building. This approach reduces sensing infrastructure costs by relying on machine learning techniques to monitor electric loads. However, the ability to evaluate and benchmark the proposed approaches across different datasets is key for enabling the generalization of research findings and consequently contributes to the large‐scale adoption of this technology. Still, only recently researchers have focused on creating and standardizing the existing datasets in order to deliver a single interface to run NILM evaluations. Furthermore, there is still no consensus regarding, which performance metrics should be used to measure and report the performance of NILM systems and their underlying algorithms. This paper provides a review of the main datasets, metrics, and tools for evaluating the performance of NILM systems and technologies. Specifically, we review three main topics: (a) publicly available datasets, (b) performance metrics, and (c) frameworks and toolkits. The review suggests future research directions in NILM systems and technologies, including cross‐datasets, performance metrics for evaluation and generalizable frameworks for benchmarking NILM technology.

POSEIDON - Passive-acoustic Ocean Sensor for Entertainment and Interactive Data-gathering in Opportunistic Nautical-activities

Recent years demonstrate an increased interest in Passive Acoustic Monitoring (PAM) applications when studying cetaceans. However, they remain expensive underwater systems and targeted for industrial and military purposes. While the usage of smartphones as acoustic sensors has been observed in terrestrial environments, ocean and nautical PAM applications remain greatly unexplored. This paper presents the design, deployment and testing of a POSEIDON system, used for real-time augmentation of whale-watching experiences. We collect and use cetaceans vocal call acoustic samples (clicks, moans and whistles) and apply machine learning for offline model training and prediction. When discriminating the calls, we find that Extra Trees and Gradient Boosting outperform other classifiers (>0.95 confidence threshold). Collected samples are at disposal to citizen scientists and marine biologists. Future studies involve real-time on-boat user testing.

Fragments of laura: incorporating mobile virtual reality in location aware mobile storytelling experiences

As 360° immersive Mobile Virtual Reality (MVR) experiences are reaching a wider public, thanks to inexpensive and more powerful mobile technology available on the shelves, it is also rapidly growing as a research arena to investigate how to best design such VR experiences. Adding to this research direction, in this paper we report on an exploratory study designed to evaluate users perception in Fragments of Laura, a mobile location-aware storytelling experience that uses MVR in a public setting. Results from the study encourage us to pursuit investigation as the experience was well received by the participants and provided an enjoyable experience, however, we discovered that there are still design challenges to overcome in order for MVR to be widely adopted within a public setting. Such as, finding the balance between multimedia and immersive multimedia, providing an onboarding time for such media and proper locations for the content consumption.

DreamScope: Evaluating Interactive 360° Virtual Reality in a Physical Narrative Art Installation

The widespread affordability of digital technologies is encouraging artists to experiment and make use of them in their artworks. This paper describes the design and in-the-wild evaluation of DreamScope, a bespoke 360° mobile virtual reality (VR) interface designed for immersive interaction in the Lucid Peninsula physical narrative art installation. We evaluated the VR interface by means of a questionnaire based on the Positive and Negative Affect Scale (PANAS) and individual items related with the interaction within the exhibition. Results extracted from 53 subjects highlight how the interface supported engagement, positive affect and reflections on the narrated scenario of the art installation. The study reported in this paper provides evidence of the positive role of 360° mobile VR in strengthening the narrative and the artists intent in the Lucid Peninsula artistic installation.