Stefano Bromuri

Managing residential energy efficiently without compromising comfort is challenging. The key difficulty lies in predicting occupancy in real time. Intelligent control strategies are widely used for residential energy management. As sensors become increasingly common in homes, occupancy prediction has gained significant attention as one such approach. Traditional occupancy prediction methods analyze temporal patterns within individual rooms but overlook spatial connectivity. This lack limits the accuracy of occupancy-forecasting methods. Incorporating spatial information could improve prediction accuracy and enhance the effectiveness of occupancy-based energy management. One promising approach to represent such spatial information is through the use of embeddings, which encode spatial connectivity in a data-driven vector space. Another complementary method is to represent the building layout as a graph, where nodes correspond to rooms and edges capture spatial connectivity. In this study, we evaluate multiple occupancy prediction architectures, including Convolutional Neural Networks (CNN) and Long Short-Term Memory networks (LSTM). These models are further enhanced by positional encoding (PE) and Graph Neural Networks (GNN) that explicitly encode spatial connectivity among rooms. Results indicate that spatial embeddings improve predictive performance, with graph-based methods providing the highest performance. Among all models, the hybrid GNN + PE + CNN yielded the best overall performance. However, considering practical heating control scenarios, balancing comfort, energy savings, and computational complexity, we found that LSTM + PE emerges as the most suitable choice.

Purpose: This literature review examines how core intelligent technological capabilities, including artificial intelligence, machine learning, deep learning, and natural language processing, augment rather than replace service agents across the various stages and tasks of customer service interactions. Design/methodology/approach: This study provides a systematic and cross-disciplinary review of empirical studies on applications of technology in human-to-human service interactions published over the last decade. The resulting 99 studies are analyzed using thematic mapping and interpreted through the lens of socio-technical systems theory and service-dominant logic. Findings: The thematic mapping revealed six central themes on the integration of technology in human-to-human service interactions: (1) pre-service optimization, (2) interaction intelligence, (3) service agent well-being, (4) service agent monitoring, (5) emotion work and (6) collaborative service. These themes are organized into a conceptual framework that highlights key tensions and boundary conditions, forming the basis for a comprehensive future research agenda. Practical implications: The findings underscore the importance of a human-centered approach, demonstrating how technology can enhance service agent roles, promote well-being, and enable collaboration in customer service. Originality/value: The review shifts the focus from technology replacing service agents to technology augmenting their roles. It contributes new insights into how augmentation enhances service agent effectiveness, efficiency and well-being, offering a foundation for sustainable improvements in service interactions.

The integration of emerging AI technologies, particularly Large Language Models (LLMs), is fundamentally reshaping the landscape of software engineering. LLMs offer a wide array of capabilities that can enhance various aspects of software development, including design assistance, automated code analysis and synthesis, and testing. Consequently, their integration into software engineering practices holds the potential to greatly improve efficiency and code quality, marking a notable paradigm shift towards the creation of more intelligent and adaptive systems. This allows the definition of an extended perspective when establishing requirements for building software engineering solutions by incorporating evidence from literature resources and practice. By synthesizing requirements and evidence from scholarly and practitioner sources using LLMs, software engineers can ensure that their solutions are not only technically sound but also align with best practices in fostering social norms and values like inclusivity. Adopting such an approach supports the creation of responsible and inclusive educational software that caters to diverse learning needs and promotes equitable access to educational resources. Moreover, by using LLMs to inform decision-making throughout the software development lifecycle, software engineers can iteratively refine and enhance their solutions based on emerging research findings, thereby ensuring continuous improvement in fostering inclusive educational environments. Hence, this research aims to develop a novel evidence-based software engineering method informed by insights from scientific literature. As a use case, we design and implement a dyslexia-oriented educational software application that supports children in learning to read, guided by this new methodological approach.

Simulating realistic time-domain observations of gravitational waves (GWs) and other events of interest in GW detectors, such as transient noise bursts called glitches, can help in advancing GW data analysis. Simulated data can be used in downstream data analysis tasks by augmenting datasets for signal searches, balancing datasets for machine learning applications, validating detection schemes, and constructing mock data challenges. In this work, we present a conditional derivative GAN (cDVGAN), a novel conditional model in the generative adversarial network framework for simulating multiple classes of time-domain observations that represent gravitational waves (GWs) and detector glitches. cDVGAN can also generate generalized hybrid samples that span the variation between classes through class interpolation in the conditioned class vector. cDVGAN introduces an additional player into the typical 2-player adversarial game of GANs, where an auxiliary discriminator analyzes the first-order derivative time series. Our results show that this provides synthetic data that better capture the features of the original data. cDVGAN conditions on three classes in the time domain, two denoized from LIGO blip and tomte glitch events from its third observing run (O3), and the third representing binary black hole (BBH) mergers. Our proposed cDVGAN outperforms four different baseline GAN models in replicating the features of the three classes. Specifically, our experiments show that training convolutional neural networks (CNNs) with our cDVGAN-generated data improves the detection of samples embedded in detector noise beyond the synthetic data from other state-of-the-art GAN models. Our best synthetic dataset yields as much as a 4.2% increase in area-under-the-curve (AUC) performance, maintaining the same CNN architecture, compared to synthetic datasets from baseline GANs. Moreover, training the CNN with class-interpolated hybrid samples from our cDVGAN outperforms CNNs trained only on the standard classes, when identifying real samples embedded in LIGO detector background between signal-to-noise ratios ranging from 1 to 16 (4% AUC improvement for cDVGAN). We also illustrate an application of cDVGAN in a data augmentation example, showing that it is competitive with a traditional augmentation approach. Lastly, we test cDVGAN’s BBH signals in a fitting-factor study, showing that the synthetic signals are generally consistent with the semianalytical model used to generate the training signals and the corresponding parameter space.