Despite the fact that Artificial Intelligence (AI) has boosted the achievement of remarkable results across numerous data analysis tasks, however, this is typically accompanied by a significant short-coming in the exhibited transparency and trustworthiness of the developed systems. In order to address the latter challenge, the so-called eXplainable AI (XAI) research field has emerged, which aims, among others, at estimating meaningful explanations regarding the employed model’s reasoning process. The current study focuses on systematically analyzing the recent advances in the area of Multimodal XAI (MXAI), which comprises methods that involve multiple modalities in the primary prediction and explanation tasks. In particular, the relevant AI-boosted prediction tasks and publicly available datasets used for learning/evaluating explanations in multimodal scenarios are initially described. Subsequently, a systematic and comprehensive analysis of the MXAI methods of the literature is provided, taking into account the following key criteria: a) The number of the involved modalities (in the employed AI module), b) The processing stage at which explanations are generated, and c) The type of the adopted methodology (i.e. the actual mechanism and mathematical formalization) for producing explanations. Then, a thorough analysis of the metrics used for MXAI methods’ evaluation is performed. Finally, an extensive discussion regarding the current challenges and future research directions is provided.

Excessive domestic energy usage is an impediment towards energy efficiency. Developing countries are expected to witness an unprecedented rise in domestic electricity in the forthcoming decades. A large amount of research has been directed towards behavioral change for energy efficiency. Thus, it is prudent to develop an intelligent system that combines the proper use of technology with behavior change research in order to sustainably transform end-user behavior at a large scale. This paper presents an overview of our AI-based energy efficiency framework for domestic applications and explains how micro-moments can provide an accurate understanding of user behavior and lead to more effective recommendations. Micromoments are short-term events at which an energy-saving recommendation is presented to the consumer. They are detected using a variety of sensing modules placed at prominent locations in the household. A supervised machine learning classifier is then used to analyze the acquired micro-moments, identify abnormalities, and formulate a list of energy-saving recommendations. Each recommendation is presented through the end-user mobile application. The results so far include a mobile application in the front-end and a set of sensing modules in the backend that comprise, an ensemble bagging-trees micro-moment classifier (achieving up to 99.64% accuracy and 98.8% F-score), and a recommendation engine.