Abstract
Traffic flow forecasting is an essential component of an intelligent transportation system to mitigate congestion. Recurrent neural networks, particularly gated recurrent units and long short-term memory, have been the stateof-the-art traffic flow forecasting models for the last few years. However, a more sophisticated and resilient model is necessary to effectively acquire long-range correlations in the time-series data sequence under analysis. The dominant performance of transformers by overcoming the drawbacks of recurrent neural networks in natural language processing might tackle this need and lead to successful time-series forecasting. This article presents a multi-head attention based transformer model for traffic flow forecasting with a comparative analysis between a gated recurrent unit and a long-short term memory-based model on PeMS dataset in this context. The model uses 5 heads with 5 identical layers of encoder and decoder and relies on Square Subsequent Masking techniques. The results demonstrate the promising performance of the transform-based model in predicting long-term traffic flow patterns effectively after feeding it with substantial amount of data. It also demonstrates its worthiness by increasing the mean squared errors and mean absolute percentage errors by (1.25 - 47.8)% and (32.4 - 83.8)%, respectively, concerning the current baselines.

Abstract
This study explores the resistive switching phenomena present in 4 mu m(2) amorphous Indium-Gallium-Zinc Oxide (IGZO) memristors. Despite being extensively reported in the literature, not many studies detail the mechanisms that dominate conduction on the different states of IGZO-based devices. In this article, we demonstrate that resistive switching occurs due to the modulation of the Schottky barrier present at the bottom interface of the device. Furthermore, thermionic field emission and field emission regimes are identified as the dominant conduction mechanisms at the high resistive state of the device, while the bulk-limited ohmic conduction is found at the low resistive state. Due to the high complexity associated with creating compact models of resistive switching, a data-driven model is drafted taking systematic steps. (C) 2022 Author(s).

Abstract
The instruments developed for the upcoming Extremely Large Telescopes (ELTs) will need efficient adaptive optics (AO) systems to correct the effects of the atmospheric turbulence and allow imaging at the highest angular resolution. One of the most important requirements for ELT AO-assisted instruments will be to deliver diffraction-limited images in a significant part of the sky. For that, the instruments will be equipped with laser guide stars (LGSs) providing most of the information required by AO instruments. But even with LGSs, AO systems still require the use of natural guide stars (NGSs) to compensate for image motion (jitter) and some low order aberrations. These NGSs are eventually limiting the fraction of the sky that can be achieved by AO systems, the so-called sky coverage (SC). We first present the SC assessment methods used for high angular resolution monolithic optical and near-infrared integral field spectrograph (HARMONI) and multiconjugate adaptive optics relay/multi-AO imaging camera for deep observations (MAORY/MICADO), that are both instruments for the ELT of the European Southern Observatory (ESO). They are based on a semianalytical description of the main contributors in the AO error budget, allowing for a fast estimation of the residual jitter. As such, these methods are well suited for statistical estimation of the SC on multiple science fields and/or to efficiently explore the system parameter space. We then compute the SC of the two instruments in cosmological fields from the cosmic assembly near-IR deep extragalactic legacy survey catalog. The goal is to provide an insight on the possibilities given by two different types of tomographic AO systems, i.e., laser tomography AO with HARMONI and multiconjugate AO with MAORY, on the same telescope. In particular, we show that HARMONI and MAORY/MICADO are complementary, meaning that the overall SC of ESO's ELT is much improved for applications common to both systems. (C) The Authors. Published by SPIE under a Creative Commons Attribution 4.0 International License.

Abstract
With the increasing penetration of distributed generations (DGs) in three-phase distribution networks, the traditional radial passive distribution networks are gradually transformed into active distribution networks (ADNs) with DGs, accompanied by a dramatic increase in operational scenarios. To fast analyze the massive scenarios of ADNs, this brief proposes a multi-scenario-based power flow analytical method based on the multi-dimensional holomorphic embedding method. First, DGs are partitioned according to their spatial correlation, and then a power flow analytical method is cast for unbalanced three-phase ADNs. Finally, the method is applied to the ADN power flow analysis for multiple scenarios. Comparisons with other power flow calculation methods validate the high efficiency and computational tractability of the proposed method.

Abstract
The search for a pleasant home has concerned people ever since. Paradoxically, people are facing strong difficulties in finding a decent place to settle their lives in cities. As such, the housing market regained momentum in connection with the development of Smart Cities, where life quality of residents is strongly emphasized. Well-being in the metropolis is affected by a wide variety of factors with housing supply being among the most important, hence stirred by financing costs, construction costs, vacancy rate, sales delay, inflation rate, housing stock, price of agricultural land, and regulation. The present article reviews empirical studies on housing supply for a better understanding of the dynamics in this market, shedding some light on the expectable outcomes of policy actions in the promotion of sustainable housing towards the smart city transition. Our review shows that the long-run price elasticity of housing supply is larger than the short-run, as well as the existence of substantial differences in the price elasticity across countries and regions. As such, overall, the hypothesis of a perfectly elastic supply is rejected. In addition, our review highlights that housing supply is negatively related to financial costs, inflation, sales delay, and the existence of regulatory or physical constraints. Also, the elasticity is lower when there are regulatory constraints. Newfangled strategic interaction models, though overlooked in the literature, reinforce that housing does not fit the perfect competition frame. The review proves that we are in face of a non-competitive market in which policy intervention is required to maximize social welfare; policy packages to grant people access to the housing market may be required. However, policy interventions should be carefully designed and clear, to mitigate their potentially negative impact on the housing supply as adverse results may be harmful to the transition towards a smart city.

Abstract
E-Learning has been widely implemented in educational institutions to improve the learning process and meet the new challenges imposed by technology in the 21st century. The implementation and use of learning management systems (LMS) in organizations have grown exponentially in recent years and have strongly impacted education, especially in higher education. The present systematic review aims to identify studies and research topics that are being addressed to improve the learning process in the e-Learning context in higher education institutions and map the studies on the implementation and use of LMS. The review process examined 30 articles that met the inclusion and exclusion criteria on the topic over the past five years, adopting Kitchenham's systematic review design methodology. The results showed that technology alone would not be enough for e-Learning to become a dominant teaching standard. The digital literacy of the stakeholders is crucial in this process.