Nowadays, Generative Large Language Models (GLLMs) have made a significant impact in the field of Artificial Intelligence (AI). One of the domains extensively explored for these models is their ability as generators of functional source code for software projects. Nevertheless, their potential as assistants to write the code needed to generate and model Machine Learning (ML) or Deep Learning (DL) architectures has not been fully explored to date. For this reason, this work focuses on evaluating the extent to which different tools based on GLLMs, such as ChatGPT or Copilot, are able to correctly define the source code necessary to generate viable predictive models. The use case defined is the forecasting of a time series that reports the indoor temperature of a greenhouse. The results indicate that, while it is possible to achieve good accuracy metrics with simple predictive models generated by GLLMs, the composition of predictive models with complex architectures using GLLMs is still far from improving the accuracy of predictive models generated by human data scientists.

Countries all around the world recognise the impact of air quality on public health, advocating for city centre decarbonisation and pollutant monitoring via Internet of Things technologies. Using data collected from these systems, it is possible to generate models that predict pollution based on regular patterns where mobility data can enhance the accuracy and robustness of these advanced machine learning models. This paper follows this approach, utilising vehicle traffic data from image recognition, on-site vehicle detectors, and synthetic data to maximise prediction accuracy in various urban environments. The results reveal that this proposal improves prediction for traffic-related pollutants, such as ${\text{SO}}_{2}$ and ${\text{PM}}_{2.5}$, which are linked to severe respiratory diseases. These results also highlight the role of synthetic data in enhancing prediction performance under limited datasets.

Nowadays, air pollution is one of the most relevant environmental problems in most urban settings. Due to the utility in operational terms of anticipating certain pollution levels, several predictors based on Graph Neural Networks (GNN) have been proposed for the last years. Most of these solutions usually encode the relationships among stations in terms of their spatial distance, but they fail when it comes to capturing other spatial and feature-based contextual factors. Besides, they assume a homogeneous setting where all the stations are able to capture the same pollutants. However, large-scale settings frequently comprise different types of stations, each one with different measurement capabilities. For that reason, the present article introduces a novel GNN framework able to capture the similarities among stations related to the land use of their locations and their primary source of pollution. Furthermore, we define a methodology to deal with heterogeneous settings on the top of the GNN architecture. Finally, the proposal has been tested with a nation-wide Spanish air-pollution dataset with very promising results.

According to many analysts, February 2022 is dated as the exacerbation of the armed conflict between Ukraine and Russia. Of course, this shocking event has an impact on mood of people. In this context, we studied the music listening habits of people in Eastern European countries. We applied timeseries and statistical analysis of mood-relate features (danceability, valence and energy) to a Spotify dataset in order to define new paradigms of emotion-aware systems that can be applied to detect the population’s state of mind. The results show that the trend in the type of music changes notably towards songs with less genre diversity and reflecting less positivity.

Abstract We are witnessing the digitalization era, where artificial intelligence (AI)/machine learning (ML) models are mandatory to transform this data deluge into actionable information. However, these models require large, high-quality datasets to predict high reliability/accuracy. Even with the maturity of Internet of Things (IoT) systems, there are still numerous scenarios where there is not enough quantity and quality of data to successfully develop AI/ML-based applications that can meet market expectations. One such scenario is precision agriculture, where operational data generation is costly and unreliable due to the extreme and remote conditions of numerous crops. In this paper, we investigated the generation of synthetic data as a method to improve predictions of AI/ML models in precision agriculture. We used generative adversarial networks (GANs) to generate synthetic temperature data for a greenhouse located in Murcia (Spain). The results reveal that the use of synthetic data significantly improves the accuracy of the AI/ML models targeted compared to using only ground truth data.

Due to the COVID-19 pandemic, most universities have adapted their learning infrastructure to an increasing demand for online training modalities. However, this type of learning, usually through Learning Management Systems (LMSs), suffer from a lack of direct feedback between students and the educational staff. For that reason, the present work introduces the EMO-learning project, whose key goal is to capture the emotions of students. This is done by means of a deep learning approach, able to timely analyse the face expressions of the students during online lectures. The module has been tested with different students during the academic year 2020-21, showing quite promising results.

Nowadays, cities are the most relevant type of human settlement and their population has been endlessly growing for decades. At the same time, we are witnessing an explosion of digital data that capture many different aspects and details of city life. This allows detecting human mobility patterns in urban areas with more detail than ever before. In this context, based on the fusion of mobility data from different and heterogeneous sources, such as public transport, transport-network connectivity and Online Social Networks, this study puts forward a novel approach to uncover the actual land use of a city. Unlike previous solutions, our work avoids a time-invariant approach and it considers the temporal factor based on the assumption that urban areas are not used by citizens all the time in the same manner. We have tested our solution in two different cities showing high accuracy rates.

Energy efficiency is in the interest of everyone, from individuals to governments, since it yields economical savings, reduces greenhouse gas emissions and alleviates energy poverty. Buildings are one of the largest consumers of primary energy and attaining their efficiency is, therefore, an important goal. The Internet of Things currently provides vast amounts of data that can be used to extract knowledge of all kinds, including that regarding energy prediction. This has motivated us to test wether the prior information on the physics of building heat transfer, that is currently available is now redundant owing to the completeness of the data from the system. We propose a machine learning approach and a grey-box model approach with which to test this hypothesis. The former is blind to the physiscs of the problem, while the latter is greatly influenced by it. The energy consumption prediction models were created with both approaches and then used to estimate energy consumption in a normal operation state and compare it with energy consumption when an energy efficiency campaign is run. Our black-box method, which is based on a combination of statistical and machine learning models and on a time series structurization of the data, shows better prediction accuracy than the so-called grey-box methods that include basic physical equations. This shows that also a data driven approach outperforms more informed methods in this, like other fields.

This paper presents the main foundations of big data applied to smart cities. A general Internet of Things based architecture is proposed to be applied to different smart cities applications. We describe two scenarios of big data analysis. One of them illustrates some services implemented in the smart campus of the University of Murcia. The second one is focused on a tram service scenario, where thousands of transit-card transactions should be processed. Results obtained from both scenarios show the potential of the applicability of this kind of techniques to provide profitable services of smart cities, such as the management of the energy consumption and comfort in smart buildings, and the detection of travel profiles in smart transport.

In the envisioned Internet of Things ecosystems, Smart objects are intended to create groups of devices in order to provide higher level services to be leveraged by citizens. However, because of the dynamic nature of such scenarios, the discovery, management and operation of such dynamic coalitions taking into account security and privacy concerns, is a challenging task that has not been properly addressed yet. In this sense, the present proposal devises a novel approach to automatically compose opportunistic aggregations of objects (bubbles) based on Complex Event Processing (CEP) and fuzzy clustering. While the former detects certain events that could give raise to discover new bubbles, the latter allows compose aggrupations of similar objects acting as candidate bubbles. Finally, the application of the proposal in an educational domain is put forward.

Currently, distributed traffic information systems have come up as one of the most important approaches for detecting traffic flow problems on a road. For that purpose, they usually make use of the location information that vehicles share among them through periodical messages that are transmitted across a vehicular ad hoc network (VANET). This paper puts forward an event-driven architecture (EDA) as a novel mechanism to get insight into VANET messages to detect different levels of traffic jams; furthermore, it also takes into account environmental data that come from external data sources, such as weather conditions. The proposed EDA has been developed through the complex-event-processing technology. Simulation tests show that the proposed mechanism can detect traffic congestions, which involve different numbers of lanes and lengths with short delay.

One of the goals of Advanced Driver Assistance Systems (ADASs) is to identify the role of a vehicle in a scene even without Global Positioning System (GPS) information. Some researches solve the problem by implementing different kinematic models for the vehicle along with a mechanism to decide the most suitable model at the current instant. In this work, a Fuzzy Rule Based Classification System (FRBCS) takes such decision starting from the measures coming from different sensors in the vehicle. The FRBCS is obtained through Data Driven Fuzzy Modeling (DDFM) techniques. In fact, several FRBCSs with promising results are generated. Most discovered models achieve better classification rates than previous researches while being simpler. Therefore, they are more suitable for implementation into an ADAS. Besides, some FRBCSs are far simpler while achieving similar rates. Finally, some tests have been done. They show the feasibility and suitability of this approach behind different situations.