Search Results (20)

Background and Objective: The intricate connection between daily behaviours and health necessitates robust monitoring, particularly with the advent of Internet of Things (IoT) systems. This study introduces an innovative approach that exploits the synergy of information from various IoT sources to assess the alignment of behavioural routines with health guidelines. The goal is to improve the readability of behaviour models and provide actionable insights for healthcare professionals. Method: We integrate data from ambient sensors, smartphones, and wearable devices to acquire daily behavioural routines by employing process mining (PM) techniques to generate interpretable behaviour models. These routines are grouped according to compliance with health guidelines, and a clustering method is used to identify similarities in behaviours and key characteristics within each cluster. Results: Applied to an elderly care case study, our approach categorised days into three physical activity levels (Insufficient, Sufficient, Desirable) based on daily step thresholds. The integration of multi-source data revealed behavioural variations not detectable through single-source monitoring. We demonstrated that the proposed visualisations in calendar and timeline views aid health experts in understanding patient behaviours, enabling longitudinal monitoring and clearer interpretation of behavioural trends and precise interventions. Notably, the approach facilitates early detection of behaviour changes during contextual events (e.g., COVID-19 lockdown and Ramadan), which are available in our dataset. Conclusions: By enhancing interpretability and linking behaviour to health guidelines, this work signifies a promising path for behavioural analysis and discovering variations to empower smart healthcare, offering insights into patient health, personalised interventions, and healthier routines through continuous monitoring with IoT-driven data analysis. Full article

Traffic crash prediction (TCP) is a fundamental problem for intelligent transportation systems in smart cities. Improving the accuracy of traffic crash prediction is important for road safety and effective traffic management. Owing to recent advances in artificial neural networks, several new deep-learning models have been proposed for TCP. However, these works mainly focus on accidents in regions, which are typically pre-determined using a grid map. We argue that TCP for roads, especially for crashes at or near road intersections which account for more than 50% of the fatal or injury crashes based on the Federal Highway Administration, has a significant practical and research value and thus deserves more research. In this paper, we formulate TCP at Road Intersections as a classification problem and propose a three-phase data-driven deep learning model, called Road Intersection Traffic Crash Prediction (RoadInTCP), to predict traffic crashes at intersections by exploiting publicly available heterogeneous big data. In Phase I we extract discriminative latent features called topological-relational features (tr-features), of intersections using a neural network model by exploiting topological information of the road network and various relationships amongst nearby intersections. In Phase II, in addition to tr-features which capture some inherent properties of the road network, we also explore additional thematic information in terms of environmental, traffic, weather, risk, and calendar features associated with intersections. In order to incorporate the potential correlation in nearby intersections, we utilize a Graph Convolution Network (GCN) to aggregate features from neighboring intersections based on a message-passing paradigm for TCP. While Phase II serves well as a TCP model, we further explore the signals embedded in the sequential feature changes over time for TCP in Phase III, by exploring RNN or 1DCNN which have known success on sequential data. Additionally, to address the serious issues of imbalanced classes in TCP and large-scale heterogeneous big data, we propose an effective data sampling approach in data preparation to facilitate model training. We evaluate the proposed RoadInTCP model via extensive experiments on a real-world New York City traffic dataset. The experimental results show that the proposed RoadInTCP robustly outperforms existing methods. Full article

We have designed a real-world smart building energy fault detection (SBFD) system on a cloud-based Databricks workspace, a high-performance computing (HPC) environment for big-data-intensive applications powered by Apache Spark. By avoiding a Smart Building Diagnostics as a Service approach and keeping a tightly centralized design, the rapid development and deployment of the cloud-based SBFD system was achieved within one calendar year. Thanks to Databricks’ built-in scheduling interface, a continuous pipeline of real-time ingestion, integration, cleaning, and analytics workflows capable of energy consumption prediction and anomaly detection was implemented and deployed in the cloud. The system currently provides fault detection in the form of predictions and anomaly detection for 96 buildings on an active military installation. The system’s various jobs all converge within 14 min on average. It facilitates the seamless interaction between our workspace and a cloud data lake storage provided for secure and automated initial ingestion of raw data provided by a third party via the Secure File Transfer Protocol (SFTP) and BLOB (Binary Large Objects) file system secure protocol drivers. With a powerful Python binding to the Apache Spark distributed computing framework, PySpark, these actions were coded into collaborative notebooks and chained into the aforementioned pipeline. The pipeline was successfully managed and configured throughout the lifetime of the project and is continuing to meet our needs in deployment. In this paper, we outline the general architecture and how it differs from previous smart building diagnostics initiatives, present details surrounding the underlying technology stack of our data pipeline, and enumerate some of the necessary configuration steps required to maintain and develop this big data analytics application in the cloud. Full article

The advancement of technology has significantly improved energy measurement systems. Recent investment in smart meters has enabled companies and researchers to access data with the highest possible temporal disaggregation, on a minute-by-minute basis. This research aimed to obtain data with the highest possible temporal and spatial disaggregation. This was achieved through a process of energy consumption measurements for six devices within seven houses, located in different communes (counties) of the Metropolitan Region of Chile. From this process, a data panel of energy consumption of six devices was constructed for each household, observed in two temporal windows: one quarterly (750,000+ observations) and another semi-annual (1,500,000+ observations). By applying a panel data econometric model with fixed effects, calendar-temporal patterns that help explain energy consumption in each of these two windows have been studied, obtaining explanations of over 80% in some cases, and very low in others. Sensitivity analyses show that the results are robust in a short-term temporal horizon and provide a practical methodology for analyzing energy consumption determinants and load profiles with panel data. Moreover, to the authors’ knowledge, these are the first results obtained with data from Chile. Therefore, the findings provide key information for the planning of production, design of energy market mechanisms, tariff regulation, and other relevant energy policies, both at local and global levels. Full article

Short-term load forecasting (STLF) is essential for urban sustainable development. It can further contribute to the stable operation of the smart grid. With the development of renewable energy, improving STLF accuracy has become a vital task. Nevertheless, most models based on the convolutional neural network (CNN) cannot effectively extract the crucial features from input data. The reason is that the fundamental requirement of adopting the convolutional neural network (CNN) is space invariance, which cannot be satisfied by the received data, limiting the forecasting performance. Thus, this paper proposes an innovative ensemble model that comprises a densely residual block (DRB), bidirectional long short-term memory (Bi-LSTM) layers based on the attention mechanism, and ensemble thinking. Specifically, the DRB is adopted to extract the potential high-dimensional features from different types of data, such as multi-scale load data, temperature data, and calendar data. The extracted features are the input of the Bi-LSTM layer. Then, the adopted attention mechanism can assign various weights to the hidden state of Bi-LSTM and focus on the crucial factors. Finally, the proposed two-stage ensemble thinking can further improve model generalization. The experimental results show that the proposed model can produce better forecasting performance compared to the existing ones, by almost 3.37–5.94%. Full article

Currently, most machine learning applications follow a one-off learning process: given a static dataset and a learning algorithm, generate a model for a task. These applications can neither adapt to a dynamic and changing environment, nor accomplish incremental task performance improvement continuously. STEP perpetual learning, by continuous knowledge refinement through sequential learning episodes, emphasizes the accomplishment of incremental task performance improvement. In this paper, we describe how a personalized temporal event scheduling system SmartCalendar, can benefit from STEP perpetual learning. We adopt the interval temporal logic to represent events’ temporal relationships and determine if events are temporally inconsistent. To provide strategies that approach user preferences for handling temporal inconsistencies, we propose SmartCalendar to recognize, resolve and learn from temporal inconsistencies based on STEP perpetual learning. SmartCalendar has several cornerstones: similarity measures for temporal inconsistency; a sparse decomposition method to utilize historical data; and a loss function based on cross-entropy to optimize performance. The experimental results on the collected dataset show that SmartCalendar incrementally improves its scheduling performance and substantially outperforms comparison methods. Full article

Lithium-ion batteries on electric vehicles have been increasingly deployed for the enhancement of grid reliability and integration of renewable energy, while users are concerned about extra battery degradation caused by vehicle-to-grid (V2G) operations. This paper details a multi-year cycling study of commercial 24 Ah pouch batteries with Li(NiMnCo)O₂ (NCM) cathode, varying the average state of charge (SOC), depth of discharge (DOD), and charging rate by 33 groups of experiment matrix. Based on the reduced freedom voltage parameter reconstruction (RF-VPR), a more efficient non-intrusive diagnosis is combined with incremental capacity (IC) analysis to evaluate the aging mechanisms including loss of lithium-ion inventory and loss of active material on the cathode and anode. By analyzing the evolution of indicator parameters and the cumulative degradation function (CDF) of the battery capacity, a non-linear degradation model with calendar and cyclic aging is established to evaluate the battery aging cost under different unmanaged charging (V0G) and V2G scenarios. The result shows that, although the extra energy throughput would cause cyclic degradation, discharging from SOC 90 to 65% by V2G will surprisingly alleviate the battery decaying by 0.95% compared to the EV charged within 90–100% SOC, due to the improvement of calendar life. By optimal charging strategies, the connection to the smart grid can potentially extend the EV battery life beyond the scenarios without V2G. Full article

In the precision viticulture domain, data recorded by monitoring devices are large-scale processed to improve solutions for grapes’ quality and global production and to offer various recommendations to achieve these goals. Soil-related parameters (soil moisture, structure, etc.) and atmospheric parameters (precipitation, cumulative amount of heat) may facilitate crop diseases occurrence; thus, following predictive analysis, their estimation in vineyards can offer an early-stage warning for farmers and, therefore, suggestions for their prevention and treatment are of particular importance. Using remote sensing devices (e.g., satellites, unmanned vehicles) and proximal sensing methods (e.g., wireless sensor networks (WSNs)), we developed an efficient precision agriculture telemetry platform to provide reliable assessments of atmospheric phenomena periodicity and crop diseases estimation in a vineyard near Bucharest, Romania. The novelty of the materials and methods of this work relies on providing comprehensive preliminary references about monitored parameters to enable efficient, sustainable agriculture. Comparative analyses for two consecutive years illustrate an excellent correlation between cumulative and daily heat, precipitation quantity, and daily evapotranspiration (ET). In addition, the platform proved viable for wine-grapes disease estimation (powdery mildew, grape bunch rot, and grape downy mildew) and treatment recommendations based on the elaborated phenological calendar. Our results, together with continuous monitoring for the upcoming years, may be used as a reference to perform productive, sustainable smart agriculture in terms of yield and crop quality in Romania. In the Conclusion section, we show that farmers and personnel from cooperatives can use this information to make assessments based on the correlation of the available data to avoid critical damage to the wine-grape. Full article

In this paper, we analyze the drivers of the adoption of climate-smart agricultural (CSA) practices and the impact of their adoption on farm net returns and exposure to risks. We use recent farm-level data from three agroecological zones of Pakistan to estimate a multinomial endogenous switching regression for different CSA practices used to reduce the adverse impact of climate change. These strategies include changing input mix, changing cropping calendar, diversifying seed variety, and soil and water conservation measures. The empirical results show that the adoption of different CSA practices is influenced by average rainfall, previous experience of climate-related shocks, and access to climate change information. The findings further reveal that adoption of CSA practices positively and significantly improves farm net returns and reduces farmers’ exposure to downside risks and crop failure. The results also reveal significant differences in the impacts of CSA practice adoption on farm net returns in different agroecological zones. Thus, policies aimed at achieving sustainability in agricultural production should consider agroecological, specific, climate-smart solutions. Full article

As COVID-19 solidifies its presence in everyday life, the interest in mental health is growing, resulting in the necessity of sentiment analysis. A smart mirror is suitable for encouraging mental comfort due to its approachability and scalability as an in-home AI device. From the aspect of natural language processing (NLP), sentiment analysis for Korean lacks an emotion dataset regarding everyday conversation. Its significant differences from English in terms of language structure make implementation challenging. The proposed smart mirror LUX provides Korean text sentiment analysis with the deep learning model, which examines GRU, LSTM, CNN, Bi-LSTM, and Bi-GRU networks. There are four emotional labels: anger, sadness, neutral, and happiness. For each emotion, there are three possible interactive responses: reciting wise sayings, playing music, and sympathizing. The implemented smart mirror also includes more-typical functions, such as a wake-up prompt, a weather reporting function, a calendar, a news reporting function, and a clock. Full article

Current changes in climate conditions due to global warming affect the phenological behavior of economically important cultivable plant species, with consequences for the food security of many countries, particularly in small vulnerable islands. Thus, the objective of this study was to evaluate the thermal viability of Solanum tuberosum (L.) through the behavior of the Thermal Index of Biological Development (ITDB) of two cultivation areas in Cuba under different climate change scenarios. For the analysis, we elaborated bioclimatic scenarios by calculating the ITDB through a grounded and parameterized stochastic function based on the thermal values established for the phenological development of the species. We used the mean temperature values from the period 1980 to 2010 (historical reference period) of the Meteorological Stations: 78320 “Güira de Melena” and 78346 “Venezuela”, located at the western and central of Cuba respectively. We also used modeled data from RCP 2.6 scenarios; 4.5 and 8.5 from the PRECIS-CARIBE Regional Climate Model, which used global outputs from the ECHAM5 MCG for the period 2010 to 2100. As result, the scenarios showed that the annual average ITDB ranges from 0.7 to 0.8, which indicates that until 2010 there were temporary spaces with favorable thermal conditions for the species, but not for the period from 2010 to 2100 in RCP 4.5 and 8.5. In these scenarios, there was a progressive decrease in the indicator that warned of a marked loss of Viability of S. tuberosum, reduction of the time-space to cultivate this species (particularly the month of April is the most inappropriate for the ripening of the tuber). These results showed that Cuba requires the establishment of an adaptation program with adjustments in the sowing and production calendar, the use of short-cycle varieties of less than 120 days, the management of genotypes adaptable to high temperatures, and the application of “Agriculture Climate Smart”, to reduce risks in food safety. Full article

Objective: To create a mobile application able to help patients follow medical treatments properly. Methods: We designed and developed a custom Android/iOS App to remind patients of the pharmaceutical drugs to be taken, of the visits and exams to attend, and to detect their compliance with their personal therapeutic plan. In this paper we describe the App development, UX/UI design, Gamification. TuTOR is an Android and iOS application designed to remind patients of the drugs to be taken, giving them all the information related to their therapeutic plans in a simple and non-invasive way. Thanks to a dedicated back-office, specially designed to meet specific medical information needs, the App can also help physicians detect their patients’ compliance with their treatments and modify prescriptions in real time. The App also ensures a state-of-the-art approach to data security and privacy protection. The main feature of TuTOR is the smart therapy assistant, which features dedicated alarms to remind users of taking their prescription drugs. Thanks to the automatic synchronization with a local database, the alert system works even without connection to the Internet. Particular attention was paid during the App’s design process: we looked to create an intuitive interface to ensure absolute ease of use, with state-of-the-art visual design aimed at maximizing user experience. Other relevant features include the App’s ability to givevisual evidence of the most important drugs to be taken and its note-taking feature, which gives patients the possibility to note down indications on why a specific drug was skipped. The App also keeps track of upcoming medical exams, laboratory tests, and visits on a devoted calendar. It also helps patients by listing therapy contacts, such as physicians’ phone numbers, and indicates all medical references by showing, for example, locations of relevant clinics and pharmacies on a map. Thanks to specific visual progress indicators and an innovative gamification approach, the App encourages users to faithfully follow therapy guidelines. With TuTOR, assessing the therapy’s state of completion is quick and easy.Thanksto the privacy-by-design approach used, all data managed by the system is compliant with the European Privacy Regulation and it is not available to third parties. Expected results: A mobile App for medication adherence might increase objectively and subjectively measured adherence. Full article

The exponential growth in population and their overall reliance on the usage of electrical and electronic devices have increased the demand for energy production. It needs precise energy management systems that can forecast the usage of the consumers for future policymaking. Embedded smart sensors attached to electricity meters and home appliances enable power suppliers to effectively analyze the energy usage to generate and distribute electricity into residential areas based on their level of energy consumption. Therefore, this paper proposes a clustering-based analysis of energy consumption to categorize the consumers’ electricity usage into different levels. First, a deep autoencoder that transfers the low-dimensional energy consumption data to high-level representations was trained. Second, the high-level representations were fed into an adaptive self-organizing map (SOM) clustering algorithm. Afterward, the levels of electricity energy consumption were established by conducting the statistical analysis on the obtained clustered data. Finally, the results were visualized in graphs and calendar views, and the predicted levels of energy consumption were plotted over the city map, providing a compact overview to the providers for energy utilization analysis. Full article

Understanding human behavior can assist in the adoption of satisfactory health interventions and improved care. One of the main problems relies on the definition of human behaviors, as human activities depend on multiple variables and are of dynamic nature. Although smart homes have advanced in the latest years and contributed to unobtrusive human behavior tracking, artificial intelligence has not coped yet with the problem of variability and dynamism of these behaviors. Process mining is an emerging discipline capable of adapting to the nature of high-variate data and extract knowledge to define behavior patterns. In this study, we analyze data from 25 in-house residents acquired with indoor location sensors by means of process mining clustering techniques, which allows obtaining workflows of the human behavior inside the house. Data are clustered by adjusting two variables: the similarity index and the Euclidean distance between workflows. Thereafter, two main models are created: (1) a workflow view to analyze the characteristics of the discovered clusters and the information they reveal about human behavior and (2) a calendar view, in which common behaviors are rendered in the way of a calendar allowing to detect relevant patterns depending on the day of the week and the season of the year. Three representative patients who performed three different behaviors: stable, unstable, and complex behaviors according to the proposed approach are investigated. This approach provides human behavior details in the manner of a workflow model, discovering user paths, frequent transitions between rooms, and the time the user was in each room, in addition to showing the results into the calendar view increases readability and visual attraction of human behaviors, allowing to us detect patterns happening on special days. Full article

There is a rapid growing demand for highly sensitive, easy adaptive and low-cost pressure sensing solutions in the fields of health monitoring, wearable electronics and home care. Here, we report a novel flexible inductive pressure sensor array with ultrahigh sensitivity and a simple construction, for large-area contact pressure measurements. In general, the device consists of three layers: a planar spiral inductor layer and ferrite film units attached on a polyethylene terephthalate (PET) membrane, which are separated by an array of elastic pillars. Importantly, by introducing the ferrite film with an excellent magnetic permeability, the effective permeability around the inductor is greatly influenced by the separation distance between the inductor and the ferrite film. As a result, the value of the inductance changes largely as the separation distance varies as an external load applies. Our device has achieved an ultrahigh sensitivity of 1.60 kPa⁻¹ with a resolution of 13.61 Pa in the pressure range of 0–0.18 kPa, which is comparable to the current state-of-the-art flexible pressure sensors. More remarkably, our device shows an outstanding stability when exposed to environmental interferences, e.g., electrical noises from skin surfaces (within 0.08% variations) and a constant pressure load for more than 32 h (within 0.3% variations). In addition, the device exhibits a fast response time of 111 ms and a good repeatability under cyclic pressures varying from 38.45 to 177.82 Pa. To demonstrate its practical usage, we have successfully developed a 4 × 4 inductive pressure sensor array into a wearable keyboard for a smart electronic calendar application. Full article