Ubiquitous Computing and Ambient Intelligence

Networked embedded systems are impacting the way we interact with the world around us. They are at the core of the advancements in information and communication technologies which have been driving the fourth generation revolution in today’s industry and networks. This technology allows for better integrated communications, integrated local and global control, supervision and maintenance. Billions of smart devices are being implemented - from smart TVs and cars to other smart health monitors and wearable technologies and this resulted in the development of smart interconnected environments. The core vision of the aforementioned smart interconnected environment is the realization of a reliable and secure two way communication between smart devices. However, the high level of heterogeneity, coupled with the wide scale of smart embedded systems, has magnified the security threats. Traditional security countermeasures and privacy cannot be enforced directly on such systems due to their limited computing capabilities and their diverse set of hardware architectures. In this paper we are proposing a multi-level security approach for smart interconnected environments/networks. We address the security at three main pillars: application level, system level, and network level.

In almost every forum around the world, country leaders are discussing the necessity of creating smart cities. However, even the term “smart city” is diffuse nowadays. Some countries want their cities to become smarter and others want to create smart cities from scratch. Several mappings have been developed around the world to locate the smartest cities. We believe that, since Latin American and Caribbean countries are exploring in the creation of smart cities, a proper mapping and plan is necessary to assure that the efforts in creating smart cities are not a waste. Using a literature review and a survey, we try to determine the state of smart cities development and its technical readiness in the Region.

Event filtering is of paramount importance in large-scale ur- ban sensing, where an enormous quantity of data is generated. Multiple criteria can be considered for filtering, location being one of the most valuable ones. Obtaining high-quality (trustworthy, accurate) location information helps to contextualize the event content and provides trust both on the source producer and on the publication itself. However, IoT-based urban services rely often on cloud architectures, which have no means to support location certification. To meet the need for location certification support in urban distributed event-based systems (DEBS), we propose three different fog architectures targeted at scenarios with mobile event producers.

In this article, we research on the service recommendation strategy in the IoT. The user attribute similarity and the attribute correlation of user and device service are computed, and the recommendation system is recommended based on the calculation results. In order to solve the cold start problem, we propose the tensor linear regression model. The experiment results show the recommendation strategy was effective.

The quality of the time travel prediction is a key factor in the transport of people and goods. This prediction is used in different facets related to management and planning of the transport activity, having special influence in the service quality in public transport. In this paper a methodology to analyse the factors which affect to travel time prediction in routes of road public transport is presented. This methodology uses vehicles GPS data to identify the causes of the travel time variability, georeferencing these causes. The infrastructure elements required, data used and the processing techniques are explained. The methodology was applied to analyse the travel time of a line of a public transport company, presenting the results of this test.

Maintenance of systems such as infrastructure, services and equipment on city context requires a complex management process in order to provide quality services, comply with regulations and extend lifespan of urban equipment and infrastructure without neglecting the proper use of resources to execute maintenance jobs. Cities can generate hundreds or thousands of maintenance jobs during a particular time period. These jobs can be generated automatically by equipment or can be reported by citizens/users. This work introduces a scheduling architecture for automatic management of maintenance jobs. The proposal is able to handle preventive and corrective maintenance jobs looking for available human and instrumental resources during the common time period required by the job to be executed. The scheduler uses intelligent strategies to satisfy constraints of each job in order to get a scheduling according to the criterion of the manager of maintenance. A case study is applied to smart city.

During the past decades, new advances in agriculture automation systems have gained more and more importance and capabilities. In parallel, Internet of Things represents the pervasive presence of a variety of objects (devices, sensors, actuators, mobile phones), which are able to interact with each other, cooperate and create new interfaces (human-machine and machine-machine), to reach common goals. Automation on agriculture systems, or precision agriculture, uses control and communication paradigms to develop new systems: devices, sensors and actuators are interoperable. This technologies (software and hardware) improves the capacity of agricultural installations and agronomists take advantage of new services. This paper presents a research on user-centered design integrated with Internet of Things paradigm. A model-driven development of interactive interfaces that can be adapted and modified by the agronomists in their own agriculturals production is proposed. Additionally, control rules and human-computer interfaces are co-designed by agronomist to adapt the needs of each type of crop and other local and temporal condition (climate, water, energy, nutrients, soil).

This paper presents a study of the impact of dynamic factors on indoor positioning. A positioning system is presented that provides advanced information services based on two subsystems: Wi-Fi and Bluetooth Low Energy (BLE). The first subsystem was intended to position users with not very high levels of accuracy and precision, but not too far from reality, and the second one was intended to position users with greater precision. It is designed for use in stations and terminals of public transportation systems in which the conditions are “hostile” or unfavourable. Experimental results demonstrate that, using different devices for both offline and online phase, RSS differences, Euclidean distance and comparing fingerprints with Weighted k-Nearest Neighbours (WKNN) algorithm, the system is able to position users with reasonable values of accuracy and precision: for Wi-Fi, with only 3 samples, depending on the orientation and compared with 3 neighbours, an average accuracy between 4.15 and 4.58 m and a precision in the range 4–7 m or less 90% of the time were obtained; for BLE, best accuracy results were obtained by comparison with 2 neighbours, giving a position error of 1.59 m and a CDF value of 2.83 m or less 90% of the time.

Adoption of the pervasive and ubiquitous computing paradigm is increasing. One application of this paradigm is Ambient Assistive Living where ICT-based assistive services are deployed in an environment. In this study, a Smart Home in a Box (SHIB) has been devised as an AAL solution. This SHIB relies upon Internet of Things (IoT) devices, such as thermal vision sensors, within an environment to function. Each SHIB deployment consists of IoT devices taken from a common pool of devices. The exact device manifest of each SHIB deployment will change as dictated by end-user requirements. Additionally, this pool will grow and expand when devices are removed or added from previous SHIB deployments. To efficiently manage this pool of devices, and SHIB deployments, a remote management solution is required. Current remote management solutions have a range of deficiencies that render them unsuitable for use with the current SHIB platform. To address these deficiencies, a secure mechanism to manage the IoT devices has been devised. This paper presents this mechanism and evaluates its applicability to SHIB deployments. The evaluation focused on assessing the security of the mechanism and the viability of its Low-Power Wireless Wide-Area Network-based communications. This evaluation showed that the proposed solution was reliable and suitable for the intended deployment scenario. An extended evaluation and system improvements have been identified and proposed within future work.

In this proposal an optimized system designed to help the greatest number of injured people in emergency situations is described, using the shortest possible time and cost. It is composed of a mobile application (assigned to medical staff and helpers), a web service and Near Field Communication wristbands assigned to victims. The mobile application is devoted to providing medical staff with the geolocation of victims as well as with an assistant indicating the best route to follow in order to take care of them based on the severity of their conditions and based on a triage method. Resolution of the routes is solved based on a classical problem, a Travelling Salesman Problem, using a k-partition algorithm to divide the huge number of victims in different clusters. Thus, each doctor has a specific area to assist victims. Besides, doctors can use a functionality of the application to contact their peers through a chat when additional help is needed. An IDentity-Based Signcryption is used for communication confidentiality, authenticity and integrity, both among peers, and between server and medical staff.

Increasingly research interests within the area of pervasive and ubiquitous computing, such as activity recognition, rely upon storage and retrieval of sensor data. Due to the increase in volume, velocity and variation of such sensor data its storage and retrieval has become a big data problem. There are a number of current platforms that are intended to store large amount of sensor data, however, they lack research oriented features. To address these deficiencies this study introduces a research oriented, device agnostic sensor, data platform called SensorCentral. This platform incorporates several research oriented features such as offering annotation interfaces, metric generation, exporting experimental datasets, machine learning services, rule based classification, forwarding live sensor records to other systems and quick sensor configuration. The current main installation of this platform has been in place for over 18 months, has been successfully associated with 6 sensor classes from 13 vendors and currently holds over 500 million records. Future work will involve offering this platform to other researchers and incorporating direct integration with the Open Data Initiative enabling better collaboration with other researchers on an international scale.

Service creation and customization are expensive tasks that require specialists. In many cases these simple services could be implemented by domain experts with no programming abilities. This research work proposes the application of a prosumer model to facilitate access to information and creation of new knowledge to domain experts without experience in the use of semantic AmI platforms. This paper presents a prosumer tool called DataQuest to gather and analyze data from an AmI platform based on ontologies using simple services. An empirical study was carried out in order to demonstrate that, based on the defined prosumer model; a domain expert can create services with acceptable quality for a given domain. The authors performed a case study where 24 people used the prosumer tool to create simple services over a prosumer infrastructure for retrieving and analyzing data from a semantic AmI platform; the aim of this study consists of analyzing the quality of the services built and studying the effort needed to build prosumer services compared with traditional development.

Different methods exist for the estimation of trips in public transportation systems based on automatized fare collection data. Some of the widely adopted estimation strategies are based on the traceability of passenger transfers. They may be effective in large cities, but fall short for trip estimation of single-routes, which are mostly the case in small-to-medium cities. We present a model to estimate the Origin-Destination Matrix, independent of transfers and passenger identification, using only boarding and alighting counts. The model is based on a system of linear diophantine equations, together with a method that computes the general solution and reduces the evaluation space, achieving useful estimations in polynomial time.

Smart transportation systems have now been implemented in many cities. The implementation of these systems requires having a solid infrastructure and specialized devices. However, the implementation of these systems does not consider the infrastructure of other countries. As result, its implementation can be costly. Specifically, in Mexico, Urban Passenger Transport has few transportation units to meet the demand of the population. Also, these do not provide precise information of arrival times of buses. In this research work, we present a software system that combats the inconveniences of public transportation in Mexico, providing information in real time that will allow the passengers to make informed or correct decisions regarding their journey. The information provided to the passengers will be the availability of seats and the arrival times of the buses.

Internet of the things (IoT) is a new scenario that aims to integrate into the global network “things” or devices such as sensors or embedded electronics equipment, sharing information and enabling interaction with them from anywhere. A very important aspect in IoT applications is the communication protocols used by sensors for sending and exchange information. Traditional protocols used so far do not have the scope and ability to manage a growing up computational elements used in modern IoT applications. New communications schemes based on spread spectrum and Long Range (LoRa) coverage, offer new horizons for design and implement low power and long-range sensor networks. In this paper, we describe a new, robust and lightweight communication protocol implemented over the physical layer of LoRa, providing effectiveness and low power consumption suitable for IoT applications. A coverage analysis in both urban and rural environments confirms the success in the selection of this technology and opens new challenges for developing useful applications in several domains as for example smart cities, smart-health, smart factory, remote monitoring, precision agriculture, energy management and remote monitoring and control.

This paper describes the use of a new extension of the Bluetooth connection protocol, called FatBeacon, which faces the problem of obtaining information where no Internet connection is available. Rather than advertising a URL to load a web page, the FatBeacon protocol has the ability to broadcast any basic web contents actually hosted on the device. In particular, FatBeacons are here used to improve the tourist experience in places with no Internet coverage through a new application of the Internet of Things (IoT). Thanks to the fact that the web content is emitted by the own FatBeacon, any smartphone with Bluetooth Low Energy (BLE) can be used to receive touristic information, even in uncovered areas, such as rural or mountain destinations. This work does not only show the applicability of the new FatBeacon protocol, but it also presents a performance comparison of different BLE technologies used for similar touristic applications.

Industry 4.0 refers a new industrial paradigm based on Cyber-Physical Systems principles. In these new, complex and highly interdependent systems, the traditional definition of “cyber-attack” is not enough to represent all the situations may occur. Furthermore, traditional security policies and defense strategies are not designed to be effective in scenarios mixing cyber and physical elements. In this context, this work presents a new idea about what cyber-physical attacks are, and a technological solution to protect and compensate the malicious effects of these attacks in Industry 4.0 systems. The proposal is based on a specific description language (CP-ADL) for cyber-physical attacks, and a mathematical framework allowing a decision making about the most adequate defense strategy. Finally, and experimental validation is provided, showing with our proposal the impact of cyber-physical attacks is highly reduced.

The need to share and reuse information has grown in the new era of Internet of things and ubiquitous computing. Researchers in ontology and schema matching use mapping approaches in order to achieve interoperability between heterogeneous sources. The use of multiple similarity measures that take into account lexical, structural and semantic properties of the concepts is often found in schema matching for the purpose of data integration, sharing and reusing. Mappings identified by automatic or semi-automatic tools can never be certain. In this paper, we present a fuzzy-based approach to combine different similarity measures to deal with scenarios where ambiguity of terms hinder the process of alignment and add uncertainty to the match.

Increasingly more companies support their strategies and value propositions offering their clients some services that require physical mobility and teamwork by their staff. This article proposes a type of Location-based Services system called Geomanagement, which supports the fulfilment of business strategy for a real estate broker agency, based on mobility and collaborative work of their employees. Geomanagement application acts as a valuable support to explore real estate and potential client information in each showing appointment, to share information among realtors and to help new and old realtors in developing teamwork skills while they are in movement. Clients also value the great contribution of Geomanagement regarding the collaboration between realtors. This study demonstrates the effectiveness of Geomanagement by generating value in, an uncommon scenario, as is a real state broker agency. We show that the implementation of Geomanagement increases both collaboration among realtors, and strategic control of managers and heads. Strengthening the link between collaboration and management control systems supported by Geomanagement, turned out to be crucial in business strategy follow-up and monitoring.

The new internet of things paradigm allows for small devices with sensing, processing and communication capabilities to be designed, which enable the development of sensors, embedded devices and other ‘things’ ready to understand the environment. In this paper, a distributed model and an architecture for internet of things paradigm is proposed to perform complex computational tasks and run advanced applications. This novel computing system defines a network design with different levels which combines sensing and processing capabilities based on the Mobile Cloud Computing paradigm.

Internet of Things is a promising paradigm for designing context-aware systems. The information provided can be used for understanding the environment and providing intelligent services to the citizen. There are several studies on the information acquisition of Smart Cities. This work is focused on citizens’ traceability technologies and the data structures involved in representing and querying this information. A location and route data structure is proposed for handling data from different sources and devices, and meet the requirements of citizens’ flow management in Smart Cities.

The quality of service has a main relevance in mass transit systems, being the reliability a key factor for this quality. A system model for the continuous transport data acquisition and computing of these data to improve the quality of service of road mass transit systems is presented in this contribution. This proposal has been conceived to provide services adapted to the needs of the travellers by a continuous monitoring of the transport activity. The data obtained by buses on boarded systems have a special relevance in the proposed model, specially the data provided by the on boarded sensors, such as GPS positioning system. The system model has been applied to analyse the reliability of the operation scheduling of a road mass transit operator, and the results of this test are presented in this paper.

In this work, we provide an analysis of BLE channel-separate fingerprinting using different distance and similarity measures. In a 168 m$$^{2}$$ testbed, 12 beacons with Eddystone and iBeacon protocols set were deployed, taking into account the orientation of users and considering 10 distance/similarity measures. We have observed that there is an orientation that offers the best positioning performance with the combination of iBeacon protocol, channel 38 and Mahalanobis distance. Taking 8 samples in the online phase, accuracy values obtained are in the range 1.28 m–1.88 m, and precision values are within 1.90 m–3.76 m or less, 90% of the time and depending which orientation the observer is facing.

Millions of people are suffering from Diabetes Mellitus today. This amount is expected to increase over the next few years due to multiple factors, not only genetic ones, but also because of our sedentary lifestyle according to the World Health Organization.This work presents a cloud-based system which consists of a web platform, a mobile application and a set of services to facilitate a centralised study of the most relevant parameters involved in diabetes self-care. The system was evaluated by a group of diabetic patients in which 75% of them showed their satisfaction using this system for diabetes self-control. Also, the acceptance level between user and system was studied by means of an usability analysis focused on several evaluation techniques.

In this article, we want to present a platform that allows the integration of different applications for the follow-up of patients with chronic diseases. We are developing two elements: the first one, is a mobile application that allows the capture and processing of vital sign for patient with high blood pressure (hypertension). This application allows to store the patient data obtained, the show historical and trends of the stored measures, a list of foods to consume, alerts and recommendations according to ranges of measures obtained. The second one, is a web application, this application allows the doctor and patients relatives to have updated information of the disease behavior through the measures obtained, food consumed, and others. Through this web application, we can also generate statistics about average measures in a given time, by age, by region, by specific date.

Occupational stress has become a problem that increasingly affects the health of workers. Suffering stress continuously can lead to more serious behavioural disorders such as anxiety or depression. Our work focuses in the proposal of a system which can monitor the stress thanks to subjective information from the workers and objective data from the environment. The information can be consulted by a specialist, and thus, assist the worker in a personalised way and help in the detection and prevention of acute stress cases.

Several evidences showed poor motor coordination in individuals with autism. Thus, in this paper, we present a touch-based hand-eye coordination video game which shows clear difference between children with autism and typically developed children. The game is tested on five children with autism and seven typically developed children. These groups presented totally different patterns from each other; hence they can be distinguished with only two features in the game environment. This promising result encourages us to use this video game for initial screening of hand-eye coordination problem, especially in children with autism, which can be used everywhere without any costs.

A lot of studies linking gait to different pathologies exists. However, few have addressed the automatic classification of such pathologies through computer vision. In this paper, a method to classify different gait pathologies is proposed. Using a smartphone camera, a sagittal view of the subject’s gait is recorded. This record is processed by a computer vision algorithm that extract different gait parameters. These parameters are then used to perform a classification between 5 types of gait: normal, diplegic, hemiplegic, neuropathic and parkinsonian. Using a standard smartphone camera allows to simplify the data capturing step making this method suitable for Ambient Assisted Living. The experiments performed show an accuracy rate of 74% with a hierarchical classifier using Support Vector Machine combining Gait Energy Images and legs angle time series. The accuracy is improved to an 80% by applying data augmentation techniques during test, i.e., obtaining one sample per gait cycle and then combining the results to provide a more robust classification of the entire record.

Epilepsy and falls incur a great social and economic cost globally. Automatically detecting their occurrence would help mitigate the myriad of issues that arise from not receiving assistance after such an event. Despite existing research showing the potential advantages in using the ever-improving sensor technology incorporated within commercially available smartphone and smartwatch devices for human activity recognition, most available solutions for fall and seizure detection are still offered with dedicated hardware, which is often more expensive and less practical. This paper presents a comparison and evaluation of algorithms for detecting convulsions and falls, separately and combined, using smartphone and smartwatch devices. With a dataset of ordinary activities and simulated falls and convulsions, recorded by 15 test subjects, we found the devices a viable option for the successful detection of the activities, achieving accuracy rates between 89.7% and 98.5% with C4.5 decision tree algorithms.

Distinguishing tip-toe walking from normal walking, in human locomotion patterns, becomes important in applications such as Autism disorder identification. In this paper, we propose a novel approach for tip-toe walking detection based on the walk’s Central Pattern Generator (CPG) parameters. In the proposed approach, the tip-toe walking is modeled by a CPG. Then, the motions of subjects are recorded and skeleton data are extracted using the first-generation Microsoft Kinect sensor. The CPG parameters of these motions are determined and compared to the given patterns to distinguish between tip-toe walking and normal walking. The accuracy of classification is promising while further data will improve the accuracy rate.

In mobile sensing, modern phones allow scientists obtain the information about the participants and their surroundings. At times, obtaining raw sensor data from mobile devices demands their collection through sensing campaigns. Often, processing these data requires data processing components in the mobile device. Some of the data processing components pertain to mathematical functions that can be reused to form other functions. These types of functions are usually crafted at a design stage by the programmers. In this work, we present a novel way in which components can be coupled at the design of the sensing campaign, without the need to redeploy the app. That is, scientists can couple two existing data processing components into a new, high-level component. The results of this paper can facilitate code re-use, code maintenance, and flexibility to a mobile sensing campaign.

In current times, virtual agents, also known as avatars, are being used for many different tasks, from helping guide a user experience, to aiding in the diagnosis and treatment of different health issues, both physical and mental. In past work, we have explored the use of an affective avatar that responds to tactile interaction with a pc–tablet, the response from the users and the possible uses for aiding the diagnosis and treatment of Social Communications Disorders. In this paper, we present our proposal as a work in progress, based on our previous work with an affective avatar, but adding verbal interaction. We will also discuss what future work we have thought of doing with this particular interaction of the model.

Nowadays, job stress is very common and it has a high cost in terms of employees’ health, absenteeism and lower performance. It is so big the impact of this psychological disease that the WHO recognizes it as one of the great epidemics of modern life. This paper presents a job stress predictive model from monitoring employees’ behavior and physiological features. The monitoring was carried out through their job computer and a wrist-worn sensor. The proposed model obtained an accuracy of 94%, a precision of 0.943, a recall and a F-Measure of 0.914. Also, the results obtained of the evaluation of the selected model are presented.

Appointment lead-time is a pivotal parameter in elderly outpatient clinics. In this regard, delayed medical care may represent complications in the elderly population and the development of more severe diseases. However, healthcare managers are not skilled in methods effectively reducing waiting times. Therefore, this paper presents the computer simulation modelling to tackle this problem. In this regard, the real-world system was initially simulated and then, three improvement scenarios were designed and validated operationally and financially. The results evidenced that Scenario 2 was the best choice since it provided a low investment per reduced day and a significant reduction (47.1%) regarding the probability of waiting for more than 8 days per appointment. With this proposal, the quality of medical care in elderly population can be meaningfully increased and decision-making process can be effectively supported.

In this paper we present the results of the evaluation of a low-level prototype implementing the concept of Personal Spaces for Multisensory Stimulation (PS4MS). A wizard of Oz evaluation was conducted in terms of the identification of the scenarios presented and the evocation of memories through the stimuli provided to study participants. Evaluation results provide evidence regarding: (i) that participants were able to recognize the scenarios 97.11% of the times and evoke memories 73.26% of the times using the tool, based on the provided olfactory, auditory and visual stimuli; (ii) understanding that although visual stimuli were determinant for scenario recognition, olfactory stimuli were perceived by participants as the most important for memory evocation; and (iii) understanding that the familiarity of the scenario is key both for scenario identification and for the evocation of memories in the participants. This results encourage us to continue working on the construction and evaluation of a high-level version of our PS4MS prototype, which will include tactile (haptic) stimuli in addition to the stimuli already provided.

Unwholesome lifestyles can reduce lifespan by several years or even decades. Therefore, raising awareness and promoting healthier behaviors prove essential to revert this dramatic panorama. Virtual coaching systems are at the forefront of digital solutions to educate people and procure a more effective health self-management. Despite their increasing popularity, virtual coaching systems are still regarded as entertainment applications with an arguable efficacy for changing behaviors, since messages can be perceived to be boring, unpersonalized and can become repetitive over time. In fact, messages tend to be quite general, repetitive and rarely tailored to the specific needs, preferences and conditions of each user. In the light of these limitations, this work aims at help building a new generation of methods for automatically generating user-tailored motivational messages. While the creation of messages is addressed in a previous work, in this paper the authors rather present a method to automatically extract the semantics of motivational messages and to create the ontological representation of these messages. The method uses first natural language processing to perform a linguistic analysis of the message. The extracted information is then mapped to the concepts of the motivational messages ontology. The proposed method could boost the quantity and diversity of messages by automatically mining and parsing existing messages from the internet or other digitised sources, which can be later tailored according to the specific needs and particularities of each user.

Waiting time is a crucial performance metric in A&E departments. In this regard, longer waiting times are related to low patient satisfaction, high mortality rates and more severe physical health complications. To analyze patient flow in these departments, discrete-event simulation (DES) has been used; however, its application has not been extended to evaluate the impact of improvement strategies. Therefore, this paper aims to design and pretest operational strategies for better ED care delivery using DES. First, input data analysis is carried out. Afterward, the DES model is developed and validated to establish whether it is statistically comparable with the real-world. Then, performance indicators of the current system are computed and analyzed. Finally, improvement strategies are proposed and evaluated by simulation modelling and statistical tests. A case study of an A&E department from a district general clinic is presented to validate the proposed framework. In particular, we will validate the effectiveness of introducing a triage system (Scenario 3), a strategy that is not currently adopted by the clinic. Results demonstrate that waiting times could be meaningfully diminished based on the proposed approaches within this paper.

Cardiopulmonary resuscitation is an emergency procedure that can save the life of a person, which combines manual chest compressions on the sternum and insufflations. The depth of the compressions and the cadence at which they are performed are two key factors for the effectiveness of the technique. This study presents a novel support system for cardiopulmonary resuscitation, helping the rescuer to perform the compressions with a correct cadence. To do this, a wearable device adapts to the initial pace of the rescuer, leading him gradually to reach the appropriate cadence.

Missing data is a common problem in clinical datasets due to the large amount of information generated that must be handled, mostly in places where data is entered manually by staff or patients or when sensors or devices for data collection are faulty or damaged. In this work we compare different supervised learning algorithms with an incomplete chronic kidney disease dataset. The aim of this comparison is to select an algorithm to use with missing data from hypertensive patients. In this way, we want to be able to prevent or diagnose chronic kidney disease in hypertensive patients, while we are monitoring their lifestyle through a clinical process improvement based on personalised recommendations using multiple physiological and environmental variables.

Activity recognition is a promising field of research aiming to develop solutions within smart environments to provide relevant solutions on ambient assisted living, among others. The process of activity recognition aims to recognize the actions and goals of one or more person in a environment with a set of sensors are deployed, basing on the sensor data stream that capture a series of observations of actions and environmental conditions. This contributions presents the initial results from a new methodology that considers the use of ontologies to expand the set of feature vector, which is computed by using the sensor data stream, that is used in the process of activity recognition by data-driven approaches. The obtained results indicates that the use of extended feature vectors provided by the use of ontology offers a better accuracy regarding the original feature vectors used in the process of activity recognition with different data-driven approaches.

As the average age of the urban population increases, cities must adapt to improve the quality of life of their citizens. The City4Age H2020 project is working on the early detection of the risks related to Mild Cognitive Impairment and Frailty and on providing meaningful interventions that prevent those risks. As part of the risk detection process we have developed a multilevel conceptual model that describes the user behaviour using actions, activities, intra-activity behaviour and inter-activity behaviour. Using that conceptual model we have created a deep learning architecture based on Long Short-Term Memory Networks that models the inter-activity behaviour. The presented architecture offers a probabilistic model that allows to predict the users next actions and to identify anomalous user behaviours.

The Open Data Initiative (ODI) has previously been proposed as a framework for the collection, annotation, management and sharing of data gathered through research in pervasive health and smart environment systems. It includes the provision of open access protocols for the conduct of experiments and a standard format for the exchange of datasets. In this paper we formalize the structure of the ODI repository through development of an ontology which seeks to unify the representation of ODI objects, experimental protocols, and event logs. An XML-based standards approach to storing event logs, using eXtensible Event Stream (XES) enables data sharing, manipulation and ontology integration. Typical usage scenarios are presented based on published experimental data to validate the concepts of the presented work. Related SPARQL queries are used to illustrate the outputs which can be derived. Based on these initial results, we outline a system architecture for prototyping further ODI related work.

We present a new dataset, called Orange4Home, of activities of daily living of one inhabitant in a smart home environment. We collected data from 236 heterogeneous sensors in a fully integrated instrumented apartment. Data collection spanned 4 consecutive weeks of working days for a total of around 180 h of recording. 20 classes of varied activities were labeled in situ. We report the methodology adopted to establish a representative, challenging dataset, as well as present the apartment and sensors used to collect this data.

In the near future, the world’s population will be characterized by an increasing average age, and consequently, the number of people requiring for a special household assistance will dramatically rise. In this scenario, smart homes will significantly help users to increase their quality of life, while maintaining a great level of autonomy. This paper presents a system for Ambient Assisted Living (AAL) capable of understanding context and user’s behavior by exploiting data gathered by a pervasive sensor network. The knowledge inferred by adopting a Bayesian knowledge extraction approach is exploited to disambiguate the collected observations, making the AAL system able to detect and predict anomalies in user’s behavior or health condition, in order to send appropriate alerts to family members and caregivers. Experimental results performed on a simulated smart home prove the effectiveness of the proposed system.

This paper proposes a goal recognition and planning algorithm, HTN-GRP-PO, to enable intelligent assistant agents to recognize older adults’ goals and reason about desired further steps. It will be used in a larger system aimed to help older adults with cognitive impairments to accomplish activities of daily living independently. The algorithm addresses issues including partial observability due to unreliable or missing sensors, concurrent goals, and incorrectly executed steps. The algorithm has a Hierarchical Task Network basis, which enables it to deal with partially ordered subtasks and alternative plans. We test on simulated cases of different difficulties. The algorithm works very well on simple cases, with accuracy close to 100%. Even for the hardest cases, the performance is acceptable when sensor reliabilities are above 0.95.

It has been estimated, on a global scale, that a limb is lost every 20 s due to diabetes. Effective treatment has, however, been shown to reduce the risk of amputations and foot ulcers. Self-management and the provision of effective foot care in the clinic and at home, is viewed as instrumental in providing appropriate care for people with diabetes. This in turn can reduce the costs associated with treating and managing diabetes and its associated complications. This paper presents a usability evaluation of a smartphone based solution for the management of diabetic foot disease. The solution combines novel thermal imaging with tailored educational content and gamification elements in an attempt to improve self-management of the condition. The solution was evaluated in a workshop setting by 7 participants. Overall, the participants felt the proposed solution would be a very worthwhile endeavor. They rated the usability and utility at an acceptable level (SUS 69.1/100). Notable suggested improvements focused on how educational content was formatted and searched in addition to the physical support provided to facilitate imaging of the feet.

With the growing change to a more aged population and greater strain on our health care services, due to higher costs and demand, there is a growing need to develop solutions to help solve this health care crisis. Generally older people prefer to keep their independence as they age, which gives them a better quality of life. Independent living can be achieved both within care homes and at home, and with the assistance of technology this can be achieved more easily. In this study a Smart Home in a Box (SHIB) solution has been developed to support Ambient Assistive Living (AAL). This paper presents a solution for an interface to interact and be notified by the SHIB platform. A usability evaluation of the developed solution is also presented in the paper. This evaluation asked 8 people consisting of management staff and a Family and Relatives group from a care home, to determine how older people would feel about a solution like this. Overall the participants of the usability evaluation felt the developed solution would benefit older people living at home.

Fog Computing has emerged as a new paradigm where the processing of data and collaborative services are embedded within smart objects, which cooperate between them to reach common goals. In this work, a rule-based Inference Engine based on fuzzy linguistic approach is integrated in the smart devices. The linguistic representation of local and remote sensors is defined by protoforms, which configure the antecedents of the rules in the Inference Engine. A case study where two inhabitants with a wearable device conduct activities in a Smart Lab is presented. Each wearable device infers the daily activities within the wearable devices by means of the rule-based Inference Engine.

Diaries allow users to record personal events and experiences, and are frequently used to collect participant data in user studies. Digital diaries have several benefits over traditional paper-based diaries, reducing respondents’ burden, administrative costs, and improving navigation. However, for elderly users, there are several challenges in the use of a digital diary: they may have cognitive and motor impairments, and fewer digital skills than other populations. We implemented a digital diary called InMyDay, specifically designed for elderly users. The goal of this diary is to promote self-care and self-reflection, by allowing users to register their activities and emotions. Ten elderly users tested the diary for five days, recording entries related to their days and how they felt. All of the participants used the diary every day and after the experiment, nine declared that they would use such an application at least once a week. We found that the diary promoted reflection, that users felt that this allowed them a moment of self-care during their day, and that they felt this was especially important for them as elderly people. Future work will focus on increasing the number of participants and emotions that may be reported and exploring new mechanisms of interaction.

Human activity recognition is a crucial component of applications in the areas of pervasive computing in healthcare. Human activity recognition approaches which adopt a data-driven approach are challenged by handling uncertainty in the data. These uncertainties arise due to sensor unreliability, natural noise and variance introduced by those performing the underlying activities. In this paper we propose an approach to human activity recognition based on Radial Basis Function Neural Networks (RBFNN) trained via a minimization of Localized Generalization Error in an effort to minimize the effects of uncertainty in the data. The proposed approach minimizes the generalization error taking into consideration both the training error and the stochastic sensitivity measure, which subsequently results in an improved generalization capability and improved tolerance to the uncertainty in the data. The approached developed was evaluated using data collected from the IESim smart environment simulation tool. Eleven activities were performed in a simulated environment, with uncertainty in the data stemming from user variance in completing the activities and the sensor placements in the environment. Classification accuracy of 98.86% was achieved demonstrating that the proposed RBFNN approach is robust to minor differences in unseen samples, many of which are caused by data uncertainty, following training which offers good generalization capability.

Formal specifications are used to prove software correctness in a critical system. Ambient Assisted Living (AAL) technologies require unambiguous and precise requirements as they provide critical services for home monitoring. Several AAL technologies have already been designed using scenario-driven approach. But most of them do not precise how to ensure the correctness and conformity of the scenario related to end-user specifications. The multidisciplinary requirements brought by the design team and the assistance to the elderly need a rigorous mechanism for validating and specifying assistance scenarios. In this article, we propose a formal specification approach for scenario construction in the context of AAL technologies. We explain how to instantiate any scenario from the modeled specification, and give some results obtained by using the Alloy language and its validation module. We present a case study applied to nighttime wandering scenario.

Advances in medical science allow people to live longer and more independently than some decades ago. However, this does not directly help older adults improve their mental wellbeing. Several studies show that elderly people usually suffer from some level of social isolation that negatively impacts on their physical and mental conditions. As a way to address such a problem, this paper presents Visitrack, a pervasive and unobtrusive service conceived to monitor the social activity of older adults living at home. Based on sensing data retrieved and processed by the system, it can take several actions. For instance, informing family members and friends about long periods with no social activity at the older adult’s home. The proposed service has been evaluated through a controlled experimental study, obtaining highly accurate results.

Securing the information exchange between the network nodes is very important in wireless sensor and actor networks as actions taken by the actor nodes on the performance is dependent upon the sensed data. This paper proposes an energy-efficient rekeying protocol for WSANs named as energy-efficient hybrid key management protocol that assigns the key management responsibility to the actor nodes. The proposed protocol organizes network nodes into clusters and categorizes nodes into critical and non-critical nodes. It establishes unique pair-wise keys only for pairs of active neighboring sensor nodes after analysis of the network traffic and minimizes the storage overhead. The protocol is distributed and is able to handle compromise of both critical and non-critical nodes. Simulation results show that the technique is highly energy-efficient and enhances the network lifetime.

Sensor nodes in wireless sensor networks are prone to failures because of their sensitiveness in the harsh surroundings. Sometimes, a failure of large scale nodes may occur in the deployed network and it converts connected network into disjoint segments called network partition problem. Therefore, a deployed application demands a continuous fault repairing mechanism to repair the lost connectivity. Deployment of additional relay node in damaged network is one of the best methods to restore the network operation. However, the relay node placement problem is shown to be an NP-hard problem. In this paper, we propose a swarm intelligence based solution to find the best locations of relay node placement in polynomial time. The simulation results show the performance gain of proposed solution over the state-of-the-art solutions.

This work presents a new secure system for deploying opportunistic communications for emergency management after any disaster in which network infrastructures have become collapsed or unavailable. This proposal uses different wireless technologies such as Bluetooth Low Energy, Wi-Fi Direct and LTE Direct to enable device-to-device communications between users through their smartphones. In this way, the system allows the deployment of decentralized communications in fragmented scenarios, what is essential for disaster relief applications. The general procedure of the proposed system is based on the use of chat rooms with possibility of real-time streaming. Two different types of chat rooms are defined: public chat rooms with open access for all registered and authenticated users, and private chat rooms that can be used by different authorized groups of people. The permissions required to use different chat room are established thanks to the use of a decentralized authentication scheme based on public keys and certificate graphs, where different trust levels can be applied. The proposed scheme includes also a decision protocol to choose the most appropriate communication technology in every moment, and a power saving protocol that minimizes the impact of the system on the device’s battery consumption. The first results of field experiments made with a beta application developed for Android smartphones are promising.

Wireless sensor networks (WSNs) is a rapidly growing technology. WSNs comprises of sensor nodes having limited energy as battery powers them. These batteries cannot be changed or recharged as they are operated in a harsh environment. Energy conservation mechanism should be developed. Through study, it is found that clustering is an approach for achieving energy efficiency. In this type of protocols, cluster heads (CH) are chosen among the sensor nodes and then clusters are formed by assigning non-cluster head to the nearest cluster head. Load balancing and the distribution of the cluster heads are the major drawbacks. To deal with the mentioned difficulties, a double optimization based on fuzzy logic approach and harmony search algorithm is proposed in this paper known as fuzzy logic and improved harmony search based clustering (FLIHSBC) algorithm. The proposed algorithm not only balances the energy consumption but also helps in maximizing the network lifetime. Simulation results proved that the proposed algorithm performs better in prolonging the lifetime of the sensor network.

Several initiatives propose the use of opportunistic networks and heterogeneous devices to help overcome the communication and coordination limitations evidenced during first response activities in disaster relief scenarios. These solutions tend to create an Internet of Things ecosystem in which most components are mobile, autonomous and interact with other in a loosely-coupled fashion. Regardless the benefits provided by these infrastructures, the message delivery on them does not consider time constraints. This aspect is particularly relevant in this scenario since the time to conduct the first response activities is quite short, therefore they must be done quickly and coordinately. Trying to help address this challenge, this paper proposes a message propagation model for opportunistic networks that considers heterogeneous devices and guarantees the real-time behavior of the network by bounding the maximum delay for messages transmission. The message propagation is modeled using an analytical approach. Two different scheduling policies are used to analyze the model and their feasibility conditions are proved.

Landslides and large floods are serious natural disasters that every year cause multiple deaths and loss in property around the world. When these events occur in areas like the “favelas” or mountain regions in coastal cities like Rio de Janeiro, the situation becomes critical as buildings and infrastructures are not prepared to withstand them. Search and rescue teams in such disaster areas need to rely on real-time communication, which often cannot be adequately provided by cell or radio networks. In this paper, we argue that flying ad-hoc networks can provide the support needed in these scenarios and propose a new solution towards that goal, termed Flying Witness Units. We make our case by presenting real-time schedulability analysis of message delivery for a disaster scenario.

The ever strengthening symbiosis between software and society calls for increasing attention on the emotions of the users in the engineering of software. In the context of agile software development, this paper proposes a preliminary framework for a user-centered and conceptual model-based process for engineering affective user stories, and illustrates a part of it by an example.

This article presents a novel ontology aiming to connect an Indoor Positioning System (IPS) to Europeana, the European Union digital platform for cultural heritage. The main purpose of this system is to deliver information about Cultural Heritage Objects (CHO) to users navigating in museums, when they approach certain pieces of art. Although different semantic works have been previously published regarding the problem of finding optimal paths with IPS, the novelty of this work is the combination of indoor positioning and a semantic view of cultural objects. This ontology enriches the experience of users and offers a new way of enjoying art. The paper shows the effectiveness of the proposed ontology to connect a widely known database to a wireless positioning system. The potential of the developed method is shown using data obtained from the Royal Museums of Fine Arts of Belgium, one of the most important European art galleries, with more than six thousand master pieces listed in Europeana. Some experiments have been also carried out in the Old masters Museum, one of the constituent museums of the Royal Museums that is dedicated to European painters from the $$15^{th}$$ to the $$18^{th}$$ centuries.

The user interfaces (UIs) of interactive systems become increasingly complex since many heterogeneous and dynamically changing context-of-use parameters regarding user profile, platform, and usage environment have to be supported. Self-adaptive UIs have been promoted as a solution for context variability due to their ability to automatically adapt to the context-of-use at runtime. Context modeling and context management are important prerequisites for supporting self-adaptive UIs, but introduce additional complexity since context information has to be captured using sensors from heterogeneous sources and dynamic context changes have to be monitored to enable UI adaptation at runtime. To overcome the complex and cumbersome task of context management, we present a model-driven approach for developing a flexible context manager supporting self-adaptive UIs. Our approach consists of a new context modeling language, named ContextML, for specification of various context-of-use situations. Based on the specified context model, our approach enables automatic generation of context services for monitoring context-of-use parameters. The benefit of our approach is demonstrated by a case study, where generated context services provide context information and trigger the adaptation of UIs for a university library web application.

Enhancing computers with the facility to perceive and recognise the user feelings and abilities, as well as aspects related to the task, becomes a key element for the creation of Intelligent Human-Computer Interaction. Many studies have focused on predicting users’ cognitive and affective states and other human factors, such as usability and user experience, to achieve high quality interaction. However, there is a need for another approach that will empower computers to perceive more about the task that is being conducted by the users. This paper presents a study that explores user-driven task-based classification, whereby the classification algorithm used features from visual-based input modalities, i.e. facial expression via webcam, and eye gaze behaviour via eye-tracker. Within the experiments presented herein, the dataset employed by the model comprises four different computer-based tasks. Correspondingly, using a Support Vector Machine-based classifier, the average classification accuracy achieved across 42 subjects is 85.52% when utilising facial-based features as an input feature vector, and an average accuracy of 49.65% when using eye gaze-based features. Furthermore, using a combination of both types of features achieved an average classification accuracy of 87.63%.

Older adults can suffer from neurodegenerative diseases, which can make them dependent on family members or trained personnel in geriatric centers. In this work, we aimed at identifying the main information elements needed for decision making in geriatric centers. We present the results of a qualitative study using grounded theory as the analytic lens for a contextual study carried out with workers of a geriatric center. The main contribution of this paper is a set of information elements that are needed for different types of decisions taken at a geriatric center. The results of this work can be used for designing better interfaces for the decision making support in geriatric centers.

Human posture and activity levels are indicators for assessing health and quality of life. Maintaining improper posture for an extended period of time can lead to health issues, e.g. improper alignment of the vertebrae and accelerated degenerative disc. This, in turn, can be the cause of back pain, neurological deterioration, deformity, and cosmetic issues. Some wearable prototypes have been proposed for spine posture monitoring, however, there has not been enough consideration for the users’ experience with these devices, to understand which characteristics are central to acceptance and long-term use. This paper presents a prototype of a low-cost spine posture wearable, along with its preliminary evaluation, which aims both to confirm that the wearable can measure spine posture and to evaluate user experience with this device. The results show that the wearable was comfortable, causing a sensation of security, and that feedback to users would be needed to help improve posture. Further work is required to make sure the device is easy to put on and remove, and discreet enough to be worn in public.

Some of the fears of small animals present in childhood are maintained and cause significant discomfort until the adolescent stage, even causing phobias. A treatment used in the traditional therapy of phobias is the live exposure to the object of fear, however one of the problems with this treatment is patient resistance and eventual therapy abandonment. The use of intelligent environments is an alternative that allows to support the therapy through the virtual, gradual and controlled exposure of the patient to the animal to which s/he is afraid. In this research, we present the design, development and evaluation of a haptic mobile augmented reality system for the treatment of small animals phobia using the TPAD haptic device. The proposed haptic system includes features that allow (i) diagnosing the level of phobia of small animals, (ii) self-adjusting the phobia treatment using support vector machines, and (iii) user progress statistics based on the estimated stress level and time touching the screen. A usability and performance evaluation of the system with 14 teenagers, suggests that the haptic system is perceived as useful and usable, while providing an effective and accessible way to treat the patient and to adjust the therapy challenge level.

Conversing with an older adult who suffers from dementia is challenging. There are frequent disruptions due to loss of attention, errors in interpretation, and low motivation from the person with dementia. We aim to develop a conversational robot that could assist people with dementia and their caregivers address problematic behaviors such as anxiety and depression. As an initial step in this direction we have designed a semiautonomous conversational agent. The agent is capable of participating in simple conversations with the coordination of a human operator to be perceived as an engaging speaker. We present the requirements of the semiautonomous conversational robot, and describe Eva, a prototype implemented to address these requirements. An evaluation conducted with eight caregivers of people with dementia show that the robot was perceived as an autonomous agent that could engage in a long conversation with fluidity.

Even with the availability of several tests to provide clarity in choosing our career path, the decision remains a tough one to undertake. Most of the available tests are either monotonous, resulting in a tedious effort to go through them entirely, or are just plain boring. In this paper, however, we present a new and different approach to career guidance systems. We use Google home as a speech-based interface and Telegram as a text-based interface to generate a conversation between the users and a bot for career guidance. The idea is to provide an easy and friendly interface with an interactive user experience while gathering the required data to provide career guidance. To evaluate the system, we used the University of Costa Rica’s Computer Science and Informatics Department scenario. In this scenario, students must decide between three possible emphases: Software Engineering, Information Technologies, and Computer Science. A usability and user experience evaluation of the system was performed with the participation of 72 freshmen.

The GreenSoul (GS) framework aims to provide a low-cost energy-efficient Information and Communications Technology (ICT) platform which seamlessly augments a traditional public-use building with a set of assets (apps, interactive interfaces, device adaptors, smart meters and a Decision Support Engine), which mediate in the interactions of users with their environments and the energy consuming devices or systems present in them. GreenSoul envisions public use buildings as ecosystems of GreenSoul-ed devices which cooperate with other devices, standard Smart Meters and, very importantly, with eco-educated and eco-aware users to minimize the unnecessary energy consumption. GS architecture is supported by a socio-economic behavioural model, which aids on behaviour understanding to turn energy consuming devices into active pro-sustainability agents that manifest to their surrounding users how well or badly they are being manipulated (energy-wise), offer tips about how to use them more efficiently and even adapt their own functioning to avoid energy waste.

Frosts are one of the main risks faced by farmers during the winter and spring seasons. These events can cause significant damage to diverse types of crops. In Chile, these frost generates significant losses in the agricultural production sector, causing crop losses of an entire year and compromising the income of the following year, especially fruit and wine makers. In this work we developed a prediction model based on historical agrometeorological information able to predict efficiently and up to 12 h earlier the occurrence of a frost event in the Maule Region of Chile. Various algorithms and machine learning methods were evaluated, we found that Random Forest exhibits the best results overall. The results obtained in the frost prediction reach over (90%) of efficiency in most of the evaluated scenarios.

Hourly electrical energy demand predictions improve grid reliability, stability, and minimize costs by maintaining system frequency and optimizing unit commitment and economic dispatch. Weather data, specifically ambient temperature and humidity, is commonly used as a predictor for demand. This paper utilizes the data from a recent demand response and behavioral energy efficiency pilot in Managua, Nicaragua in order to evaluate the relationship between household temperature and demand data, city-wide temperature and demand data, and the potential for utilizing household-level data to predict city-wide demand. Results from this paper indicate that temperature and humidity data can help to inform both household-level and city-wide prediction of electricity demand. Further, the available household level data was found to have a limited relationship with city-wide demand.

Patients suffering from Social Cognition Deficits have difficulties when trying to understand its interlocutor emotional status. In order to contribute to the treatment of this deficit, we have developed a distributed application to offer remote therapies and using the concept of avatars. By using this application, therapist embody avatars that convey their emotions, voices and gestures. Therefore, this application enables patient to recognize the avatars emotions which, in turn, are controlled by the therapist. For this aim, a distributed software has been developed along with different devices such as a Kinect v2 for motion tracking and a facial expression analyzer. Unity has been used for the development of this application to make this type of remote therapy possible.

Schizophrenia, a health and social problem of enormous importance and cost, is a serious mental disorder that has defied researchers for many years. Virtual and augmented reality (VR/AR) is proving to be a powerful experimental tool to study such complex cognitive processes. But there are no VR/AR tools for the systematic treatment of schizophrenia. Avatars do not incorporate adaptation, learning and evolution able to produce natural communication and interaction with humans. To significantly advance the state of art, it is proposed to supplement VR/AR with brain-computer interfaces (BCI) based on obtaining electrical brain signals. The project “Human-avatar symbiosis in cognitive cybertherapies: Proof of concept for auditory verbal hallucinations” assesses whether, through the complementary use of VR/AR and BCI, it is possible to achieve human-machine symbiosis (between a patient who hears voices and “his/her” avatar), which would provide an unforeseen advance. It relies on automatic adaptation, learning and evolution over BDI (beliefs-desires-intentions) models, proper of the intelligent agency paradigm inspired in social behaviour. The proof of concept will show if the proposal is an alternative/complement to conventional therapies of auditory verbal hallucinations and can guide the patient from a negative socio-emotional behaviour toward successful social situations.

Development of algorithms for automatic detection of emotions is essential to improve affective skills of human-computer interfaces. In the literature, a wide variety of linear methodologies have been applied with the aim of defining the brain’s performance under different emotional states. Nevertheless, recent findings have demonstrated the nonlinear and dynamic behavior of the brain. Thus, the use of nonlinear analysis techniques has notably increased, reporting promising results with respect to traditional linear methods. In this sense, this work presents a review of the latest advances in the field, exploring the main nonlinear metrics used for emotion recognition from EEG recordings.

Nowadays, emotional intelligence plays a key role in improving human-machine interaction (HMI). The main objective of HMI is to fill the gap between human emotional states and the reaction of a computer in accordance with this feeling. However, there is a lack of mathematical emotional models to implement affective computing systems into real applications. Consequently, this paper explores the properties of the nonlinear methodology based on Quadratic Sample Entropy (QSE) for the recognition of different emotional subspaces. Precisely, 665 segments of 32-channel electroencephalographic recordings from 32 subjects elicited with different emotional stimuli have been analyzed to validate the proposed model. Results conclude that QSE is a promising feature to be taken into account. Indeed, this metric has reported a discriminant ability around 72% using a support vector machine classifier. This result is comparable with the outcomes reported by other more complex methodologies which use multi-parametric analysis.

In this paper, we propose a mood detection approach which is crucial for human-computer and human-human interaction. In the proposed method, the facial emotional changes are observed through a camera while users use a social network messenger. The advantage of this approach, over the previously proposed approaches, is in its natural setup in which people facially express their feelings, while they read and interact in the social network. This setup eliminates the need for artificial stimulus since social networks are normally filled with different stimulus. The proposed approach is implemented on the Telegram social media messenger. The results show good performance in determining the mood of users. A very promising usage of the proposed approach is in helping human-human relation by providing the mood of one person to another person before an encounter.

In this paper, we present a 3D Kinect-based gaze region estimation module to add gaze pattern information in a driving simulator. Gaze region is estimated using only face orientation cues, similarly to other previous approaches in the literature. An initial user-based calibration stage is included in our approach. The module is able to detect the region, out of 7 in which the driving scene was divided, that a driver is gazing on route every processed frame. 8 people tested the module, which achieved an accuracy of 88.23%. The information provided by the gaze estimation module enriches the driving simulator data and makes it possible a multimodal driving performance analysis.

The proliferation of mobile devices and the ubiquitous nature of cameras today serve to increase the importance of Emotionally Aware Computational Devices. We present a tool to help clinicians and mental health professionals to monitor and assess patients by providing an automated appraisal of a patient’s mood as determined from facial expressions. The App takes video as input from a patient and creates an annotated, configurable record for the clinician as output accessible from mobile devices, Internet or IoT devices.

This paper presents the design, development and preliminary evaluation of an affective wearable device for the transmission of affection gestures between grandparents and grandchildren who are geographically separated. A preliminary evaluation of the device with 33 subjects showed that it is possible to evoke affection gestures. Moreover, such gestures of affection can be tuned in such a way to appear to be real, in order to allow a closer and constant relationship between geographically separated loved ones.

The number of cases reported with Autism Spectrum disorder (ASD), as a developmental disorder, has increased sharply in recent decades. Early diagnosis of ASD in children is essential for proper treatment and intervention. The difficulties in early detection of autism encouraged the authors to design a novel intelligent toy car for autism screening. The toy car is equipped with an accelerometer, which records a subject’s usage behavior in terms of accelerations in three dimensions. A set of features, consisting of forty-four movement characteristics, has been extracted which can be used to discriminate between children with autism and normal children. The intelligent toy car has been tested on 25 children with autism and 25 normal children as the test and control groups respectively. Support Vector Machine (SVM) is used to distinguish between the children with autism and other children. The system has 85% correct classification rate, 93% sensitivity and 76% specificity. The results are the same for boys and girls indicating the possible widespread use of this system among all children.

The huge amount of location tracker data generated by electronic devices makes them an ideal source of information for detecting trends and behaviors in their users’ lives. Learning these patterns is very important for recommender systems or applications targeted at behavior prediction. In this work we show how user location history can be processed in order to extract the most relevant visited locations and to model the user’s profile through a weighted finite automaton, a probabilistic graphical structure that is able to handle locations and temporal context compactly. Our condensed representation gives access to the user’s routines and can play an important role in recommender systems.

The Gaia mission (ESA) is collecting huge amounts of information about the objects that populate our Galaxy and beyond. Such data must be processed and analyzed before being released, and this work is carried out by the Data Processing and Analysis Consortium (DPAC) through several work packages. One of these packages is Outlier Analysis, devoted to the study, by means of unsupervised clustering, of all the objects that cannot be fitted into any of the existent models. An algorithm based on optimized Self-Organized Maps (SOM) is proposed and implemented for taking advantage of distributed computing platforms, such as the MapReduce paradigm for Apache Hadoop and Apache Spark. Finally, the processing times of the sequential implementation of the algorithm is compared to the Hadoop and Spark based ones.

The dissemination of opinions is a very important phenomenon in modern societies because it shapes politics, policies and then guides the evolution of societies in the future. The scheme is very simple: an opinion is transmitted by a leader to other people in his (her) neighborhood. People can be convinced or not and so, in turn, they convey their updated opinion (or their previous one) to their neighbors. This paper presents a computational framework based on a set binary local rules that have the capability to model the transmission of opinion between neighbors. Several neighborhood types are considered, such as 4- and 8-neighbours and knight (chess) neighbors. Different behavioral patterns are analyzed in relation to rule type, neighborhood type, and leader type.