Tomato is one of the most grown and the second most consumed vegetable in the world. Alternaria solani is recognized as the most dangerous tomato pathogen. Currently, the diagnostics of this disease requires the proper symptoms assessment of plant tissue damages. Therefore, it is necessary to propose a fast and reliable method able to assess the degree of plants' damage. Hyperspectral measurements and machine learning algorithms are one of the possibilities to address the problem of finding fast and even more important nondestructive plant diseases detection method. The presented work describes the application of two ensemble learning algorithms: Decision tree and Random Forest adapted for Alternaria solani detection for two varieties of tomatoes cultivated under foil tunnels. The final model was trained on the hyperspectral measurements from 3502500nm spectral range. With a resulting accuracy of the method:0.78 and 0.98 for decision tree and random forest algorithms, respectively.
Language Identification (LI) is an important first step in several speech processing systems. With a growing number of voicebased assistants, speech LI has emerged as a widely researched field. To approach the problem of identifying languages, we can either adopt an implicit approach where only the speech for a language is present or an explicit one where text is available with its corresponding transcript. This paper focuses on an implicit approach due to the absence of transcriptive data. This paper benchmarks existing models and proposes a new attention based model for language identification which uses log-Mel spectrogram images as input. We also present the effectiveness of raw waveforms as features to neural network models for LI tasks. For training and evaluation of models, we classified six languages (English, French, German, Spanish, Russian and Italian) with an accuracy of 95.4% and four languages (English, French, German, Spanish) with an accuracy of 96.3% obtained from the VoxForge dataset. This approach can further be scaled to incorporate more languages.
Machine Learning (ML) has been applied to enable many life-assisting applications, such as abnormality detection in daily routines and automatic emergency request for the solitary elderly. However, in most cases ML algorithms depend on the layout of the target Internet of Things (IoT) sensor network. Hence, to deploy an application across Heterogeneous Sensor Networks (HSNs), i.e. sensor networks with different sensors type or layouts, it is required to repeat the process of data collection and ML algorithm training. In this paper, we introduce a novel framework leveraging deep learning for graphs to enable using the same activity recognition system across HSNs deployed in different smart homes. Using our framework, we were able to transfer activity classifiers trained with activity labels on a source HSN to a target HSN, reaching about 75
Demand for leisure activities has increased due to some reasons such as increasing wealth, ageing populations and changing lifestyles, however, the efficiency of public transport system relies on solid demand levels and well-established mobility patterns and, so, providing quality public transportation is extremely expensive in low, variable and unpredictable demand scenarios, as it is the case of non-routine trips. Better prediction estimations about the trip purpose helps to anticipate the transport demand and consequently improve its planning. This paper addresses the contribution in comparing the traditional approach of considering municipality division to study such trips against a proposed approach based on clustering of dense concentration of services in the urban space. In our case, POIs (Points of Interest) collected from social networks (e.g. Foursquare) represent these services. These trips were associated with the territory using two different approaches: ‘municipalities’ and ‘clusters’ and then related with the likelihood of choosing a POI category (Points-of-Interest). The results obtained for both geographical approaches are then compared considering a multinomial model to check for differences in destination choice. The variables of distance travelled, travel time and whether the trip was made on a weekday or a weekend had a significant contribution in the choice of destination using municipalities approach. Using clusters approach, the results are similar but the accuracy is improved and due to more significant results to more categories of destinations, more conclusions can be drawn. These results lead us to believe that a cluster-based analysis using georeferenced data from social media can contribute significantly better than a territorial-based analysis to the study of non-routine mobility. We also contribute to the knowledge of patterns of this type of travel, a type of trips that is still poorly valued and difficult to study. Nevertheless, it would be worth a more extensive analysis, such as analysing more variables or even during a larger period.
Nowadays, the normalization culture is the usual strategy in enterprises. The formalization has demonstrated its utility for creating efficient, traceable and optimized processes in all the stages of manufacturing. This in being applied to medicine protocols in order to ensure a better quality of service to patients. This protocols can be used to deploy prevention plans on factories improving the current policies by implementing a individualized and holistic approach. Nevertheless, the deployment of these protocols in factory conditions is very complex due to the lack of easy to configure, simple, understandable and efficient systems that could be integrated on the process factory. In this paper, a workflow based solution that enable occupational health professionals is presented. This system enables occupational health professionals to create individualized prevention protocols that allows an easy control of specific workers integrated on the available infrastructure in the factory.
In this paper a study is reported for investigating the effects of a lighting atmosphere on emotional expressiveness and cognitive processing. An experimental lighting atmosphere was created for hospital consultation rooms to better support the shared decision making process of patient and clinician. The lighting atmosphere consists of two phases: (1) indirect, dimmed-warm light (supporting emotional expressiveness) and (2) direct, cold-bright light (supporting cognitive processing). The ambient lighting atmosphere was compared with a standard office lighting atmosphere involving 54 male participants. Participants took part in the experiment in pairs. During the first phase, they watched two emotion inducing film fragments and then discussed these fragments with each other. Under warm-dimmed lighting conditions significantly more emotions were expressed with less negative valence. During the second phase, participants performed two cognitive tasks. No statistical significant effects of lighting condition on both attention and concentration tasks were found. The results showed that participants’ emotions and anxiety level was influenced negatively by the film fragments in both conditions. During the discussion, participants in the intervention condition had significantly more eye contact and showed fewer negative expressions than participants in the control condition. On the cognitive tasks, there was no difference between the conditions, indicating that attention and concentration were not influenced by the intervention.
Climate change and the need for sustainable development have become part of our daily lives. In this context, it is crucial to involve the educational community to the discussion, both students and teachers; by increasing awareness about these issues and the ways school communities can contribute to energy savings, we can kick-start a change towards more sustainable practices in our societies. The Green Awareness in Action (GAIA) H2020 research project implemented an IoT-based approach in several European schools for sustainability awareness and energy efficiency, while at the same time aiming for increasing students' digital skills. By using gamification, competitions and IoT-based educational activities, GAIA engaged directly with teachers and students in order to realize energy-saving activities in their environment. We report here on the use of gamification and competition among schools in this context, and how they helped together with IoT-based lab activities to engage students and educators to participate in the project more actively. We provide details on the implementation of GAIA's intervention in specific school settings to showcase our approach. Our findings, backed up by evaluation data and answers to a survey by 30 educators in Greece and Italy, confirm that the inclusion of competition and gamification aspects can significantly increase students' engagement, especially when having groups/schools competing with each other. Moreover, IoT-based educational activities can supplement existing educational activities in interesting ways, with students evaluating positively the experience and educators reporting increased overall student engagement in their class during the intervention period, and, on average, better class performance compared to previous periods.
Smart Meter infrastructures are emerging systems that measure, collect, and analyze utility data and communicate with the network’s backbone on a fixed schedule. Such infrastructures are a vital part towards real Intelligent Cities. In this article we propose an edgeprocessing oriented Internet of Things architecture for smart meter networks that helps reduce data communication while keeping the system secure, reliable and responsive. We discuss our system architecture based on a real-world water metering deployment of 48 water meters inside a University Campus, using off-the-shelf wM-Bus water meters. We also provide a study of how our solution can face the same problems regardless of the size of the water meter network, scaling up to cities of millions of citizens and measuring points, reducing traffic and data sizes event by 80%.
This study investigates the monitoring of the Quality of Life (QoL) of senior citizens using a teddybear with built in sensors. Utilizing an accelerometer, a motion sensor and a sound sensor, we attempt to employ anomaly detection methods to track the activities of a senior citizen and estimate their QoL. The goal is to keep family members and caregivers updated with the QoL status and let them know if their support is required. In this work, a prototype of the teddybear was kept by a senior citizen for a week while keeping a diary of daily activities. Results show that the collected data can be correlated to daily activities relevant to measuring QoL levels.
Modern planning and management of urban spaces is an essential topic for smart cities and depends on up-to-date and reliable information on urban land use. In the last years, driven by increased availability of georeferenced data from social or embedded sensors and remote sensing (RS) images, various methods become popular for land use analysis. This paper addresses the various methods that are employed in this context, as well as data types needed for these techniques. From our study we concluded that even using the same methods and the same kind of datasets, results depend on spatial configuration of the data, accordingly to the specificity of each region. The work described in this paper is intended to provide relevant contributions to the selection of methods for knowledge discovery for city planning and management.
A lot of activity is being devoted to studying issues related to energy consumption and efficiency in our buildings, and especially on public buildings. In this context, the educational public buildings should be an important part of the equation. At the same time, there is an evident need for open datasets, which should be publicly available for researchers to use. We have implemented a real-world multi-site Internet of Things (IoT) deployment, comprising 25 school buildings across Europe, primarily designed as a foundation for enabling IoT-based energy awareness and sustainability lectures and promoting data-driven energy-saving behaviors. In this work, we present some of the basic aspects to producing datasets from this deployment and discuss its potential uses. We also provide a brief discussion on data derived from a preliminary analysis of thermal comfort-related data produced from this infrastructure.
