Serious games are receiving increasing attention in the field of Cultural Heritage (CH) applications. A special field of CH and education is Intangible Cultural Heritage and particularly dance. Machine learning (ML) tools are necessary elements for the success of a serious game platform since they introduce intelligence in processing and analysis of users' interactivity. ML provides intelligent scoring and monitoring capabilities of the user's progress in a serious game platform. In this paper, we introduce a deep learning model for motion primitive classification. The model combines a convolutional processing layer with a bi-directional analysis module. This way, RGB information is efficiently handled by the hierarchies of convolutions, while the bi-directional properties of an LSTM model are retained. The resulting Convolutionally Enhanced Bi-directional LSTM (CEBi-LSTM) architecture is less sensitive to skeleton errors, occurring using low-cost sensors, such as Kinect, while simultaneously handling the high amount of detail when using RGB visual information.

For a long time on-the-job training has been considered as the single point of contact of technical education with the real world job market. Indeed, traditional on-the-job training activities are of great educational value and complement uniquely any classroom-based learning activity. However, it has been observed that several obstacles arise when integrating on-the-job training with traditional learning and that new network-based tools can be used to improve the situation. Following this concept, in this paper we introduce a network-assisted educational structure for business simulation activities which can be considered as an alternative or a complement to on-the-job training activities in technical professional education. Business Simulation Activities are based on the execution of business scenarios in a controlled educational environment with support by network tools. A simulation of the real world business environment is created and virtual companies are established in order to participate in a virtual marketplace using network-based learning support and aiming to provide on-the-job experience without the shortcomings of traditional on-the-job training. An operational structure, network distribution, implementation issues, as well as some experience from a pilot implementation of the concept, are presented in this paper.

The Open Geospatial Consortium (OGC) is committed to developing quality open standards for the global geospatial community, thus enhancing the interoperability of geographic information. In the domain of sensor networks, the Sensor Web Enablement (SWE) initiative has been developed to define the necessary context by introducing modeling standards, like ‘Observation & Measurement’ (O&M) and services to provide interaction like ‘Sensor Observation Service’ (SOS). Land surveying measurements on the other hand comprise a domain where observation information structures and services have not been aligned to the OGC observation model. In this paper, an OGC-compatible, aligned to the ‘Observation and Measurements’ standard, model for land surveying observations has been developed and discussed. Furthermore, a case study instantiates the above model, and an SOS implementation has been developed based on the 52° North SOS platform. Finally, a visualization schema is used to produce ‘Web Map Service (WMS)’ observation maps. Even though there are elements that differentiate this work from classic ‘O&M’ modeling cases, the proposed model and flows are developed in order to provide the benefits of standardizing land surveying measurement data (cost reducing by reusability, higher precision level, data fusion of multiple sources, raw observation spatiotemporal repository access, development of Measurement-Based GIS (MBGIS)) to the geoinformation community.

When there is an absence of disaster prevention measures, natural hazards can lead to disasters. An essential part of disaster risk management is the geospatial modeling of devastating hazards, where data sharing is of paramount importance in the context of early-warning systems. This research points out the usefulness of Spatial Data Infrastructures (SDIs) for disaster risk reduction through a literature review, focusing on the necessity of data unification and disposal. Initially, the principles of SDIs are presented, given the fact that this framework contributes significantly to the fulfilment of specific targets and priorities of the Sendai Framework for Disaster Risk Reduction 2015–2030. Thereafter, the challenges of SDIs are investigated in order to underline the main drawbacks stakeholders in emergency management have to come up against, namely the semantic misalignment that impedes efficient data retrieval, malfunctions in the interoperability of datasets and web services, the non-availability of the data in spite of their existence, and a lack of quality data, while also highlighting the obstacles of real case studies on national NSDIs. Thus, diachronic observations on disasters will not be made, despite these comprising a meaningful dataset in disaster mitigation. Consequently, the harmonization of national SDIs with international schemes, such as the Group on Earth Observations (GEO) and European Union’s space program Copernicus, and the usefulness of Artificial Intelligence (AI) and Machine Learning (ML) for disaster mitigation through the prediction of natural hazards are demonstrated. In this paper, for the purpose of disaster preparedness, real-world implementation barriers that preclude SDIs to be completed or deter their functionality are presented, culminating in the proposed future research directions and topics for the SDIs that need further investigation. SDIs constitute an ongoing collaborative effort intending to offer valuable operational tools for decision-making under the threat of a devastating event. Despite the operational potential of SDIs, the complexity of data standardization and coordination remains a core challenge.