Destinations, disasters and public relations: Stakeholder engagement in multi-phase disaster management

Research about disasters in tourism has emerged in earnest since the 1990s covering insights for preparedness and response. However, recently, authors have called for more systematic and holistic approaches to tourism disaster management research. To address this gap, this study adopted a public relations perspective to refocus attention to relationships and stakeholder expectations of destination communities across multiple phases of disaster management. The authors used a mixed method approach and developed a battery of disaster management attributes by conducting interviews and analyzing industry documents and the extant literature. These attributes formed part of a survey of tourism businesses. Exploratory factor analysis resulted in a two factor solution: - i) business disaster preparedness, and; - ii) destination disaster response and recovery. Findings also show that participants reported a gap between the importance and destination performance of these attributes. In particular, tourism businesses perceived destinations did not adequately engage in disaster preparedness activities, which had implications for disaster response and recovery.

Elliptical summary randomisation for sensor-based human activity recognition

Many conventional statistical machine learning al- gorithms generalise poorly if distribution bias ex- ists in the datasets. For example, distribution bias arises in the context of domain generalisation, where knowledge acquired from multiple source domains need to be used in a previously unseen target domains. We propose Elliptical Summary Randomisation (ESRand), an efficient domain generalisation approach that comprises of a randomised kernel and elliptical data summarisation. ESRand learns a domain interdependent projection to a la- tent subspace that minimises the existing biases to the data while maintaining the functional relationship between domains. In the latent subspace, ellipsoidal summaries replace the samples to enhance the generalisation by further removing bias and noise in the data. Moreover, the summarisation enables large-scale data processing by significantly reducing the size of the data. Through comprehensive analysis, we show that our subspace-based approach outperforms state-of-the-art results on several activity recognition benchmark datasets, while keeping the computational complexity significantly low.