Expert-driven Approach Research Articles

Enhancing hotel and restaurant business efficiency through digital technologies

The significance of this research is deeply rooted in the rapid technological advancements of the modern era, which have the capacity to transform and enhance business operations within the hospitality industry. This study aims to assess the impact of digital technologies on the efficiency of hotel and restaurant businesses, highlighting their key benefits while identifying the challenges and risks associated with their implementation. To achieve these objectives, the study employed a comprehensive analysis of literature, alongside methods of generalization and systematization, to examine the role of digital technologies. The primary objective is to achieve an optimal balance among efficiency, adaptability, and service personalization while addressing the weaknesses and potential threats associated with the implementation of digital technologies in hotel and restaurant operations. The evaluation of their impact was conducted using an expert-driven approach based on SWOT analysis criteria. The Pearson correlation tool, as part of the JASP statistical program, was used to determine the degrees of influence of the criteria. A comprehensive analysis revealed that experts rate the strengths and opportunities of implementing digital technologies in the hospitality sector higher than the associated weaknesses and threats. Moreover, the high degree of interconnection between increasing the throughput of establishments and the personalisation of customer service, high dependence on technology and equipment, and personalisation in customer service, as well as the need for constant technology updates and changes in the internal culture of the organisation and staff skills, indicates the necessity for a systematic approach to the implementation of digital technologies in the activities of hotel and restaurant enterprises. The objective is to achieve an optimal balance between efficiency, adaptability, and service personalisation by taking into account the weaknesses and potential threats that may arise when digital technologies are implemented in the activities of hotel and restaurant enterprises.

ADDRESSING ADOLESCENT ANGER: A STUDY OF THE NOMINAL GROUP TECHNIQUE AS AN EFFECTIVE SOLUTION

This study utilized the Nominal Group Technique (NGT) to identify critical factors influencing adolescent anger management, guided by Albert Bandura's Social Learning Theory (SCT). The primary objective was to pinpoint core components for designing effective anger management strategies for adolescents. Nine experts-comprising trained counseling teachers and mental health professionals-evaluated key factors across three main categories: behavioral, environmental, and personal factors. The key findings revealed that imitation (77.78%) was the most significant behavioral factor, highlighting the crucial role of observational learning in managing anger. The school environment (74.07%) was identified as a vital environmental factor, indicating the significant influence of educational settings on adolescent anger regulation. Additionally, cognitive skills (74.07%) emerged as a key personal factor, underscoring the importance of strengthening problem-solving and critical thinking skills as part of comprehensive anger management strategies. This study provides valuable insights by identifying key intervention factors through a structured expert-driven approach facilitated by NGT. Future research should explore how traditional factors can be incorporated into these strategies and assess the long-term effectiveness of the interventions developed.

Vulnerability-oriented risk identification framework for IoT risk assessment

The proliferation of Internet of Things (IoT) systems across diverse applications has led to a notable increase in connected smart devices. Nevertheless, this surge in connectivity has induced a broad spectrum of vulnerabilities and threats, jeopardizing the security and safety of IoT applications. Security risk assessment methods are commonly employed to analyze risks. However, traditional IT and existing IoT-tailored security assessment methods often fail to fully address key IoT aspects: complex assets intercommunication, dynamic system changes, assets’ potential as attack platforms, safety impacts of security breaches, and assets resource constraints. Such oversights lead to significant risks being overlooked in the IoT ecosystem. In this paper, we propose a novel vulnerability-oriented risk identification framework comprising a four-step process as a core element of IoT security risk assessment, applicable to any IoT system. Our process enhances both traditional and IoT-specific security risk assessment methods by providing tailored approaches that address their crucial oversights for comprehensive IoT risk assessment. We validate our process with a case study of an IoT smart healthcare system using a proposed expert-driven approach. The results confirm that our process effectively identifies critical attack scenarios originating from the lack of proper security measures, mobility, and intercommunication processes of IoT devices in the healthcare system. Furthermore, our analysis reveals potential attacks that exploit the IoT devices as platforms to target the backend and user domains. We demonstrate the feasibility of our process for identifying realistic risks by conducting simulations of two derived attack scenarios using the Contiki Cooja network simulator.

Sequential Modeling and Knowledge Source Integration for Identifying the Structure of a Bayesian Network for Multistage Process Monitoring and Diagnosis

Abstract As a popular applied artificial intelligence tool, Bayesian networks are increasingly being used to model multistage manufacturing processes for fault diagnosis purposes. However, the major issue limiting the practical adoption of Bayesian networks is the difficulty of learning the network structure for large multistage processes. Traditionally, Bayesian network structures are learned either with the help of domain experts or by utilizing data-driven structure learning algorithms through trial and error. Both approaches have their limitations. On the one hand, the expert-driven approach is costly, time-consuming, cumbersome for large networks, and susceptible to errors in assessing probabilities; on the other hand, data-driven approaches suffer from noise, biases, and inadequacy of training data and often fail to capture the physical causal structure of the data. Therefore, in this article, we propose a Bayesian network structure learning approach where popular manufacturing knowledge sources like the failure mode and effect analysis (FMEA) and hierarchical variable ordering are used as structural priors to guide the data-driven structure learning process. In addition, to introduce modularity and flexibility into the learning process, we present a sequential modeling approach for structure learning so that large multistage networks can be learned stage by stage progressively. Furthermore, through simulation studies, we compare and analyze the performance of the knowledge source–based structurally biased networks in the context of multistage process fault diagnosis.

Protective effect of restorative possibilities on cognitive function and mental health in children and adolescents: A scoping review including the role of physical activity

BackgroundThe exposome approach can be a powerful tool for understanding the intertwining of social, physical, and internal influences that shape mental health and cognitive development throughout childhood. To distil conceptual models for subsequent analyses, the EU-funded project Early Environmental quality and Life-course mental health effects (Equal-Life) has conducted literature reviews on potential mediators linking the exposome to these outcomes. We report on a scoping review and a conceptual model of the role of restorative possibilities and physical activity.MethodsPeer-reviewed studies published since the year 2000 in English, on the association between the exposome and mental health/cognition in children/adolescents, and quantitatively investigating restoration/restorative quality as a mediating variable were considered. Database searches were last updated in December 2022. We used an unstructured expert-driven approach to fill in gaps in the reviewed literature.ResultsFive records of three distinct studies were identified, indicating a scarcity of empirical evidence in this newly developing research area. Not only were these studies few in numbers, but also cross-sectional, lending only tentative support to the idea that perceived restorative quality of adolescent's living environment might mediate the association between greenspace and mental health. Physical activity emerged as a mediator leading to better psychological outcomes in restorative environments. We provide a critical discussion of potential caveats when investigating the restoration mechanism in children and propose a hierarchical model including restoration, physical activity, and relational dynamics between children and their environment, including social context, as well as restorative environments other than nature.ConclusionsIt is justified to further explore the role of restoration and physical activity as mediators in the association between early-life exposome and mental health/cognitive development. It is important to consider the child perspective and specific methodological caveats. Given the evolving conceptual definitions/operationalizations, Equal-Life will attempt to fill in a critical gap in the literature.

Towards systematic read-across using Generalised Read-Across (GenRA)

Read-across continues to be a popular data gap filling technique within category and analogue approaches. One of the main issues hindering read-across acceptance is the notion of addressing and reducing uncertainties. Frameworks and formats have been created to help facilitate read-across development, evaluation, and residual uncertainties. However, read-across remains an expert-driven approach with each assessment decided on its own merits with no objective means of evaluating performance or quantifying uncertainties. Here, the underlying motivation of creating an algorithmic approach to read-across, namely the Generalised Read-Across (GenRA) approach, is described. The overall objectives of the approach were to quantify performance and uncertainty. Progress made in quantifying the impact of each similarity context commonly relied upon as part of read-across assessment are discussed. The framework underpinning the approach, the software tools developed to date and how GenRA can be used to make and interpret predictions as part of a screening level hazard assessment decision context are illustrated. Future directions and some of the overarching issues still needed in this field and the extent to which GenRA might facilitate those needs are discussed.

Evaluation of Retrofitting Techniques for Historical Adobe Constructions Using a multi-criteria Decision Analysis: The Case Study of Chile

ABSTRACT The evaluation of retrofitting techniques against multiple (often conflicting) criteria requires an expert-driven approach with in-depth context-based knowledge. This research evaluates the expected performance of retrofitting techniques for historical adobe constructions by integrating literature-data and expert assessment. Consideration is given to performance-based criteria (structural behaviour, material compatibility), values around cultural built heritage (degree of intrusion, retreatability), and local constraints (cost-effectiveness). A multi-criteria decision analysis using MACBETH (Measuring Attractiveness by a Categorical-Based Evaluation TecHnique) is undertaken in this study in order to facilitate the selection process for retrofitting techniques within distinct damage scenarios. The MACBETH model is applied to three historical adobe buildings in Chile using a weighted criteria matrix, while a sensitivity analysis is conducted to determine the stability of interventions. The best-scoring options for each damage pattern are then compared with national guidelines and the current practice. Among the best-ranking solutions obtained from MACBETH models, the use of wooden corner keys makes it possible to effectively counteract flexural cracks or mid-height cracks, damage at intersections of perpendicular walls, corner-damage vertical cracks, and in-plane damage. Wooden tie beams are effective solutions for corner-damage vertical cracks. Geo-mesh also obtains a high score yet is limited by its cost and low reversibility.

The Erdoğan Government's Response to the COVID-19 Pandemic: Performance and Actuality in an Authoritarian Context

AbstractThis article analyses the Erdoğan government's policy response to the coronavirus pandemic. Despite the abundant use of moral antagonisms in his discourse, Erdoğan did not attempt to politicize the pandemic, instead framing it as a global health crisis and presenting the government's public health policies as expert-driven and competent. However, this expert-driven approach was largely a performance. Without a system of democratic oversight or a free media to scrutinize government policies, the Erdoğan government could systematically undercount COVID-19 cases and disregard its own public health restrictions, all the while spreading its narrative of competence and success. Competitive policymaking by opposition-controlled municipalities and criticism from a strong doctors' association had relatively limited ability to discredit the government. The public opinion data we present reveal broad-based support for the government's COVID policies. Our article highlights how authoritarian institutions allow governments to sustain a gap between performance and actuality, granting their leaders greater possibilities to claim policy success.

Applicability of the global TanDEM-X elevation data for terrain modelling of a forested karst area: a case study from Slovak Karst

New interferometric radar data of the TanDEM-X space mission have become recently available as a global digital elevation model providing 0.4 arc second spatial resolution (ca. 12 meters). The TanDEM-X dataset brings new options into geoscientific research across multiple scales. However, the accuracy and suitability of this data have not been evaluated in such an extensive manner as, for example, the widely used SRTM data which resolution is 1 arc second (ca. 30 m). We present a validation of the vertical accuracy of TanDEM-X DEM product and an evaluation of its suitability for landform classification in a forested karst area. The DEM segmentation using geomorphons was used for the automated object-based landform classification. We focused on the identification of dolines for which polygons of dolines mapped by an expert-driven approach were used for validation. Airborne lidar data in the form of DSM and DTM were used as the reference dataset for validation of the TanDEM-X DEM vertical accuracy. The results from the study area show that the vertical RMSE of the TanDEM-X data is 3.42 m with respect to the lidar DSM and 9.64 m in comparison with lidar DTM. The identification of dolines by the geomorphon approach achieved 73 % with TanDEM-X, lower than for the lidar DTM (85 %). The TanDEM-X elevation errors were strongly correlated with the canopy height derived from the lidar data suggesting limited suitability of the TanDEM-X data for mapping fine-scale geomorphological features under forests while there was a good match with the lidar DTM terrain in open areas. URL: https://www.gcass.science.upjs.sk/

Cardio-ML: Detection of malicious clinical programmings aimed at cardiac implantable electronic devices based on machine learning and a missing values resemblance framework

Patients with life-threatening arrhythmias are often treated with cardiac implantable electronic devices (CIEDs), such as pacemakers and implantable cardioverter defibrillators (ICDs). Recent advancements in CIEDs have enabled advanced functionality and connectivity that make such devices (particularly ICDs) vulnerable to cyber-attacks. One of the most dangerous attacks on CIED ecosystems is a data manipulation attack from a compromised programmer device that sends malicious clinical programmings to the CIED. Such attacks can affect the CIED functioning and impact patient's survival and quality of life. In this paper, we propose Cardio-ML – an automated system for the detection of malicious clinical programmings that is based on machine learning algorithms and a novel missing values resemblance framework. Our system is designed to detect new variants of existing attacks and, more importantly, new unknown (zero-day) attacks, aimed at ICDs. We collected 1651 legitimate clinical programmings from 514 patients, over a four-year period, from programmer devices at two medical centers. Our collection also includes 28 core malicious functionalities created by cardiac electrophysiology experts that were later used to create different variants of malicious programmings. Cardio-ML was evaluated extensively in three comprehensive experiments and showed high detection capabilities in most attack scenarios. We achieved perfect classification results for detecting newly created variants of existing core malicious functionalities, with an AUC of 100%; for completely new unknown (zero-day) malicious clinical programmings, an AUC of 80% was obtained, which is 14% better than the state-of-the-art method. We were able to further improve our detection results by identifying the best combination of legitimate and zero-day malicious programmings in the dataset, achieving an AUC of 87%. CIED clinical programmings have many parameters without values for a large number of samples (programmings). To cope with the extreme amount of missing values in our dataset, we developed a novel missing values-based resemblance framework and evaluated it using three dataset-creation approaches: a standard expert-driven approach, our novel data-driven approach, and a combined approach incorporating both approaches. The results showed that our novel framework handles missing values in the data better than the expert-driven approach which yields an empty dataset. In particular, the combined approach showed a 40% improvement in data utilization compared to the data-driven approach.

Workflow for the Generation of Expert-Derived Training and Validation Data: A View to Global Scale Habitat Mapping

Our ability to completely and repeatedly map natural environments at a global scale have increased significantly over the past decade. These advances are from delivery of a range of on-line global satellite image archives and global-scale processing capabilities, along with improved spatial and temporal resolution satellite imagery. The ability to accurately train and validate these global scale-mapping programs from what we will call “reference data sets” is challenging due to a lack of coordinated financial and personnel resourcing, and standardized methods to collate reference datasets at global spatial extents. Here, we present an expert-driven approach for generating training and validation data on a global scale, with the view to mapping the world’s coral reefs. Global reefs were first stratified into approximate biogeographic regions, then per region reference data sets were compiled that include existing point data or maps at various levels of accuracy. These reference data sets were compiled from new field surveys, literature review of published surveys, and from individually sourced contributions from the coral reef monitoring and management agencies. Reference data were overlaid on high spatial resolution satellite image mosaics (3.7 m × 3.7 m pixels; Planet Dove) for each region. Additionally, thirty to forty satellite image tiles; 20 km × 20 km) were selected for which reference data and/or expert knowledge was available and which covered a representative range of habitats. The satellite image tiles were segmented into interpretable groups of pixels which were manually labeled with a mapping category via expert interpretation. The labeled segments were used to generate points to train the mapping models, and to validate or assess accuracy. The workflow for desktop reference data creation that we present expands and up-scales traditional approaches of expert-driven interpretation for both manual habitat mapping and map training/validation. We apply the reference data creation methods in the context of global coral reef mapping, though our approach is broadly applicable to any environment. Transparent processes for training and validation are critical for usability as big data provide more opportunities for managers and scientists to use global mapping products for science and conservation of vulnerable and rapidly changing ecosystems.

Adaptation strategies for coral reef ecosystems in Small Island Developing States: Integrated modelling of local pressures and long-term climate changes

Planning and decision-making vastly benefit from a holistic and systematic understanding of the long-term impacts of climate change and other non-climatic stressors on the health and resilience of coral reef ecosystems, and the efficacy of adaptation strategies and management interventions on mitigating these impacts and maintaining ecosystem condition and associated ecosystem service. This study reports on an approach to modelling coral reef stressors and possible adaptation interventions using the coral reef ecosystem of Port Resolution on Tanna Island, Vanuatu as the case study serving as a microcosm of endangered Pacific Small Island Developing States (SIDS). A novel participatory modelling framework was developed and followed in a stepwise manner to integrate local and long-term climate change pressures by coupling structural analysis and the Bayesian Network (BN) techniques. The BN model was quantified through an advanced consolidated data-induced, evidence-based, and expert-driven approach that incorporated: (1) projections of future climate conditions and changing human activities; (2) the influences of multiple stressors including physical environmental and sociological factors; and (3) spatial variability in the key processes and variables. The first and second phases conceptualised the whole system by providing a graphical presentation of system variables within the Driver-Pressure-State-Impact (DPSI) framework using the structural analysis technique. In the third phase, the BN technique was used to integrate the outcomes of multidisciplinary assessments and analysis with experts’ opinion. The BN modelling phase was completed based on evidence extracted from literature which reported the results of regional and downscaled climate models, GIS-based analysis, parametrised data obtained from the region, and tacit knowledge elicited from experts. The validated model was employed to anticipate the future health and resilience condition of coral reefs under different sets of climatic trajectories and adaptive responses scenarios. The results predict the risks to the health and resilience of the Port Resolution coral reef system from the adverse impacts of climate change and harmful human activities and the possible success of adaptations strategies. A sobering conclusion was that despite the current satisfactory condition of coral reefs in the case study zone, their health and resilience would be severely threatened by 2070 in the absence of implementing adaptation strategies and associated sustainable management interventions.

Heritage waste management

PurposeThe considerable volume of rubble generated by the 2016–2017 earthquakes in central Italy reveals a significant issue in the post-disaster reconstruction phase. Drawing from the experience of Macerata province and the city of Camerino, the purpose of this paper is to explore a possible change of attitude in the reuse of heritage waste materials in the reconstruction process of damaged historical villages and towns in Italy.Design/methodology/approachThis research outlines a comparison between national and regional directives on the rubble management on the one hand, and the praxis on the other, carrying out semi-structured interviews with experts who have been involved in the reconstruction process of Macerata province and Camerino, in Marche region.FindingsThe research reveals that the current vision in Italy for the management of disaster waste is still very close to the traditional paradigm that gives heritage waste an intrinsic value, worthy of great efforts for its collection, catalogue and preservation in view of the likely philological restoration of the damaged heritage. The most recent experiences in Camerino show that institutions responsible for the conservation of cultural heritage may accept a possible paradigm shift towards a more innovative and less expert-driven approach to heritage waste materials and their possible upcycling.Originality/valueWithin a critique of the traditional restoration paradigm, this article links disaster waste management to the Sendai Framework for Disaster Risk Reduction, to enhance the long-term sustainability of historical villages and towns in Italy.

Expert-Based Approach on Mapping Ecosystem Services Potential Supply Incircling a Protected Areas by Integrating Matrix Model Assessment

The concept of ecosystem services was introduced into global policy to consolidate both conservation and development to interconnect the discourse between biodiversity and sustainable development for the benefit of human well-being; this encourages conservation and land use planners to achieve both ecological and social goals. Mapping the potential supply of ecosystem services is necessary for continuous monitoring of such services to ensure the sustainable management of natural resources to support decision-making. Therefore, this study presents a multifaceted methodological framework for mapping ecosystem services potential supply surrounding a protected area. We integrate a non-monetary ‘Matrix Model’ assessment with an expert-driven approach to assess the potential supply from a holistic perspective. Experts and stakeholders dealt with Likert scales to weigh multiple land cover supply of ecosystem services. Our study shows that using an integrated approach to map ecosystem services not only minimizes uncertainties but is also important for the evaluation of ecosystem services. The integration process can also be an interactive approach between scientists and stakeholders to improve mutual understanding regarding resource management. The selected services and land cover data can be an effective medium to visually exhibit forest function and the services of a protected area; these can work as scheming tools that can contribute to an effective policy measurement for the sustainable conservation of protected areas.

Democratising planetary boundaries: experts, social values and deliberative risk evaluation in Earth system governance

ABSTRACTRecent debates about the concept of planetary boundaries recall longstanding concerns about whether ecological limits are compatible with ecological democracy. The planetary boundaries framework (originally set out in Rockström et al., 2009a, 2009b) defines values for key Earth-system processes such as climate change and biodiversity that aim to maintain a ‘safe’ distance from thresholds or levels that could endanger human wellbeing. Despite having a significant impact in policy debates, the framework has been criticised as implying an expert-driven approach to governing global environmental risks that lacks democratic legitimacy. Drawing on research on deliberative democracy and the role of science in democratic societies, we argue that planetary boundaries can be interpreted in ways that remain consistent with democratic decision-making. We show how an iterative, dialogical process to formulate planetary boundaries and negotiate ‘planetary targets’ could form the basis for a democratically legitimate division of labour among experts, citizens and policy-makers in evaluating and responding to Earth-system risks. Crucial to this division of evaluative labour is opening up space for deliberative contestation about the value judgments inherent in collective responses to Earth-system risks, while also safeguarding the ability of experts to issue warnings about what they consider to be unacceptable risks.

Expert-Driven and Citational Approaches to Assessing Journal Publications of Brazilian Political Scientists

In this study, we seek to contribute to discussions on how the quality of academic production in the field of political science should be evaluated using Brazil as a case study. We contrast the 'expert-driven approach' that is followed by CAPES, an agency of the Brazilian federal government with the 'citational' approach, which is based on the ranking of journals by mainstream indices of scientific research impact. With data provided by CAPES from 2010 to 2014, we examine to what extent journals that are ranked as having high quality by CAPES also have high impact indexes in the SCImago Journal rank index (SJR), the Hirsch index (h-index) calculated by SCImago, the h5-index and h5-median (based on the h-index period 05 years, calculated by Google Scholar Metrics), and the SNIP indicator (calculated by the CWTS Journal Indicators, included in the Scopus database). Our findings show that there is a positive, but weak correlation between citational criteria and the Qualis evaluation of the same journals. In ordered logistic regressions, we show that a journal's past Qualis scores are the most important factor for explaining its grades in the next evaluation. We show that once a journal's past Qualis score is considered, a journal's citational ranking does not influence its Qualis score with the exception of the SJR in the 2013-4 evaluation. Moreover, a journal's Qualis score is not influenced by the country of publication, language, or social science focus, all else equal.

An Expert-driven Ontology-based Approach to the Collaborative Acquisition of Information

This article presents design decisions for an ontologybased framework supporting an expert-driven approach to the collaborative acquisition, integration and analysis of information. It describes a three-tier organization in which components facilitate establishing an information system to assist domain experts in addressing multi-causal dynamic situations where heterogeneous information sources must be integrated and the interaction of a variety of actors must be supported in large geographical areas with possibly low or intermittent connectivity levels. The approach and envisioned framework are illustrated with a case-study in the design of bikeways in urban zones.

Challenges in User Modeling and Personalization

Personalization has a long history, dating back to the “master-apprentice” approach of individual tutoring that sought to pass on knowledge and skills from one generation to the next. Through user modeling and adaptation, we try to capture the tutor’s human intelligence and turn it into artificial intelligence. Over the last decades, this research has evolved from an expert-driven approach toward a data-driven approach. This evolution comes with an interesting challenge: How can we continue to understand what an automated tutor is doing when the process of collecting and interpreting data about users is fully automated and the adaptation and recommendation decisions are “deduced” from individual users’ behavior as well as the behavior of all users combined? This article discusses the challenges of scrutability, repeatability, and meta-adaptation (aka adaptation of the adaptation), important research issues for the coming years.

Automatic ICD-10 multi-class classification of cause of death from plaintext autopsy reports through expert-driven feature selection.

ObjectivesWidespread implementation of electronic databases has improved the accessibility of plaintext clinical information for supplementary use. Numerous machine learning techniques, such as supervised machine learning approaches or ontology-based approaches, have been employed to obtain useful information from plaintext clinical data. This study proposes an automatic multi-class classification system to predict accident-related causes of death from plaintext autopsy reports through expert-driven feature selection with supervised automatic text classification decision models.MethodsAccident-related autopsy reports were obtained from one of the largest hospital in Kuala Lumpur. These reports belong to nine different accident-related causes of death. Master feature vector was prepared by extracting features from the collected autopsy reports by using unigram with lexical categorization. This master feature vector was used to detect cause of death [according to internal classification of disease version 10 (ICD-10) classification system] through five automated feature selection schemes, proposed expert-driven approach, five subset sizes of features, and five machine learning classifiers. Model performance was evaluated using precisionM, recallM, F-measureM, accuracy, and area under ROC curve. Four baselines were used to compare the results with the proposed system.ResultsRandom forest and J48 decision models parameterized using expert-driven feature selection yielded the highest evaluation measure approaching (85% to 90%) for most metrics by using a feature subset size of 30. The proposed system also showed approximately 14% to 16% improvement in the overall accuracy compared with the existing techniques and four baselines.ConclusionThe proposed system is feasible and practical to use for automatic classification of ICD-10-related cause of death from autopsy reports. The proposed system assists pathologists to accurately and rapidly determine underlying cause of death based on autopsy findings. Furthermore, the proposed expert-driven feature selection approach and the findings are generally applicable to other kinds of plaintext clinical reports.

From Top-Down to “Community-Centric” Approaches to Early Warning Systems: Exploring Pathways to Improve Disaster Risk Reduction Through Community Participation

Natural hazards and their related impacts can have powerful implications for humanity, particularly communities with deep reliance on natural resources. The development of effective early warning systems (EWS) can contribute to reducing natural hazard impacts on communities by improving risk reduction strategies and activities. However, current shortcomings in the conception and applications of EWS undermine risk reduction at the grassroots level. This article explores various pathways to involve local communities in EWS from top-down to more participatory approaches. Based on a literature review and three case studies that outline various levels of participation in EWS in Kenya, Hawai'i, and Sri Lanka, the article suggests a need to review the way EWS are designed and applied, promoting a shift from the traditional expert-driven approach to one that is embedded at the grassroots level and driven by the vulnerable communities. Such a community-centric approach also raises multiple challenges linked to a necessary shift of conception of EWS and highlights the need for more research on pathways for sustainable community engagement.