Using an ontology based big data architecture for predicting pandemic outbreak risk (Preprint)

BACKGROUND Contact tracing is one of the prevention methods for reducing the spread of the pandemic. The existing contact tracing methods mainly use the contact network to collect information of the contacted person, time, location, frequency and duration to predict the risk of pandemic’s spread and its transmission routes. The online and offline questionnaires, mobile phone, wearable wireless sensors, RFID, and GPS are some commonly used methods to collect the information. However, the risk of spread of the pandemic is not only attributed to contacts of people, but also some environmental factors such as cultural behaviors, government policies, public education, technologies usage, etc. OBJECTIVE The objectives of this research are to identify the environmental factors causing the spread of pandemic, and to propose an ontology-based information architecture to collect and filter this information for further analysis. METHODS The research methods include an empirical study and a conceptual research. A review for identifying the environmental factors was done. The EBSCOHost databases (e.g. Medline, ERIC, Library Information Science & Technolog, etc) from 2019 to 2022 were used. The keywords of contact tracing model, spread of pandemic, fear, hygiene measures, government policy, prevention program, pandemic program, information disclosure, economic, COVID-19, Omicron, etc were used to archive the discussion on the spread of pandemic. Content analysis was carried out. The identified environmental factors were used to build the conceptual framework of ontology-based big data information architecture. RESULTS There are 588 archived articles of which 84 articles are relevant to spreading risk topics. The major environmental factors influencing the spread of pandemic include risk perception (n=14), hygiene behaviors (n=5), attitude of pandemic prevention programs/culture (n=12), health education program (n=2), government policies (n=25), technologies (n=18), information disclosure (n=6), and economic strategy (n=2). An ontology-based big data architecture was proposed to capture this information into the contact tracing network. A cluster-based ontology was designed to define and relate the environmental factors and contacts information. Since the spreading of pandemic does not have a stationary pattern, artificial intelligence and network analysis methods were proposed to determine the related environmental factors regarding a person’s contacted network in pandemic risk prediction. CONCLUSIONS The major contribution of this research is that some identified environmental factors have not been considered nor explored before in the spread of pandemic prediction literature. Moreover, the ontology-based big data architecture can integrate environmental data and contact information for pandemic risk prediction and transmission routes and patterns discovery. It helps policy makers to identify the reasons of the spread in the community and its causes-and-consequence relationships that the traditional contact network analysis models did not address, and to plan the pandemic prevention strategy. CLINICALTRIAL NIL