FORECASTED MODELING FOR AIR QUALITY INDEX IN VIETNAMESE TOURIST DESTINATIONS: LEVERAGING DEEP LEARNING APPROACHES

Tóm tắt

Severe air pollution in Vietnam's tourism areas has become a significant economic issue in recent years. While many studies have found a link between population exposure to air pollution and poor health outcomes, short-term exposure to air pollutants in high-pollution zones can result in acute health consequences; thus, poor air quality jeopardizes visitors' health and well-being and threatens the tourism industry's sustainability. As a result, attempts to correctly estimate the air quality index (AQI) are crucial for effective air quality management, a challenge that smart cities must address as they become more developed soon. However, there are some challenges to predicting AQI. First, the results are influenced by various factors that low-cost sensors frequently skip due to the nonlinear and dynamic nature of multivariate air quality time-series data, leaving a gap for enhancements. Second, standard prediction algorithms often use the training data at fixed intervals and require as many available attributes as possible. This work reviews these issues by applying many Recurrent Neural Network (RNN) deep-learning models for the AQI dataset from PAM AIR stations in 10 Vietnamese tourism areas. Then, it compares each model's impact on the data set by leveraging deep learning models for early predictions based on limited but crucial parameters such as particulate matter 2.5 microns (PM2.5) levels, humidity, and temperature. It presents an appealing method for tackling air pollution problems while dataset quality is uncertain. These findings will result in a fast, efficient, cost-effective, and reliable model that would help reduce the impact on health and add to the literature on meteorology and air pollution while giving theoretical insights and practical guidance in assessing AQI and its dangers. It would support the government in adopting efficient pollution control measures to minimize emissions from various sources by making informed decisions proactively to address air pollution challenges before they increase.