PolRam: Proposed Prediction Framework Based on Pollen Outbreak & a Critical Review

Pollen allergies are a major health challenge in India, and the prevalence of allergies is rising due to climate change. Symptoms of pollen allergies, such as coughing, sneezing, and nasal congestion, can hinder work productivity and overall well-being. This study proposes the future implementation of PolRam, a web application that predicts pollen outbreaks. To forecast pollen outbreaks, PolRam will make use of variables including humidity, moisture, temperature, precipitation, and wind speed. The methodology in this study includes data collection, feature extraction, machine learning model training, evaluation, and the development of a user-friendly web application. In the research, historical pollen data and real-time weather data were collected, and relevant features were extracted for training a regression model. Performance parameters including the F1 score, accuracy, and the average squared error are used to evaluate the model. The trained model will be incorporated into the PolRam web application which will provide an accuracy of 90-95% for predicting personalized pollen forecasts for different regions in India. PolRam aims to enhance allergy management and public health by empowering individuals with timely and accurate predictions of pollen outbreaks.