Read Aloud the Text Content

This audio was created by Woord's Text to Speech service by content creators from all around the world.

Text Content or SSML code:

Pandemic Alert is a system trying to detect onset of pandemics by analyzing medical reports and self-reported symptoms, Machine and deep learning techniques are utilized to find spreading patterns of health problems, thereby predicting whether a pandemic is going to occur or its just a local event. Since the dataset should handle growth in the amount of data and users, 500 million data points that are supposedly collected from millions of users were initially generated using a program constructed in Java programming language. The dataset are stored locally in CSV format, each CSV file contains 100 thousand data record. Pandas python-based software library was used to import these data to allow for further manipulation. The imported data are represented in Pandas DataFrames and were split into Training and Testing sets. Then three supervised models were evaluated including Deep Neural Networks, Random Forest and naive bayes on the data to identify the superior classifier. To build a more concrete intuition of how the data looks, here is a scatterplot of the eleven categories after reducing its dimensions using UMAP. The dataset is expandable, which means that in the case of new data entries, to which existing class they belong to is checked and added to it. If they don't belong to any category, a clustering pipeline is implemented to obtain labels to each similar group and add them to the existing data categories. A pandemic is alerted when the disease rate exceeds 15 percent of the data capacity.