Data Integration and Predictive Analysis System (IPAS)

The goal of IPAS is to design and develop Data Integration and Predictive Analysis System (IPAS) that enables prediction, analysis, and response management of incidents of human infectious diseases.IPAS collects and integrates comprehensive datasets of previous disease incidents and potential influencing factors such as environmental, vector prevalence, demographic, health conditions, other disease prevalence, trade and travel patterns, social media, news feed signal patterns, and Smartphone based sentinel data to facilitate multivariate, predictive analysis.IPAS provides comprehensive analytical support for different stages of epidemiological analysisexploratory, spatial and temporal correlation, hypothesis testing, prediction, and intervention analysis.Innovative machine learning and predictive analytical techniques like support vector machines (SVM) and decision tree based random forests and boosting are used to predict the disease epidemic curves based on detected patterns and influencing factors.Innovative natural language processing (NLP) techniques and KBSIs Intelligence Products Mosaic (IPM) tool are used to extract specific disease incidents, syndromic and zoonotic details from news sources and medical publications and to automate the generation of biosurveillance reports.IPAS also supports the analysis of intervention and prophylaxis response planning in terms of human and medical resource requirements based on the predicted disease epi curve.