RIBEIRO, L. A. P. A.; GARCIA, A. C. B.; DOS SANTOS, P. S. M. Temporal and Causal Relations on Evidence Theory: an Application on Adverse Drug Reactions. The International FLAIRS Conference Proceedings, v. 34, p. 1-4, 2021. doi:10.32473/flairs.v34i1.128546
Temporal and Causal Relations on Evidence Theory: an Application on Adverse Drug Reactions
- Luiz Alberto Pereira Afonso Ribeiro (UNIRIO)
- Ana Cristina Bicharra Garcia (UNIRIO)
- Paulo Sérgio Medeiros dos Santos (UNIRIO)
The use of big data and information fusion in electronic health records (EHR) allowed the identification of adverse drug reactions (ADR) through the integration of heterogeneous sources such as clinical notes (CN), medication prescriptions, and pathological examinations. This heterogeneity of data sources entails the need to address redundancy, conflict, and uncertainty caused by the high dimensionality present in EHR. The use of multisensor information fusion (MSIF) presents an ideal scenario to deal with uncertainty, especially when adding resources of the theory of evidence, also called Dempster–Shafer Theory (DST). In that scenario there is a challenge which is to specify the attribution of be-lief through the mass function, from the datasets, named basic probability assignment (BPA). The objective of the present work is to create a form of BPA generation using analysis of data regarding causal and time relationships between sources, entities and sensors, not only through correlation, but by causal inference.
Keywords: Dempster-Shafer Theory, linear regression, time series, adverse drug reactions, information fusion, machine learning, natural language processing