Conference Papers Year : 2019

Estimating the Believability of Uncertain Data Inputs in Applications for Alzheimer's Disease Patients

Abstract

Data believability estimation is a crucial issue in many application domains. This is particularly true when handling uncertain input data given by Alzheimer’s disease patients. In this paper, we propose an approach, called DBE_ALZ, to estimate quantitatively the believability of uncertain input data in the context of applications for Alzheimer’s disease patients. In this context, data may be given by Alzheimer’s disease patients or their caregivers. The believability of an input data is estimated based on its reasonableness compared to common-sense standard and personalized rules and the reliability of its authors. This estimation is based on Bayesian networks and Mamdani fuzzy inference systems. We illustrate the usefulness of our approach in the context of the Captain Memo memory prosthesis. Finally, we discuss the encouraging results de-rived from the evaluation of our approach.
No file

Dates and versions

hal-02465512 , version 1 (04-02-2020)

Identifiers

Cite

Fatma Ghorbel, Elisabeth Metais, Fayçal Hamdi. Estimating the Believability of Uncertain Data Inputs in Applications for Alzheimer's Disease Patients. Natural Language Processing and Information Systems, Jun 2019, Salford, United Kingdom. pp.208-219, ⟨10.1007/978-3-030-23281-8_17⟩. ⟨hal-02465512⟩
80 View
0 Download

Altmetric

Share

More