Paper: Graph-based alignment of narratives for automated neurological assessment

ACL ID W12-2401
Title Graph-based alignment of narratives for automated neurological assessment
Venue Workshop on Biomedical Natural Language Processing
Session
Year 2012
Authors

Narrative recall tasks are widely used in neu- ropsychological evaluation protocols in or- der to detect symptoms of disorders such as autism, language impairment, and demen- tia. In this paper, we propose a graph-based method commonly used in information re- trieval to improve word-level alignments in order to align a source narrative to narra- tive retellings elicited in a clinical setting. From these alignments, we automatically ex- tract narrative recall scores which can then be used for diagnostic screening. The signifi- cant reduction in alignment error rate (AER) afforded by the graph-based method results in improved automatic scoring and diagnos- tic classification. The approach described here is general enough to be applied to almost any narrative recall scenario, and the reductio...