Many definitions for tinnitus exist and causes and treatments are plentiful, yet not completely understood. A database of eleven tables, as many as 555 unique patient tuples and numerous time-stamped and other features was obtained for knowledge discovery related to causes and treatments of tinnitus. The paper describes the knowledge discovery and machine learning process and introduces several new temporal features to improve tinnitus evaluation, outcomes analysis, and overall understanding. Through automated analysis it is the goal of the authors to determine unknown yet potentially useful attributes related to tinnitus research.
Published in:
Intelligent Agent Technology, 2007. IAT '07. IEEE/WIC/ACM International Conference on
Date of Conference: 2-5 Nov. 2007