Title :
The choice of information diffusion function with optimum parameters [hypertension]
Author :
Shang, Hanji ; Wu, Xiyun ; Lu, Yuchu ; Shen, Ye
Author_Institution :
Fudan Univ., Shanghai, China
Abstract :
To deal with a small sample problem which appears in estimating the prevalence rate of hypertension, this paper presents an unlimited information diffusion method with a special kind of function, which is called the "normal structural information diffusion function with parameters". Based on the supposition that the variation of the prevalence rate with age and body mass index (BMI) - which are the two main risk factors - should be smooth and have no great fluctuations, this paper introduces a new method to optimize the parameters according to the "minimal fluctuation principle"
Keywords :
fluctuations; functions; fuzzy systems; haemodynamics; information theory; optimisation; parameter estimation; physiological models; sampling methods; age; body mass index; hypertension prevalence rate estimation; minimal fluctuation principle; normal structural information diffusion function with parameters; optimum parameters; risk factors; small sample problem; smooth variation; unlimited information diffusion method; Data analysis; Diseases; Fluctuations; Hypertension; Mathematics; Optimization methods; Risk analysis; Statistical analysis; Statistical distributions; Statistics;
Conference_Titel :
IFSA World Congress and 20th NAFIPS International Conference, 2001. Joint 9th
Conference_Location :
Vancouver, BC
Print_ISBN :
0-7803-7078-3
DOI :
10.1109/NAFIPS.2001.944769