Determination of Rate of Medical Waste Generation Using RVM, MARS and MPMR

Determination of Rate of Medical Waste Generation Using RVM, MARS and MPMR

J. Jagan (VIT University, India), Pijush Samui (VIT University, India) and Barnali Dixon (University of South Florida St. Petersburg, USA)
Copyright: © 2020 |Pages: 18
DOI: 10.4018/978-1-7998-1210-4.ch022

Abstract

The prediction of medical waste generation is an important task in hospital waste management. This article uses Relevance Vector Machine (RVM), Multivariate Adaptive Regression Spline (MARS) and Minimax Probability Machine Regression (MPMR) for prediction of rate of medical waste generation. Type of hospital, Capacity and Bed Occupancy has been used as inputs of RVM, MARS and MPMR. RVM is a probabilistic bayesian learning framework. MARS builds flexible model by using piecewise linear regressions. MPMR maximizes the minimum probability that future predicted outputs of the regression model will be within some bound of the true regression function. MARS, RVM and MPMR have been used as regression techniques. The results show that the developed RVM, MPMR and MARS give excellent models for determination of rate of medical waste generation.
Chapter Preview
Top

Details Of Mars

MARS uses basis functions to model the complex relationship between input(x) and output(y). The expression of MARS model is given below.

978-1-7998-1210-4.ch022.m01
(1) where a0 is coefficient, Bm(x) is basis function, am is coefficient of Bm(x) and M is the number of basis functions. In this study, x=[C, B0, x3, x4, x5, x6] and y=[wt]. Table 1 shows the value x3, x4, x5 and x6 for the different hospitals.

Complete Chapter List

Search this Book:
Reset