A theoretical study on estimating body fat percentage by using a committee machine based on genetic algorithm optimization approach

Body fat is one of the most important parameters in determining one’s health condition. The current study focuses on estimating body fat values from routine characteristics of adults with the use of artificial intelligent systems. The methodology applied here combines the results of the individual models in a committee machine with intelligent systems (CMIS) for estimating body fat percentage. The artificial neural networks (ANNs), fuzzy logic (FL), and adaptive neuro-fuzzy inference system (ANFIS) were utilized as intelligent experts of the CMIS. The NN models were developed with four different training algorithms (Levenberg–Marquardt (LM), scaled conjugate gradient (SCG), one step secant (OSS) and resilient back-propagation (RP)) and the best one was chosen as the optimal NN expert of the CMIS. The CMIS assigns a weight factor to each individual expert by the simple averaging and weighted averaging methods. A genetic algorithm (GA) optimization technique was used to derive the weighted averaging coefficients. The results indicated that the GA-optimized CMIS performed better than the individual experts acting alone for estimating the body fat percentage from one specific set of input characteristics.