ABSTRACT
Over the past decades, Bayesian-based reasoning, a data and knowledge engineering technology (DKET) approach and reasoning method, has played an important role in knowledge-based systems (KBSs). The well-known concept of this theory is that a parameter source is generated by a random process and experience or prior probability of events of interest; the posterior probability depends not only on the likelihood but also on the history of data. In this chapter, a novel hybrid ontology called multinomial logistic regression (Markov chain Monte Carlo)–C5.0–classification and regression tree (MLR (MCMC)–C5.0–CART) for thalassemia KBS is revealed in terms of theory and comparing the performances of the proposed algorithms. The obtained results show that MLR (MCMC)–C5.0–CART provides satisfactory results with Markov chain error in the range 0.0112–0.2473 for 500,000 iterations. In the future, hybrid intelligent computing such as Bayesian-based reasoning (BBR), artificial neural networks–based reasoning (ANNBR), fuzzy-based reasoning (FBR), evolutionary-based reasoning (EBR), and MLR (MCMC)–C5.0–CART will be constructed for thalassemia KBS.
