Answers are grounded in this paper and cite their sources.
Analyzes meat quality in broiler chickens, suggesting improved evaluation methods.
Traditional methods are limited by the accuracy and objectivity of broiler meat quality evaluation, due to both the complexity of its multivariate structure and conflict inherent in the uncertainty of data. A Mamdani-type fuzzy logic-based decision support system was designed for meat quality classification based on the dataset of 48 male Ross 308 broiler chickens, slaughtered at the age of 42 days. In this structure, pH24, cooking loss (CL), drip loss (DL) and color (L*) were used as input variables, and quality was classified into low/high/medium groups as output variable. The model performance was thoroughly evaluated by examining the results against those provided by human experts, using Accuracy, Cohen’s kappa, and Weighted kappa statistics to achieve meaningful measures. Results indicated that the Accuracy rate attained by the model was between 81% to 85%, along with a Kappa coefficient of over 0.70, evidencing moderate agreement. Importantly, the weighted kappa analysis demonstrated that most classification errors happened between adjacent quality classes. Overall, the developed fuzzy logic–based model effectively addresses the complexity and vagueness of meat quality prediction, providing reliable classifications that are simple to interpret.
No takes yet. Share an insight, caveat, or question.
Bayyurt et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: