Neutrosophic Clustering Analysis with Data from Cysticercus Tenuicollis DNA Samples

Abstract

Parasitic diseases caused by Cysticercus tenuicollis significantly impact the livestock industry. The quantification of such impacts presents challenges due to the handling of data obtained through modern molecular techniques. This investigation explores the application of neutrosophic cluster analysis to the study of Cysticercus tenuicollis DNA. Neutrosophic clustering and the use of linguistic terms help interpret analyses, employing a five-step neutrosophic clustering algorithm: selection of the clustering algorithm, definition of distance metrics, data preparation, algorithm execution, and examination of results. Samples from 20 sheep exhibiting liver cysts were collected in Cotopaxi Province, Ecuador. The analysis resulted in a neutrosophic cluster where one group displayed low truth and falsity values but a moderate level of indeterminacy. The stability of the clusters was evaluated using the Adjusted Rand Index (ARI) with 100 bootstrapping repetitions, yielding a moderate value of approximately 0.52, which implies that the detected cluster structures are robust and representative of the underlying patterns in the data. The clusters show a clear differentiation based on the initial DNA concentration. Samples with high DNA concentrations tend to group together, suggesting that the quality and quantity of DNA may be critical indicators of the samples' condition and possibly the presence and state of the parasite.