Li shenghui 02 05 Large amounts of data must be accessible in order to be properly organised and grouped so that access to the data is made easier clustering algorithms are now doing this for us Semantic data clustering which is necessary for semantic interpretation of the input data has received particular attention in recent years Three modified clustering approaches are applied in this article and their outcomes are assessed On the basis of this a method is initially created that applies a few principles to avoid confusion inside clusters The provided data can be subjected to a rulebased clustering The following method then applies ontologybased semantics to carry out these rules Finally the fundamental method modifies the assumed ontology before applying rules to clusters The outcome demonstrates that the clusters formed from the data included within them were both highly similar and highly dissimilar from one another Additionally there was a notable decrease in the kdistance of these clusters and the correlation was raised Copyright (c) Journal of Engineering Technology and Sciences. All rights reserved 2026