This paper presents the DELTA-R approach that detects and classifies the changes between two versions of a linked dataset. It contributes to the state of the art firstly: by proposing a more granular classification of the resource level changes, and secondly: by automatically selecting the appropriate resource properties to identify the same resources in different versions of a linked dataset with different URIs and similar representation. The paper also presents the DELTA-R change model to represent the changes detected by the DELTA-R approach. This model bridges the gap between resource-centric and triple-centric views of changes in linked datasets. As a result, a single change detection mechanism will be able to support the use cases like interlink maintenance and dataset or replica synchronization. Additionally, the paper describes an experiment conducted to examine the accuracy of the DELTA-R approach in detecting the changes between two versions of a linked dataset. The result indicates that the accuracy of DELTA-R approach outperforms the state of the art approaches by up to 4%. It is demonstrated that the proposed more granular classification of changes helped to identifyup to 1529 additional updated resources compered to X.By means of a case study, we demonstrate the support of DELTA-R approach and change model for an interlink maintenance use case. The result shows that 100% of the broken interlinks were repaired between DBpedia person snapshot 3.7 and Freebase.