Change detection

[keme686]

So far iRap expects the changesets provided by the source dataset. This will not be the case for most of dataset in LOD cloud. Change detection can be simple or high level. A simple change detector checks non-semantic changes, i.e., what is added and what is deleted. On the other hand, high-level change detection detects semantic level changes, i.e., meaning difference, in addition to simple changes.
Interest-based change detection could also be considered. If we have specific interest in a dataset, we only have to check changes related to our interest. This approach computationally advantageous than computing the changeset for the whole dataset. On the other hand, this approach will not scale if the pub/sub approach is implemented. In pub/sub approach, it is computationally feasible (performs better), if the changeset of the original dataset is computed.