An instance associated with the ontological term in the dataset
What happens when a tag is clicked?
Show a list of documents of the tag
&Show another contextual tag cloud with this tag added to context
The initial tag cloud has an empty context, and the tags in the tag cloud reflect the absolute sizes of instances related to each tag. Here are something you can click:
Add/Remove Tags. You can add a tag by clicking it in the tag cloud; Or you can remove a tag from the current context by clicking the "X" after it.
Note that every time you change the context, you are dynamically choosing a subset of instances from the dataset for the next tag cloud.
Switch between Class/Property Views. We show tags in two different views: the class view shows the classes of the the instances (defined by rdf:type);
the property view shows the properties that the instances use in the triples.
Toggle on/off Negation. You can also see the sizes of instances that do not have a specific tag, and even add that negation tag to the context.
Change Inference Rules. We support both subsumption and domain-range inference in RDFS, and you can decide to turn them on or off.
By comparing tag clouds of the same context, under different rules, you might be able to have a better understanding of the dataset.
Search By Keywords. You can search by keywords to find any matching ontological terms you are interested in.
Change Sort Options. When you are browsing tags, we provide two sorting options. By default we sort tags by their local names.
If the system has complete its computation for all the tags, another option will be enabled: i.e., to sort by tag frequencies.
Switch between Tag Cloud/Instance View. For any given set of instances selected by the current context, you can browse by tag cloud to get the summary information,
or switch to instance view to examine raw triples of each instance in the dataset
For more technical details, please read our draft paper (submitted to BTC in ISWC 2012).
If you have any further questions or want to report a bug, please contact us: xiz307 at lehigh.edu