With a four step approach, DBpedia Spotlight performs named entity extraction, including entity detection and name resolution. It can also be used for named entity recognition , amongst other information extraction tasks.
Empower the user experience reusing, interlinking and making semantic queries among high-quality open datasets, extract meaning from unstructured data.
Identification of surface forms substrings of the original input that may be entity mentions
Selecting a set of surface forms from step 1 along with the DBpedia resources that are candidate meanings for those surface forms
Deciding on the most likely candidate resource for each selected surface form
Adjusting the annotations to task-specific requirements according to user-provided configuration
This work has been partially funded by