Generalization of Named Entities - Using Linked Data
Master thesis
Permanent lenke
http://hdl.handle.net/11250/2615809Utgivelsesdato
2015Metadata
Vis full innførselSamlinger
Sammendrag
In the climate sciences, it is not feasible for a scientist to read more than a smallfraction of all the papers currently being published.Our group is attempting to help climate scientists by making a tool fordiscovering knowledge in scientific literature. Our approach involves the extractionof causally related events involving either increase or decrease. To facilitatereasoning and search, the relevant parts of the extracted variables need to beidentified and generalized. This is done by a combination of pruning the variables,Named Entity Recognition, and using background knowledge about recognizedentities.In the semantic web, an ever increasing amount of machine readable informationis becoming freely available. With growing support for Linked Data standardsand open licensing, there is now an abundance of structured data waiting to beexplored and utilized.This study seeks to find and utilize knowledge bases in order to generalizespecific types of named entities, and finds that Linked Data resources are wellsuited for this task.