9 Named entity disambiguation
This chapter covers
- Understanding the key ideas of Named Entity Disambiguation (NED) combined with Knowledge Graph (KG) technologies
- Building a KG from multiple sources (unstructured contents and ontologies) to boost the activity of stakeholders involved in the domain of substances of human origin (SoHO)
- Performing advanced analysis to support different use cases related to the SoHO domain
Chapter 6 introduced the critical role of NLP techniques in creating KGs from unstructured data automatically. One of the main tasks in that process was Named Entity Recognition (NER). The NER plays a fundamental role in building KGs from unstructured data sources. It allows you to identify mentions of relevant named entities in raw text, assigning predefined categories such as people, organizations, locations, or diseases. However, the NER provides a necessary but insufficient contribution to precisely understanding the text in an application domain.
Imagine you are developing an Intelligent Advisory System (IAS) to support the activities of diverse stakeholders in the healthcare field. One of the critical attributes of such IAS is interactivity, which is the practical ability to exchange information with humans fruitfully via many interactions.
A subset of the essential features to enable this exchange includes:
- the capacity to detect meaningful entities in natural language;
- the ability to retrieve information on these entities from different knowledge sources;