Organizational Entity Recognition
Entity annotation is the act of locating and labeling mentions of named entities within a body of text. Common entity types include people, organizations, places, etc. However, since entities can be organized according to any unique classification system, the number of ways to classify entities is almost endless. Organizational Entity Recognition (OER) takes datasets from within your organization itself, whereas Named Entity Recognition uses well-defined datasets for people’s names, geographic locations etc.
Every organization is different and hence the type and number of datasets will vary. We call this the organization’s “DNA” – what types of information within the organization are important to its history and existence. Let us take the example of a city wanting to analyze school records. Potential datasets (topics) could be Student names, Teacher names, School names, Subject names, etc. OER annotation will locate and label entities within the content using the different datasets. The higher the frequency of located entities, the higher the relevancy of the content to the organization because it is making direct reference to topics that are relevant and important.