Chapter 5. Identifying people, places, and things
In this chapter
- The basic concepts behind named-entity recognition
- How to use OpenNLP to find named entities
- OpenNLP performance considerations
People, places, and things—nouns—play a crucial role in language, conveying the sentence’s subject and often its object. Due to their importance, it’s often useful when processing text to try to identify nouns and use them in applications. This task, often called either entity identification or named-entity recognition (NER) is often handled by a parser or chunker, as you saw in chapter 2. Though using a parser is nice for understanding a sentence, text applications often will find it more useful to focus on a subset of nouns that identify specific instances of an object such as proper nouns, also often called named entities. Furthermore, fully parsing a sentence is a process-intensive task, whereas finding proper nouns need not be so intensive.