Named Entity Recognition can identify individuals, companies, places, organization, cities and
other various type
of entities. API can extract this information from any type of text, web page or social media
network.
Komprehend NER achieves State-of-the-art results on CoNLL 2003 test dataset with Precision 0.9, Recall 0.92 and F1-Score of 0.90. It uses character as well as word level embeddings and therefore, does not reply on POS labels to detect entities making it very useful to detect entities in user-generated content (Try “obama was the third president of america” in Komprehend and Spacy)
Komprehend NER does not lookup dictionaries like Freebase or DBPedia to identify the type of entities and therefore, is very fast to meet demands from various industries.
Komprehend NER can be customized with very few training examples and therefore, it can be adapted to any domain dataset.
Named Entity Recognition API seeks to locate and classify elements in text into definitive
categories such as
names of persons, organizations, locations. It can extract this information in any type of text, be
it a web page,
piece of news or social media content.
The API uses Deep Learning technology to determine representations of character groupings.The text
to be analyzed
is broken into word groups and words are further broken down to character groups and neural network
trains on both
of these granularities. The hypothesis behind the algorithm is that there are two important aspects
which
determine if a word is a proper noun, the first is the composition of a word, what syllables it uses
and what
sounds it comprises of and the second is the adjacent words to the considered word.
Named Entity Recognition can automatically scan documents and extract important entities like people, organizations, and places. Knowing the relevant entities for each article helps to automatically categorize articles in defined hierarchies as well as enables smooth content discovery.
Named Entity Recognition can automatically categorize incoming customer chats into relevant departments based on product names or location names mentioned in them. This can reduce the manual work of support agents and they can take on other complex tasks.
