EntitySearch: the Named Entity Recognition (NER) tool by Openindex

EntitySearch is the Named Entity Recognition (NER) tool by Openindex.

This allows us to extract all kinds of relevant information from any text, such as names, companies, organizations, locations, etc. EntitySearch recognizes more and more entities in any arbitrary text by training with large available data sets.

Input can be text, pdf, a website or any other format. For tests and small quantities the input can be entered via a web form, but we also offer an API service. This allows the technology to be incorporated into automated processes.

Check demo

Openindex inside

Built with advanced search engine technology by Openindex.

Privacy guaranteed

Openindex is all about respecting privacy. All data collected is stored anonymously. No data is sold or provided to third parties.

Extract numerous entities from your text

Automatically finds people, brands, locations, time stamps and much more.

Works on any website and text source

It doesn't matter what technology is behind the websites or data sources, as long as the output is readable by our parser. The input can be any accessible website, but also text files, PDFs, etc.

Support for multiple languages

We currently support forty-eight languages for EntitySearch, but additional language models are being built. It supports about forty different entity types and integrates seamlessly with our parser.

API available

A web API is available through which entitysearch can be queried. This allows the technology to be integrated into automated processes.

Try the online demo below

Enter a URL or text in the text box and see what information is extracted.

EntitySearch uses the following techniques

Apache OpenNLP

Apache OpenNLP is a machine learning based toolkit for the processing of natural language text.

Apache Solr

Solr is the popular, blazing-fast, open source enterprise search platform built on Apache Lucene™. Openindex has its own highly customized and optimized Solr instance that server as the base of our platform.

Perceptrons

A perceptron is a neural network in which the neurons are connected in different layers. A first layer consists of input neurons, where the input signals are applied.

Maximum Entropy

The principle of maximum entropy states that the probability distribution that best reflects the current state of knowledge is the one with the greatest entropy, in the context of accurately stated previous data

PoS tagging

In corpus linguistics, part-of-speech tagging, also known as grammatical tagging, is the marking of a word in a text corresponding to a particular word portion, based on both the definition and the context.

More information

If you are interested in our EntitySearch please contact us to discuss the possibilities.