October 28nd 2018
Introducing Industry Word2Vec
The Eventn Word2Vec service provides pre-trained word2vec models to perform a lexical analysis on. The services makes it easy to keep a finger on the pulse of the most important information across industries from a single API.
The Eventn platform continuously crawls sites across the web in real-time to build vast industry focused data corpuses ready for analysis.
What Is Word2Vec?
You shall know a word by the company it keeps (Firth, J. R. 1957:11)
The idea behind Word2Vec is relatively straight forward in that it is assumes that the meaning of a word can be inferred by the company it keeps.
There are two types of Word2Vec, Skip-gram and Continuous Bag of Words (CBOW). Word2Vec takes 'raw' text as input and learns a word by predicting its surrounding context (in the case of the skip-gram model) or predict a word given its surrounding context (in the case of the cBoW model) using gradient descent with randomly initialized vectors.
NOTE: Eventn currently supports Continuous Bag of Words (CBOW) with Skip-gram coming soon. Please contact us if you have an immediate Skip-gram requirement.
The Eventn Word2Vec service can be used to surface similar concepts, find unrelated concepts, compute similarity between two words, and more. Having real-time industry corpuses always available means that current trends and topics can be extracted as well as providing product focused integrations for tasks such as SEO, automate customer-facing experiences, as supporting content creatives across agencies and journalism.
The following industries are supported by the Eventn Word2Vec service:
|Industry Name||Model Name|
Looking for an industry that is not currently supported? Feel free to contact us to get your industry of choice added.