We store the data on Amazon AWS, in the United States. On-prem option is also available - contact us

If you are using our off-the-shelf webchat widget (like this one) then its UI is mostly fixed. However, we will be happy to modify it for you.

The bot you are using right now is powered by Alterra Answers , answering questions about our products.  It is built on top of our FAQ API. We eat our own dogfood :-).

The bot continues to learn from new chats. However, it’s not fully automatic. The new user queries appear in the LABEL tab of FAQ Editor. A human would have to open the Editor and label them (assign to the right answers). Then just press the TRAIN button, and here it is.

Our software is not available for downloading. All our software resides in the cloud. Alterra Answers is cloud-based SaaS. Our APIs are served from our servers, too. However, on-prem option is also available - contact us.

Natural language understanding ( NLU ) is a subfield of natural language processing (NLP) that deals with machine reading comprehension. The goal of an NLU system is to interpret input text.

Natural language processing ( NLP ) is a subfield of artificial intelligence (AI) concerned with the ability of a computer program to understand human language as it is spoken.

We don’t have a publication on the technical details of our phrase2vec sequence embedding algorithm. In short, it's like word2vec, but for multiple-word questions and commands. Each word in the sentence is mapped to a vector with GloVe , and then a bi-directional LSTM is applied to the sequence. This way, the word sequence is mapped to a 300-dimensional vector

All our products and services are offered as cloud-based SaaS. Everything is served from the cloud. However, on-prem option is also available – please contact us at info@alterra.ai or leave your email here.

Long short term memory ( LSTM ) is a recurrent neural network ( RNN ) architecture that remembers values over arbitrary intervals. Originally invented to encode and predict time series it has been recently successfully used in Natural Language Processing, for encoding sequences of words, i.e. phrases.