Chatbot Architecture
Pavel Surmenok

Details about my NLP project are at In short…

It starts with a table of words with their parts of speech. They are then linked together to create a knowledge base (“KB”) which a bot can use to analyze input and generate output. A temporary data file can be created for each conversation and “new information” (data not already linked into the KB) can be linked into the KB by the bot. If new information is ambiguous or conflicts with information already in the knowledge base, the bot can attempt to resolve the problems with the user before modifying the KB.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.