Ai.ra is a chatbot (or chatterbot, but we can also say that Ai.ra is a natural language processor), i.e., it is a software application used to conduct an online chat conversation via text. Ai.ra is designed to simulate the way a human (expert) would behave during a round of question and answer (Q&A).
We can classify this type of system (CUS - Conversation Understanding System) into "open-domain systems" and "closed domain systems." A closed domain system, also known as a domain-specific system, focuses on a particular set of topics and has limited responses. On the other hand, an open domain system covers (in principle) any topic. For example, GPT-3 - the NLP model produced by OpenAI - is capable of "talking just about anything."
Ai.ra is a closed domain chatbot, so don't even try to ask her what the square root of 25 is and she won't be able to help you (but her calculator can!). Ai.ra is designed to provide definitions and answer questions on topics related to artificial intelligence (AI), machine learning, AI ethics, and AI safety, and this is "her domain."
Ai.ra has three iterations, the first and second iterations were trained by machine learning (using a Bayesian neural network and an LSTM trained through supervised learning), while the third iteration was created from pre-established rules (n-gram analysis + dictionary search). For its LSTM version, we used the word embeddings made available by the Repositório de Word Embeddings do NILC (Núcleo Interinstitucional de Linguística Computacional).
If you are not satisfied with Ai.ra's performance, if you would like to help us improve the capabilities of our system, or if you want to complain about any kind of message produced by Ai.ra, please contact nicholas@airespucrs.org.
Ai.ra was developed by Nicholas Kluge and Carolina Del Pino. For more information visit this repository. To access Ai.ra, follow this link!
Comentários