Scoring and evaluation of user input in voice user interfaced intelligent interviewer chatbot

Chatbot is a text based conversation agent which can interact human users through some
medium, such as chat interface. Chat interface may be Conversational user interface or voice
user interface. Conversational interface is a user interface design that allows users to chat with
either real humans or bots. Conversational interfaces introduce an opportunity to interact with a
machine using natural language. It takes two forms one is voice assistant that allows you to talk
and other one is chatbots that allow you to type. Instead of using computer specific language,
Conversational user interface enables users to talk to a computer the way they might talk to
another human. This will become more advanced and more efficient. Here I-aBro is an education
and learning chatbot that replace the conversational interface to VUI (voice user interface). VUI
allows the user to interact user with both voice and speech commands. I-aBro generates both
questions and answers and provides evaluation of users input. The main advantage of I-aBro is
that the question is generated via previous response of the user’s input. And here we use Named
Entity Recognition (NER) technology to generate the question, which made more reliable
interview approach. I-aBro, the interview chatbot help to reduce human effort and it is a very
time-consuming process. One of the challenge of I-aBro is to evaluate the interview. Evaluation
is done in two ways. One method is Chat-log is converted to PDF format which sent via mail to
of experts, and another method is evaluation by system itself which is done by Levenshtein
Distance of users’ response.

MSC MI 2016-2018 INT Mr. Pradeep Kumar K