Master Thesis

Work towards adding automated elements to asynchronous interview systems by solving the problems of interview follow-up question generation and automatic assessment of candidates using various Machine Learning techniques like text generation, text classification and transfer learning techniques for NLP.

• Automatic Follow-up Question Generation – enable a virtual interviewer to generate a follow-up question given a predetermined question and a candidate response using LSTM-based and Transformer models. (View demo)
• Automatic Assessment of Communication Skill – Supervised predictive models to assess the written communication skill scores of interview candidates in various non-conventional interview settings. (View thesis and defense slides)