Personal Cognitive Assistant: Personalisation and Action Scenarios Expansion.

This paper examines the problem of insufficient flexibility in modern cognitive assistants for choosing cars. We believe that the inaccuracy and lack of content information in the synthesised responses negatively affect consumer awareness and purchasing power. The study’s main task is to create a personalised interactive system to respond to the user’s preferences. The authors propose a unique method of supplementing the car buying scenario with deep learning and unsupervised learning technologies to solve the issues, analyse the user’s utterances, and provide a mechanism to accurately select the desired car based on open-domain dialogue interaction. The paper examines the problem of changing user interest and resolves it using a non-linear calculation of the desired responses. We conducted a series of experiments to measure the assistant’s response to temporary changes in user interest. We ensured that the assistant prototype had sufficient flexibility and adjusted its responses to classify user group interests and successfully reclassify them if they changed. We found it logical to interact with the user in open-domain dialogue when there is no certain response to the user’s utterance in the dialogue scenario.