Proceedings of the First International Conference on Advanced Scientific Innovation in Science, Engineering and Technology, ICASISET 2020, 16-17 May 2020, Chennai, India

Research Article

Development of a Cognitive Assistant to Learn Concepts for Placement Assistance

Download405 downloads
  • @INPROCEEDINGS{10.4108/eai.16-5-2020.2303970,
        author={R  Dhana Lakshmi and S  Abirami and M  Srivani},
        title={Development of a Cognitive Assistant to Learn Concepts for Placement Assistance},
        proceedings={Proceedings of the First  International Conference on Advanced Scientific Innovation in Science, Engineering and Technology, ICASISET 2020, 16-17 May 2020, Chennai, India},
        publisher={EAI},
        proceedings_a={ICASISET},
        year={2021},
        month={1},
        keywords={cognitive assistants human reasoning skills question answering systems evidence extraction},
        doi={10.4108/eai.16-5-2020.2303970}
    }
    
  • R Dhana Lakshmi
    S Abirami
    M Srivani
    Year: 2021
    Development of a Cognitive Assistant to Learn Concepts for Placement Assistance
    ICASISET
    EAI
    DOI: 10.4108/eai.16-5-2020.2303970
R Dhana Lakshmi1,*, S Abirami1, M Srivani1
  • 1: CEG Anna University
*Contact email: dhanalakshmird03@gmail.com

Abstract

A cognitive assistant helps the humans and enhance their capabilities to solve a large range of complex tasks. The main aim of this paper is to develop a pedagogical cognitive assistant to improve the reasoning abilities and decision making skills. The proposed system has been implemented to assist as a personal agent for students to learn python programming language. The cognitive assistant facilitates natural interactions with the students and it applies human reasoning skills to judge the students ability and train them further. The proposed techniques include Question Answer (QA) analyser, dynamic study plan generation by using assertion graph and accurate answer generation by using evidence extraction and inference generation. A cognitive conversation increases user’s satisfaction and easily engages them with the system and it has achieved significant higher learning gains than a non-interactive online course. The proposed system is evaluated by using the confidence weighted score evaluation metric.