Top of page

Student/Staff Portal
Global Site Navigation

School of Science

Local Section Navigation
You are here: Main Content

Related Contents

  • See Also
  • LinkedIn profile
    https://www.linkedin.com/in/sharjeel-tahir-5ba742116
  • LinkedIn profile
  • Google Scholar profile
    https://scholar.google.com/citations?user=4828qHIAAAAJ&hl=en&oi=ao
  • Google Scholar profile

Sharjeel Tahir

Overview of thesis

The standard workflow for artificial empathy consists of three stages: 1) Emotion Recognition (ER) using the retrieved features, 2) analysing the perceived emotion or degree of empathy to choose the best course of action, and 3) carrying out a response action. Recent studies that show AE being used with virtual agents or robots often include Deep Learning (DL) techniques. However, there has not been any study that presents an independent approach for evaluating AE, or the degree to which a reaction was empathetic. Most evaluations used in research are based on a Likert scale, with the Davis' IRI and Godspeed Questionnaire being popular examples. Although these questionnaires provide a good estimate of the overall empathic ability of a system, they are time-consuming, expensive, and depend on each individual's perception, raising ambiguity in results. Moreover, the existing corpus for empathy can only classify an action/emotion as empathic or non-empathic. Classification of different degrees of empathy can further improve the ability of a robot to assess and respond in a more natural and human-like manner. The objectives of this research is to improve the implementation of AE by developing state-of-the-art DL techniques, exploiting models, such as transformer networks and deep reinforcement learning models, e.g., DQN. Moreover, it aims to devise an autonomous technique that can evaluate empathy in artificial agents (ECAs or robots), independent of humans. In addition to this, it seeks to improve the effectiveness of empathy recognition by increasing the spectrum of classes (i.e., degree of empathy) in the data that is currently available.

The below are all optional

Qualifications

  • RMT (Research Master with Training, Murdoch University (2020-2021).
  • BSCS (Bachelor of Computer Science, LGU (2014-2018).

Research

Research Interests

  • Computer Vision
  • Robot Vision
  • Emotion Recognition
  • Artificial Empathy

Past Research employment history

  • September 2021 – January 2022: Research Associate, UWA

Other work

  • July 2020 – May 2021: Teaching Assistant (Tutor), Murdoch University

Past Teaching

  • March 2022 – May 2022: Private Tutor, ECU Joondalup

Scholarships and Awards

  • 2022 – Postgraduate Research Scholarship (PhD, ECU)

Supervisors

Dr Syed Afaq Sha, (ECU)

Contact

Sharjeel Tahir
PhD Student
Centre for Artificial Intelligence and Machine Learning
School of Science
Skip to top of page