Kaixin is a scholarship recipient of the ARC Centre for Automated Decision-Making and Society.
Each year the Centre offers a select number of PhD scholarships across our nodes to engage and prepare the next generation of researchers to make world-leading contributions in an increasingly engaged and transdisciplinary research environment.
Measuring and quantifying bias, fairness and engagement for information access systems
The Internet is the dominant source where people access information generated by the information access systems. It is important that the system provides equal opportunities and fair information and does not harm any parties. However, the users usually intentionally or unintentionally interact with the system with cognition bias, such as only looking at information that matches the existing belief in mind. This violates the fair distribution of the information, then people often make decisions or choose a side based on the wrong information. The current strategies evaluate the fairness on the system side, e.g. the content distribution on the search results; users can still adjust their behaviour to meet their bias or preferences. In addition, the interaction style of each kind of information access system is different. On the Search Engine, people interact visually. On the Intelligent Assistance, there is usually no visual presentation but voice interaction. The research aims to use a mixed-method to investigate user biased behaviours when interacting with the information access system, and propose a measuring method of user bias, fairness and engagement level. This will help the system make adjustments based on the user’s behaviour to promote healthy information access.
Prof Flora Salim, RMIT University
Prof Falk Scholer, RMIT University
Dr Damiano Spino, RMIT University