Hiruni’s research aims to understand and help eliminate the unwarranted impact of recommender systems on side stakeholders who do not necessarily interact with the system.
Recommender Systems are crucial in helping users pick better options when a wide variety of choices are available. There is a growing concern for fairness in recommender systems as they are implemented in diverse areas, such as music, restaurants, social connections, and jobs. Researchers have explored fairness in recommenders in three main aspects. C-fairness or P-fairness focuses on fairness among different consumer or provider subgroups, like age or gender. C-P-fairness addresses fairness issues between consumers and providers. A relatively new concept called S-fairness deals with the fairness of recommendations on “side stakeholders”, who are not consumers or providers but are affected by suggestions, especially in platforms like Airbnb or Uber. C-fairness, P-fairness, and C-P-fairness are studied based on the actions of consumers and providers within the system. However, there is a notable gap in researching the fairness of recommendations for side stakeholders (S-fairness). Achieving S-fairness is challenging as interactions involving side stakeholders are not visible or reported in the system. Hiruni’s work aims to address this gap by developing a framework for evaluating the fairness of recommendations for multiple stakeholders, with a particular focus on side stakeholders. Her research explores novel methodologies for identifying and assessing the impacts of recommendations on side stakeholders. The initial case study focuses on a micro-mobility sharing service involving various stakeholders like e-scooter riders, other road users and organisations.
Hiruni is a scholarship recipient of the ARC Centre for Automated Decision-Making and Society (ADM+S) and is supervised by Prof. Mark Sanderson, Prof. Flora Salim, Dr. Jeffrey Chan and Dr. Danula Hettiachchi.
References
2024
E-Scooter Dynamics: Unveiling Rider Behaviours and Interactions with Road Users through Multi-Modal Data Analysis
Electric scooters (e-scooters), characterised by their small size and lightweight design, have revolutionised urban commuting experiences. Their adaptability to multiple mobility infrastructures introduces advantages for users, enhancing the efficiency and flexibility of urban transit. However, this versatility also causes potential challenges, including increased interactions and conflicts with other road users. Previous research has primarily focused on historical trip data, leaving a gap in our understanding of real-time e-scooter user behaviours and interactions. To bridge this gap, we propose a novel multi-modal data collection and integrated data analysis methodology, aimed at capturing the dynamic behaviours of e-scooter riders and their interactions with other road users in real-world settings. We present the study setup and the analysis approach we used for an in the wild study with 15 participants, each traversing a pre-determined route equipped with off-the-shelf commercially available devices (e.g., cameras, bike computers) and eye-tracking glasses.
2023
Are footpaths encroached by shared e-scooters? Spatio-temporal Analysis of Micro-mobility Services
Micro-mobility services (e.g., e-bikes, e-scooters) are increasingly popular among urban communities, being a flexible transport option that brings both opportunities and challenges. As a growing mode of transportation, insights gained from micro-mobility usage data are valuable in policy formulation and improving the quality of services. Existing research analyses patterns and features associated with usage distributions in different localities, and focuses on either temporal or spatial aspects. In this paper, we employ a combination of methods that analyse both spatial and temporal characteristics related to e-scooter trips in a more granular level, enabling observations at different time frames and local geographical zones that prior analysis wasn’t able to do. The insights obtained from anonymised, restricted data on shared e-scooter rides show the applicability of the employed method on regulated, privacy preserving micro-mobility trip data. Our results showed population density is the topmost important feature, and it associates with e-scooter usage positively. Population owning motor vehicles is negatively associated with shared e-scooter trips, suggesting a reduction in e-scooter usage among motor vehicle owners. Furthermore, we found that the effect of humidity is more important than precipitation in predicting hourly e-scooter trip count. Buffer analysis showed, nearly 29% trips were stopped, and 27% trips were started on the footpath, revealing higher utilisation of footpaths for parking e-scooters in Melbourne.