In large online volunteer systems, inefficiency and low volunteer retention are existing challenges that compromise the success of online communities particularly given the uncertainty in volunteer participation behavior. A strategy that matches volunteers to a host of fields will alleviate these challenges, yet creating an all-in-one volunteer recommendation system is an unexplored but promising area. We propose VolRec, a session-based recommendation framework for large volunteer networks that employs temporal dynamics to capture uncertainty caused by the changing structure of volunteers’ participation behaviour. To optimize the recommendations, we construct a probabilistic volunteer network graph that denotes co-participation in an activity. We then model individual and inferred neighbours’ preferences as dynamic and context-aware sessions. VolRec can be adapted to recommend volunteers to organizers, tasks, groups and communities, creating a comprehensive and efficient recommendation system. Experiments using Pioneers data, a mobile based app launched in the wake of Covid-19 to mobilize volunteers and record their participation activities demonstrate the efficacy of this approach.

Last edited on February 6^th 2025 03:49AM (Time Zone: EST).
Powered by jemdoc + new design.