Big data from customers and non-customers through crowdsourcing, citizen science and crowdfunding

Machine learning to support citizen science in urban environmental management

Machine learning (ML) and citizen science (CS) are increasingly prevalent and rapidly evolving approaches to studying and managing environmental challenges. Municipal and other governance actors can benefit from technology advances in ML and public engagement benefits of CS but must also address validity and other quality assurance concerns in their application to particular management contexts. In this article, we take up the pervasive challenge of urban litter to demonstrate how ML can support CS by providing quality assurance in the regulatory context of California's stormwater program. We gave quantitative CS-collected data to five ML models to compare their predictions of a qualitative, site-specific, multiclass "Litter Index" score, an important regulatory metric typically only assessed by trained experts. XGBoost had the best outcome, with scores of 0.98 for accuracy, precision, recall and F-1. These strong results show that ML can provide a reliable complement to CS assessments and increase quality assurance in a regulatory context. To date, ML and CS have each contributed to litter management in novel ways and we find that their integration can provide important synergies with additional applications in other environmental management domains.

Boosting advice and knowledge sharing among healthcare professionals

Purpose

This study aims to investigate the dynamics of knowledge sharing in health care, exploring some of the factors that are more likely to influence the evolution of idea sharing and advice seeking in health care.

Design/methodology/approach

The authors engaged 50 pediatricians representing many subspecialties at a mid-size US children’s hospital using a social network survey to map and measure advice seeking and idea sharing networks. Through the application of Stochastic Actor-Oriented Models, the authors compared the structure of the two networks prior to a leadership program and eight weeks post conclusion.

Findings

The models indicate that health-care professionals carefully and intentionally choose with whom they share ideas and from whom to seek advice. The process is fluid, non-hierarchical and open to changing partners. Significant transitivity effects indicate that the processes of knowledge sharing can be supported by mediation and brokerage.

Originality/value

Hospital administrators can use this method to assess knowledge-sharing dynamics, design and evaluate professional development initiatives and promote new organizational structures that break down communication silos. This work contributes to the literature on knowledge sharing in health care by adopting a social network approach, going beyond the dyadic level and assessing the indirect influence of peers’ relationships on individual networks.