Citizen science versus professional data collection: Comparison of approaches to mosquito monitoring in Germany

• adding passively collected citizen science data to data generated by scientists improved distribution models of invasive rabbits in Australia
• citizen science can have major benefits for data collection
• receive data from private properties which scientists normally cannot access
• passive (citizen science) monitoring is very efficient in proximity to humans
• active (trapping) monitoring performs notably better in capturing diversity
• passive less suitable for short-term surveillance, spatially limited regions or selected habitats
• involvement of citizen science can play an increasingly important role in the future
• citizen science cannot be the one-fits-all solution (rather it being a suitable supplement)

Data Collection and Quality Challenges for Deep Learning

• 80-90% of machine learning development time is spent on data preparation
• even the beste algorithms cannont perform well without good data
• data quality has a major impact on model accuracy