This work is carried out with support from Lacuna Fund, an initiative co-founded by The Rockefeller Foundation, Google.org, and Canada’s International Development Research Centre. The views expressed herein do not necessarily represent those of Lacuna Fund, its Steering Committee, its funders, or Meridian Institute.
The project
The fact that supervised learning training requires labeled data has become a bottleneck towards developing new models, and curating labeled data is prohibitively expensive especially when domain experts are required in tasks like deep learning (Bach, Ratner, & Ré, 2017).
This direct and explicit labeling of data in supervised learning hinders the direct application of machine learning to many user modeling tasks (Webb, Pazzani, & Billsus, 2001).
Project motivation
The absence of water quality datasets in any known machine learning data repository for the African Catfish (Clarias gariepinus); the cost of generating a labeled African catfish dataset for machine learning without an expertly; and the challenge of time consumption and inaccurate generation of labeled data using manual methods.
Aims of the project
The project aims to generate machine learning datasets for the modeling and prediction of fish growth rate and yield capacity using IoT-based smart water quality monitoring and controller system in freshwater catfish aquaculture and aquaponics pond system.
Project objectives
The specific objectives of this project are to:
i. Set up an experimental fish pond site, consisting of 20 ponds (9 normal ponds and 9 aquaponics pond systems and 2 control ponds) for replications, which enhance a higher level of result accuracy. Each system has a control experiment with which the results of the IoT smart ponds are compared.
ii. Deploy low power and low-cost water quality sensors in a freshwater aquaponics system that monitors pH level, dissolved oxygen, temperature, ammonia, nitrate and nitrite, and turbidity of the water.
iii. Design and Implement a GSM-edge-cloud orchestration for the fish pond controller system
iv. Collect real-time data from a freshwater aquaponics and store in an indigenous cloud data repository made available to scientists.
v. Remotely monitor and manage an aquaponics system, thereby lowering the cost of manpower and production.
vi. Set up a mobile aquaponic pond system for the identification of it's plant-animal biomass relationships.
