Advanced distributed analytics facilitate to make the services smarter for a wider range of data-driven applications in many domains, including agriculture. The key to producing services at such level is timely analysis for deriving insights from data. Centralized data analytic services are becoming infeasible due to limitations in both the ICT infrastructure, timeliness of the information, and data ownership. Distributed Machine Learning (DML) platforms facilitate efficient data analysis and overcome such limitations effectively. Federated Learning (FL) is a DML concept, enables optimizing resource consumption while performing privacy-preserved timely analytics. In order to create such services through FL, there need to be innovative machine learning (ML) models as data complexity as well as application requirements limit the applicability of existing ML models. Therefore, in this paper, we propose a Neural Network (NN)- and Partial Least Square (PLS) regression-based joint FL model (FL-NNPLS). Then its predictive performance is evaluated under sequential- and parallel-updating based FL in a smart farming context, and specifically for milk quality analysis. Smart farming is a fast-growing industrial sector which requires effective analytic platforms to employ sustainable farming practices. The FL-NNPLS approach performs and compares well with a centralized approach and has state-of-the-art performance.
- Data Imbalance
- Decentralized Machine Learning
- Federated Optimization
- MIRS Milk Quality Predictions
- Neural Network