Umeå universitets logga

Time-series forecasting is predicting future values based on historical data. Applications include forecasting traffic flows, stock market trends, and energy consumption, which significantly helps to reduce costs and efficiency. However, the complexity inherent in time-series data makes accurate forecasting challenging. This article proposes a novel privacy-enhancing k-anonymous federated learning framework for time-series prediction based on microaggregation. This adaptable framework can be customised based on the client-side processing capabilities. We evaluate the performance of our proposed framework by comparing it with the centralized one using the standard metrics like Mean Absolute Error on three real-world datasets. Moreover, we performed a detailed ablation study by experimenting with different values of k in microaggregation and different client side forecasting models. The results show that our approach gives comparable a good privacyutility tradeoff as compared to the centralized benchmark.

Accurate traffic flow prediction plays an important role in intelligent transportation management and reducing traffic congestion for smart cities. Existing traffic flow prediction techniques using deep learning, mostly LSTMs, have achieved enormous success based on the large traffic flow datasets collected by governments and different organizations. Nevertheless, a lot of these datasets contain sensitive attributes that may relate to users’ private data. Hence, there is a need to develop an accurate traffic flow prediction mechanism that preserves users’ privacy. To address this challenge, we propose a federated learning-based temporal fusion transformer framework for traffic flow prediction which is a distributed machine learning approach where all the model updates are aggregated through an aggregation algorithm rather than sharing and storing the raw data in one centralized location. The proposed framework trains the data locally on client devices using temporal fusion transformers and differential privacy. Experiments show that the proposed framework can guarantee accuracy in predicting traffic flow for both the short and long term.

Federated learning is a recent trend in the field of machine learning for building a collaborative model from distributed data while preserving its privacy. The focus of existing literature is on developing supervised federated learning algorithms requiring labeled data. Whereas only a few solutions have been proposed to identify patterns in distributed unlabeled data using federated clustering methods. However, the issue of measuring the goodness of clusters remains unsolved as existing cluster validity indices cannot be applied in federated learning due to the unavailability of the entire data. To fulfill this research gap, a new metric called FedCI is proposed in the paper for measuring the performance of federated clustering methods, The rationale for FedCI is also discussed and the new metric is validated by comparing it with DB index and Silhouette score. It is found that the behavior of FedCI is consistent with existing metrics. Further, FedCI is applied to the recently proposed FedCLUS a federated clustering method. The FedCLUS algorithm has distinctive characteristics like identification of arbitrarily shaped clusters; the ability to merge, split and discard clusters reported by data owners; communication cost effectiveness. The performance of FedCLUS is compared with centralized DBSCAN using FedCI on various datasets. The results indicate that FedCLUS performs close to the centralized DBSCAN clustering algorithm. The FedCI is expected to guide in finding better clusters in federated settings.