English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Credit risk forecasting plays a crucial role for commercial banks and otherfinancial institutions in granting loans to customers and minimise thepotential loss. However, traditional machine learning methods require thesharing of sensitive client information with an external server to build aglobal model, potentially posing a risk of security threats and privacyleakage. A newly developed privacy-preserving distributed machine learningtechnique known as Federated Learning (FL) allows the training of a globalmodel without the necessity of accessing private local data directly. Thisinvestigation examined the feasibility of federated learning in credit riskassessment and showed the effects of data imbalance on model performance. Twoneural network architectures, Multilayer Perceptron (MLP) and Long Short-TermMemory (LSTM), and one tree ensemble architecture, Extreme Gradient Boosting(XGBoost), were explored across three different datasets under variousscenarios involving different numbers of clients and data distributionconfigurations. We demonstrate that federated models consistently outperformlocal models on non-dominant clients with smaller datasets. This trend isespecially pronounced in highly imbalanced data scenarios, yielding aremarkable average improvement of 17.92% in model performance. However, fordominant clients (clients with more data), federated models may not exhibitsuperior performance, suggesting the need for special incentives for this typeof clients to encourage their participation.

The Effects of Data Imbalance Under a Federated Learning Approach for Credit Risk Forecasting