𝗦𝗣𝗘𝗖𝗜𝗔𝗟 𝗜𝗦𝗦𝗨𝗘 𝗼𝗻 𝗙𝗲𝗱𝗲𝗿𝗮𝘁𝗲𝗱 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗳𝗼𝗿 𝗖𝗼𝗹𝗹𝗮𝗯𝗼𝗿𝗮𝘁𝗶𝘃𝗲 𝗙𝗿𝗮𝘂𝗱 𝗗𝗲𝘁𝗲𝗰𝘁𝗶𝗼𝗻 𝗶𝗻 𝗟𝗮𝗿𝗴𝗲-𝗦𝗰𝗮𝗹𝗲 𝗡𝗲𝘁𝘄𝗼𝗿𝗸𝘀

This special issue in 𝗢𝗽𝗲𝗻 𝗖𝗼𝗺𝗽𝘂𝘁𝗲𝗿 𝗦𝗰𝗶𝗲𝗻𝗰𝗲 (𝗜𝗙 𝟮𝟬𝟮𝟮: 𝟭.𝟱) focuses on Federated learning provides a solution for enabling collaborative model training across distributed nodes without sharing sensitive data. It addresses the privacy concerns associated with centralized approaches by allowing models to be trained regionally on individual nodes without exchanging raw data. This decentralized method uses the network's collective intelligence to detect fraud while protecting the privacy of user data. Federated learning is also appropriate for dynamic fraud detection scenarios because it allows for real-time model modifications. Federated learning involves the cooperative training of a global model on a number of decentralized nodes, like servers or devices, that individually store local data samples. First, a global model is delivered to every node and initialized. Next, using its own data, each node trains the model locally, protecting privacy by using some methods. After that, the modified model parameters are combined to create a global model, which is eventually iteratively improved over several training cycles.
Some of the techniques generally utilized in federated learning for fraud detection such as differential privacy, secure aggregation, model compression, and transfer learning. Frameworks for federated learning algorithms are provided by tools like TensorFlow Federated, PySyft, and FLAML, which also guarantee data privacy and confidentiality. The adoption of advanced encryption techniques to secure model aggregation, the enhancement of federated learning frameworks optimized for resource-constrained devices, and the incorporation of domain-specific knowledge through transfer learning are the recent trends in federated learning. Enhancing scalability to support even larger networks, improving defense against adversarial attacks, and investigating federated learning applications in developing fields like edge computing and the Internet of Things are key to the future of federated learning in fraud detection.
Despite its potential, it presents several drawbacks such as heterogeneity of data distributions across nodes, ensuring model fairness, and addressing privacy concerns in highly regulated industries like finance and healthcare. The massive volumes of data involved and the requirement for a real-time response to new threats make fraud detection in large-scale networks a considerable issue. To overcome these challenges, need to use some techniques including model personalization, adaptive aggregation strategies, and federated learning-specific optimization algorithms. However, federated learning provides a privacy-preserving and scalable approach to collaborative fraud detection in large-scale networks and addresses the dynamic nature of fraudulent activities in real-time.

List of Interested topics include, but are not limited to, the following:
• Analysis of collaborative model training in federated learning.
• Applications of secure aggregation techniques in federated learning to safeguard data privacy in distributed environments.
• Advancements in federated learning for fraud detection.
• Enhancing fraud detection in resource-constrained environments with federated learning algorithms.
• Frameworks and tools for integrating federated learning algorithms in fraud detection.
• Optimization of federated learning frameworks for resource-constrained devices.
• Implementing domain-specific knowledge through transfer learning in federated fraud detection.
• Role of federated learning in fraud detection for IoT devices.
• Utilization of transfer learning approaches for leveraging domain-specific knowledge.
• Strategies to overcome challenges in federated learning for fraud detection.

Authors are requested to submit their full revised papers complying with the general scope of the journal. The submitted papers will undergo the standard peer-review process before they can be accepted. Notification of acceptance will be communicated as we progress with the review process.

===
𝑮𝑼𝑬𝑺𝑻 𝑬𝑫𝑰𝑻𝑶𝑹𝑺

Mahmud Iwan Solihin, UCSI University, Malaysia.
Lin Guoping, Tunghai University, Taiwan.
Slamet Riyadi, Universitas Muhammadiyah Yogyakarta, Indonesia.

𝘼𝘿𝙑𝙄𝙎𝙊𝙍𝙔 𝙀𝘿𝙄𝙏𝙊𝙍
Christos Anagnostopoulos, University of Glasgow, United Kingdom

===
𝑫𝑬𝑨𝑫𝑳𝑰𝑵𝑬

The deadline for submissions is 𝗔𝗨𝗚𝗨𝗦𝗧 𝟭𝟬, 𝟮𝟬𝟮𝟰, but individual papers will be reviewed and published online on an ongoing basis.

===
𝑯𝑶𝑾 𝑻𝑶 𝑺𝑼𝑩𝑴𝑰𝑻
All submissions to the Special Issue must be made electronically via the online submission system Editorial Manager:

𝗵𝘁𝘁𝗽𝘀://𝘄𝘄𝘄.𝗲𝗱𝗶𝘁𝗼𝗿𝗶𝗮𝗹𝗺𝗮𝗻𝗮𝗴𝗲𝗿.𝗰𝗼𝗺/𝗼𝗽𝗲𝗻𝗰𝘀/𝗱𝗲𝗳𝗮𝘂𝗹𝘁𝟮.𝗮𝘀𝗽𝘅

Please choose the article type “𝙎𝙄: 𝙁𝙚𝙙𝙚𝙧𝙖𝙩𝙚𝙙 𝙇𝙚𝙖𝙧𝙣𝙞𝙣𝙜 𝙛𝙤𝙧 𝘾𝙤𝙡𝙡𝙖𝙗𝙤𝙧𝙖𝙩𝙞𝙫𝙚 𝙁𝙧𝙖𝙪𝙙 𝘿𝙚𝙩𝙚𝙘𝙩𝙞𝙤𝙣”.

===
𝑪𝑶𝑵𝑻𝑨𝑪𝑻

𝗼𝗽𝗲𝗻𝗰𝗼𝗺𝗽𝘂𝘁𝗲𝗿𝘀𝗰𝗶𝗲𝗻𝗰𝗲@𝗱𝗲𝗴𝗿𝘂𝘆𝘁𝗲𝗿.𝗰𝗼𝗺

===
𝗙𝗼𝗿 𝗺𝗼𝗿𝗲 𝗶𝗻𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝗼𝗻, 𝗽𝗹𝗲𝗮𝘀𝗲 𝘃𝗶𝘀𝗶𝘁 𝗼𝘂𝗿 𝘄𝗲𝗯𝘀𝗶𝘁𝗲.

https://www.degruyter.com/journal/key/comp/html#overview