Journal of Web Engineering

Editorial

2025-02-07T06:49:31+01:00

The international workshop on Big Data-Driven Edge Cloud Services (BECS) aims to provide a platform for scholars and practitioners to share their experiences and present ongoing work in developing data-driven AI appli- cations and services within distributed computing environments, commonly referred to as the edge cloud.

The fourth edition of the workshop (BECS 2024)1 was held in conjunc- tion with the 24th International Conference on Web Engineering (ICWE 2024)2, which took place in Tampere, Finland, from 17–20 June 2024. This special issue of the Journal of Web Engineering focuses on enhancing the efficiency of machine learning (ML)-based systems by leveraging the unique features of distributed edge cloud environments. For this issue, we selected papers from BECS 2024 that propose conceptual frameworks to improve the performance and privacy of ML-based systems and explore distributed ML-based solutions for addressing real-world challenges.

Personalized User Models in a Real-world Edge Computing Environment: A Peer-to-peer Federated Learning Framework

2025-02-07T06:22:12+01:00

As the number of IoT devices and the volume of data increase, distributed computing systems have become the primary deployment solution for large-scale Internet of Things (IoT) environments. Federated learning (FL) is a collaborative machine learning framework that allows for model training using data from all participants while protecting their privacy. However, traditional FL suffers from low computational and communication efficiency in large-scale hierarchical cloud-edge collaborative IoT systems. Additionally, due to heterogeneity issues, not all IoT devices necessarily benefit from the global model of traditional FL, but instead require the maintenance of personalized levels in the global training process. Therefore we extend FL into a horizontal peer-to-peer (P2P) structure and introduce our P2PFL framework: efficient peer-to-peer federated learning for users (EPFLU). EPFLU transitions the paradigms from vertical FL to a horizontal P2P structure from the user perspective and incorporates personalized enhancement techniques using private information. Through horizontal consensus information aggregation and private information supplementation, EPFLU solves the weakness of traditional FL that dilutes the characteristics of individual client data and leads to model deviation. This structural transformation also significantly alleviates the original communication issues. Additionally, EPFLU has a customized simulation evaluation framework, and uses the EUA dataset containing real-world edge server distribution, making it more suitable for real-world large-scale IoT. Within this framework, we design two extreme data distribution scenarios and conduct detailed experiments of EPFLU and selected baselines on the MNIST and CIFAR-10 datasets. The results demonstrate that the robust and adaptive EPFLU framework can consistently converge to optimal performance even under challenging data distribution scenarios. Compared with the traditional FL and selected P2PFL methods, EPFLU achieves communication time improvements of 39% and 16% respectively.

Privacy and Performance in Virtual Reality: The Advantages of Federated Learning in Collaborative Environments∗

2025-02-07T06:29:26+01:00

Federated Learning has emerged as a promising approach for maintaining data privacy across distributed environments, enabling training on a diverse range of devices from high-performance servers to low-power gadgets. Despite its potential, managing numerous data sources can strain these devices, particularly those with limited capabilities, leading to increased latency. This is especially critical in virtual reality, where real-time responsiveness is crucial due to the need for constant data connectivity. Historically, virtual reality systems have relied on tethered computer setups, restricting their flexibility and the benefits of wireless technology. However, recent advancements have enhanced the computational power of VR devices, allowing them to perform certain tasks independently. This work explores the feasibility of training a neural network on VR devices, using a federated learning approach, to develop a collaborative model aggregated and stored in the cloud. The goal is to assess the computational demands and explore the potential and constraints of leveraging VR devices for artificial intelligence applications.

Code Smell-guided Prompting for LLM-based Defect Prediction in Ansible Scripts

2025-02-07T06:35:29+01:00

Ensuring the reliability of infrastructure as code (IaC) scripts, like those written in Ansible, is vital for maintaining the performance and security of edge-cloud systems. However, the scale and complexity of these scripts make exhaustive testing impractical. To address this, we propose a large language model (LLM)-based software defect prediction (SDP) approach that uses code-smell-guided prompting (CSP). In some cases, CSP enhances LLM performance in defect prediction by embedding specific code smell indicators directly into the prompts. We explore various prompting strategies, including zero-shot, one-shot, and chain of thought CSP (CoT-CSP), to evaluate how code smell information can improve defect detection. Unlike traditional prompting, CSP uniquely leverages code context to guide LLMs in identifying defect-prone code segments. Experimental results reveal that while zero-shot prompting achieves high baseline performance, CSP variants provide nuanced insights into the role of code smells in improving SDP. This study represents exploration of LLMs for defect prediction in Ansible scripts, offering a new perspective on enhancing software quality in edge-cloud deployments.

Overcoming Terrain Challenges with Edge Computing Solutions: Optimizing WSN Deployments Over Obstacle Clad-Irregular Terrains

2025-02-07T06:40:21+01:00

Wireless sensor networks (WSNs) are primarily used for real time data collection and monitoring, especially in environments where direct human involvement is challenging due to harsh conditions. Optimized deployment of WSN nodes is a long standing issue and several ideas have been proposed to address this. Existing deployment strategies are mostly based on the assumption that the terrain for deployment of nodes is perfectly regular. This is an impractical assumption and in this paper we address this gap by proposing a deployment strategy for WSN nodes over irregular terrains. Such terrains comprise uneven elevations, morphology and vegetation based obstacles, rocky obstacles, and so on. Our approach comprises extraction of satellite images of the region of interest (RoI) from Google Earth and generating a KML file (Keyhole Markup Language) for the RoI containing the latitude, longitude, and elevation values of each and every point in the RoI. These points are used to generate a contour map of the RoI containing detailed terrain morphology. A radio frequency path loss model in combination with an advanced inverse distance weighted (IDW)-interpolation technique is proposed to ensure connectivity and coverage in such irregular terrains with varying nature of obstacles. The technique effectively detects occlusions and enables effective deployment. This edge computing approach involves real-time decision-making at the network edge (the sensor nodes) leading to a deterministic deployment of motes in diverse terrain conditions with various obstacles. The approach is compared with existing deployment techniques and the results validate its efficacy. To demonstrate the practicality of our approach, we have also implemented a deployment in real-world environmental conditions, validating our approach in challenging terrains.

Software Practice and Experience on Smart Mobility Digital Twin in Transportation and Automotive Industry: Toward SDV-empowered Digital Twin through EV Edge-Cloud and AutoML

2025-02-07T06:45:08+01:00

A digital twin is a virtual representation of a physical asset that serves as a pivotal convergence technology that facilitates real-time prediction, optimization, monitoring, control, and improved decision-making. It can be widely applied to various domains, such as automotive, manufacturing, logistics, and smart cities. The automotive industry, in particular, is actively integrating digital twins throughout the product life cycle, from research and development, production, sales, and services to enhance the overall customer experience. This paper presents insights and lessons learned on software practice and experience related to implementing smart mobility digital twins, focusing on the potential of transportation digital twins built from data collected by electric vehicles (EVs) with EV edge cloud and automated machine learning (AutoML). Despite current limitations in data sufficiency, we forecast that, as the SDV trend accelerates and the adoption of EVs increases, the digital twin will become essential for the intelligent transportation system (ITS) in future smart cities, enabling accurate traffic predictions even in areas with limited road infrastructure. The successful integration of real-time data, high-performance prediction models, and automated service environments will enhance the effectiveness toward an SDV edge-empowered transportation digital twin.