Traffic classification is a process which assorts computer network traffic into predefined traffic classes by utilizing packet header information or network packet statistics. Real-time traffic classification is mainly used in network management tasks comprising traffic shaping and flow prioritization as well as in network security applications for intrusion detection. Machine Learning (ML) based traffic classification that exploits statistical characteristics of traffic, has come into prominence recently, due to its ability to cope with encrypted traffic and newly emerging network applications utilizing non-standard ports to circumvent firewalls. To meet high data rates and achieve online classification with ML-based techniques, Field Programmable Gate Arrays (FPGAs) providing abundant parallelism and high operating frequency is the most appropriate platform. In this paper, we propose to use Simple Classification and Regression Trees (Simple CART) machine learning algorithm for traffic classification. However, the variations in node sizes of Simple CART decision tree caused by discretization pre-process incur memory and resource inefficiency problems when the tree is directly mapped onto the hardware. To resolve these problems, we propose to represent Simple CART decision tree by two stage hybrid data structure (Extended-Simple CART) that comprises multiple range trees in Stage 1 and a Simple CART decision tree enriched with bitmaps at its nodes in Stage 2. Our design is implemented on parallel and pipelined architectures using Field Programmable Gate Arrays (FPGAs) to acquire high throughput. Extended-Simple CART architecture can sustain 557 Gbps or 1741 million classification per second (MCPS) (for the minimum packet size of 40 Bytes) on a state-of-the-art FPGA and achieve an accuracy of 96.8% while classifying an internet traffic trace including eight application classes.
Traffic classification process categorizes internet traffic into application classes by exploiting packet header data or collected packet statistics. Real-time internet traffic classification is mostly required for network management and security applications. Machine Learning (ML) based traffic classification approaches which utilize statistical properties of traffic flows, have recently attracted great deal of attention from the researches due to its operability under encrypted traffic conditions. In this paper, we propose to use Simple Classification and Regression Trees Forest (SCF) for internet traffic classification. Our proposed scheme comprising multiple parallel trees demonstrates a substantial improvement in search delay and throughput as well as in the memory footprint when compared to a traditional single Simple CART structure. To reach high data rates for realtime classification, we also proposed a parallel and pipelined architecture on Field Programmable Gate Arrays (FPGAs) that support SCF data structure. The post place-and-route FPGA results demonstrate that our design can sustain 854 Gbps or 2669 million classification per second (MCPS) for the minimum packet size of 40 Bytes. Furthermore, our architecture shows an accuracy of 96.6719% for real internet traffic with eight application classes.
29 May 2025 in Anadolu Kliniği Tıp Bilimleri Dergisi
Aim: This study aims to develop and validate LuminaURO, a Retrieval-Augmented Generation (RAG)-based AI Assistant specifically designed for urological healthcare, addressing the limitations of conventional Large Language Models (LLMs) in healthcare applications. Methods: We developed LuminaURO using a specialized repository of urological documents and implemented a novel pooling methodology to search multilingual documents and aggregate information for response generation. The system was evaluated using multiple similarity algorithms (OESM, Spacy, T5, and BERTScore) and expert assessment by urologists (n=3). Results: LuminaURO generates responses within 8-15 seconds from multilingual documents and enhances user interaction by providing two contextually relevant follow-up questions per query. The architecture demonstrates significant improvements in search latency, memory requirements, and similarity metrics compared to state-of-the-art approaches. Validation shows similarity scores of 0.6756, 0.7206, 0.9296, 0.9223, and 0.9183 for English responses, and 0.6686, 0.7166, 0.8119, 0.9220, 0.9315, and 0.9086 for Turkish responses. Expert evaluation by urologists revealed similarity scores of 0.9444 and 0.9408 for English and Turkish responses, respectively. Conclusion: LuminaURO successfully addresses the limitations of conventional LLM implementations in healthcare by utilizing specialized urological documents and our innovative pooling methodology for multi-language document processing. The high similarity scores across multiple evaluation metrics and strong expert validation confirm the system’s effectiveness in providing accurate and relevant urological information. Future research will focus on expanding this approach to other medical specialties, with the ultimate goal of developing LuminaHealth, a comprehensive healthcare assistant covering all medical domains.