Rapid authentication of Cuscutae Semen with imbalanced sample categories using a handheld near infrared spectrometer

Journal of Near Infrared Spectroscopy, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 15, 2025

Chinese herbal medicines, primarily derived from plants and natural sources, are widely incorporated into the formulation of health foods dietary supplements. Ensuring their authenticity is crucial for maintaining therapeutic efficacy. This study introduces a method rapid authentication medicines using handheld near-infrared spectrometer coupled with chemometrics. Focusing on Cuscutae Semen, prone to market adulteration, involves spectral data collection, preprocessing, feature processing, classification algorithm. To address challenge imbalanced datasets prevalent in practice, synthetic minority over-sampling technique tomek links (SMOTETomek) comprehensive sampling was applied, enhancing model discrimination. The resulting model, combining Savitzky-Golay smoothing first derivative random forest classifier (SGFD_RF), achieved high accuracy category authentication, macro-averaged area under curve (AUC_macro) scores 0.997 (cross-validation) 0.945 (test set). And f-score recall test set reached 0.954 0.955, respectively. For content detection, SGFD_RF displayed outstanding performance, AUCs 0.995 1.000 Both 1.000. also demonstrated that competitive adaptive reweighted algorithm could reduce dimensionality training time, while providing even more precise only 8 features. approach offers reliable solution on-site medicine authentication.

Language: Английский

---

Research on Network Intrusion Detection Model Based on Hybrid Sampling and Deep Learning

Sensors, Journal Year: 2025, Volume and Issue: 25(5), P. 1578 - 1578

Published: March 4, 2025

This study proposes an enhanced network intrusion detection model, 1D-TCN-ResNet-BiGRU-Multi-Head Attention (TRBMA), aimed at addressing the issues of incomplete learning temporal features and low accuracy in classification malicious traffic found existing models. The TRBMA model utilizes Temporal Convolutional Networks (TCNs) to improve ResNet18 architecture incorporates Bidirectional Gated Recurrent Units (BiGRUs) Multi-Head Self-Attention mechanisms enhance comprehensive features. Additionally, ResNet is adapted into a one-dimensional version that more suitable for processing time-series data, while AdamW optimizer employed convergence speed generalization ability during training. Experimental results on CIC-IDS-2017 dataset indicate achieves 98.66% predicting types, with improvements precision, recall, F1-score compared baseline model. Furthermore, address challenge identification rates types small sample sizes unbalanced datasets, this paper introduces (BS-OSS), variant integrates Borderline SMOTE-OSS hybrid sampling. demonstrate effectively identifies sizes, achieving overall prediction 99.88%, thereby significantly enhancing performance

Language: Английский

---

Evaluating the Effectiveness of Machine Learning Algorithms for Network Intrusion Detection

Communications in computer and information science, Journal Year: 2025, Volume and Issue: unknown, P. 325 - 344

Published: Jan. 1, 2025

Language: Английский

---

A Novel Time‐to‐Response Based Attack Mitigation Technique for Internet of Things‐Integrated Wireless Sensor Networks

International Journal of Communication Systems, Journal Year: 2025, Volume and Issue: 38(7)

Published: March 11, 2025

ABSTRACT Various heterogeneous devices, including wireless sensor network (WSN) nodes, constitute the Internet of Things (IoT). WSNs are affected by resource constraints connected devices and face frequent security breaches data transmission loss through node‐level compromises. This study proposes a ledger‐based authenticated node‐detection technique to mitigate compromising man‐in‐middle attacks. The proposed method uses blockchain technology update response natures nodes IoT platform. Heterogeneous node operations validated periodically randomly their time‐to‐response metric. Neighbor‐to‐neighbor verification relies on periodic updates, whereas node‐to‐IoT is performed arbitrarily for global validation. validation using classification learning, prevent intervention malicious neighbors in sequences. Compromises attacks detected sequentially updates. For maximum request variant, improves detection ratio 11.18% responses 12.54% it reduces 11.17%. above results obtained comparison with existing MNIM‐CT [31], DADF [36], APTAD [29] methods described related works section.

Language: Английский

---

An interpretable dimensional reduction technique with an explainable model for detecting attacks in Internet of Medical Things devices

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: March 13, 2025

The security of Internet Medical Things (IoMT) devices is crucial for ensuring the integrity and reliability patients' medical data. These devices, operating over TCP ICMP protocols, are highly susceptible to cyberattacks. While machine learning models can detect these attacks with acceptable accuracy, their operational mechanisms remain unclear, leaving decision-making process undefined. Moreover, accuracy training time more questionable when datasets has large number sparse features class imbalances. This study introduces an interpretable feature selection technique designed enhance intrusion detection in IoMT by reducing redundant improving model efficiency. Random Forest-based explainable AI provides transparency attack classification better decision-making. simulated results employing CICIoMT2024 dataset demonstrate that proposed method significantly improves performance, Forest achieving 99% outperforming XGBoost (98%), Decision Tree (97%), Support Vector while explainability through SHAP-based analysis. Thus, simulation outcomes reveal key contributing factors various cyberattacks on IoMT, facilitating enhanced measures real-time monitoring. approach boosts interpretability, making it suitable real-world applications.

Language: Английский

---

Chaotic Adaptive Particle Swarm Optimization and Quantum-Inspired Genetic Algorithm for Robust Feature Selection in IoT Intrusion Detection

Published: Feb. 21, 2025

Language: Английский

---

Data-driven strategies for digital native market segmentation using clustering

International Journal of Cognitive Computing in Engineering, Journal Year: 2024, Volume and Issue: 5, P. 178 - 191

Published: Jan. 1, 2024

The rapid growth of internet users and social networking sites presents significant challenges for entrepreneurs marketers. Understanding the evolving behavioral psychological patterns across consumer demographics is crucial adapting business models effectively. Particularly, emergence new firms targeting adolescents future generations underscores importance comprehending online behavior communication dynamics. To tackle these challenges, we introduce a Machine Learning-based Digital Native Market Segmentation designed to cater specifically interests digital natives. Leveraging an open-access prototype dataset from (SNS), our study employs variety clustering techniques, including Kmeans, MiniBatch AGNES, Fuzzy C-means, uncover hidden teenage consumers SNS data. Through rigorous evaluation approaches by default parameters, identify optimal number clusters group with similar tastes Our findings provide actionable insights into impact critical driving marketing growth. In experiment, systematically evaluate various notably, Kmeans cluster outperforms others, demonstrating strong segmentation ability in market. Specifically, it achieves silhouette scores 63.90% 58.06% 2 3 clusters, respectively, highlighting its effectiveness segmenting

Language: Английский

---

Enhancing agricultural wireless sensor network security through integrated machine learning approaches

Security and Privacy, Journal Year: 2024, Volume and Issue: 7(6)

Published: July 2, 2024

Abstract Wireless sensor network (WSN) works with a collection of multiple nodes to fetch the data from deployed environment fulfill application whether it is agricultural monitoring, industrial etc. The region can be monitored by deploying verticals where continuous human presence not feasible. These devices are equipped limited resources and easily vulnerable various cyber‐attacks. attacker hack steal critical information WSN devices. cluster heads in play vital role process routing packets attackers launch malicious codes through sender or damage shut down entire regions. This research paper proposes framework improve security WSNs providing shield using machine learning techniques. experimental study includes comparative analysis three techniques decision tree classifier, Gaussian Naïve Bayes, random forest classifier for predicting attacks like flooding, gray hole, blackhole, TDMA that support proposed on attack dataset. achieves an accuracy 98%, Precision 97.6%, Recall F1 score 97.8% which maximum among

Language: Английский

---

Towards Reliable Diabetes Prediction: Innovations in Data Engineering and Machine Learning Applications

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: July 17, 2024

Abstract Objective Diabetes is a metabolic disorder that causes the risk of stroke, heart disease, kidney failure, and other long-term complications because diabetes generates excess sugar in blood. Machine learning (ML) models can aid diagnosing at primary stage. So, we need an efficient machine model to diagnose accurately. Methods In this paper, effective data preprocessing pipeline has been implemented process random oversampling balance data, handling imbalance distributions observational more sophisticatedly. We used four different datasets conduct our experiments. Several ML algorithms were determine best predict faultlessly. Results The performance analysis demonstrates among all algorithms, RF surpasses current works with accuracy rate 86% 98.48% for dataset-1 dataset-2; XGB DT surpass 99.27% 100% dataset-3 dataset-4 respectively. Our proposal increase by 12.15% compared without preprocessing. Conclusions This excellent research finding indicates proposed might be employed produce accurate predictions supplement preventative interventions reduce incidence its associated costs.

Language: Английский

---

ElNnGA: An Artificial Intelligence Empowered Scheme for Unknown Attack Recognition and Defense Strategy Recommendation

2022 International Conference on Networking and Network Applications (NaNA), Journal Year: 2024, Volume and Issue: unknown, P. 411 - 417

Published: Aug. 9, 2024

Language: Английский

---