This study presents an innovative hybrid evolutionary algorithm that combines the Arctic Puffin Optimization (APO) algorithm with the JADE dynamic differential evolution framework. The APO algorithm, inspired by the foraging patterns of Arctic puffins, demonstrates certain challenges, including a tendency to converge prematurely at local minima, a slow rate of convergence, and an insufficient equilibrium between the exploration and exploitation processes. To mitigate these drawbacks, the proposed hybrid approach incorporates the dynamic features of JADE, which enhances the exploration–exploitation trade-off through adaptive parameter control and the use of an external archive. By synergizing the effective search mechanisms modeled after the foraging behavior of Arctic puffins with JADE’s advanced dynamic strategies, this integration significantly improves global search efficiency and accelerates the convergence process. The effectiveness of APO-JADE is demonstrated through benchmark tests against well-known IEEE CEC 2022 unimodal and multimodal functions, showing superior performance over 32 compared optimization algorithms. Additionally, APO-JADE is applied to complex engineering design problems, including the optimization of engineering structures and mechanisms, revealing its practical utility in navigating challenging, multi-dimensional search spaces typically encountered in real-world engineering problems. The results confirm that APO-JADE outperformed all of the compared optimizers, effectively addressing the challenges of unknown and complex search areas in engineering design optimization.

This paper presents JADEDO, a hybrid optimization method that merges the dandelion optimizer’s (DO) dispersal-inspired stages with JADE’s (adaptive differential evolution) dynamic mutation and crossover operators. By integrating these complementary mechanisms, JADEDO effectively balances global exploration and local exploitation for both unimodal and multimodal search spaces. Extensive benchmarking against classical and cutting-edge metaheuristics on the IEEE CEC2022 functions—encompassing unimodal, multimodal, and hybrid landscapes—demonstrates that JADEDO achieves highly competitive results in terms of solution accuracy, convergence speed, and robustness. Statistical analysis using Wilcoxon sum-rank tests further underscores JADEDO’s consistent advantage over several established optimizers, reflecting its proficiency in navigating complex, high-dimensional problems. To validate its real-world applicability, JADEDO was also evaluated on three engineering design problems (pressure vessel, spring, and speed reducer). Notably, it achieved top-tier or near-optimal designs in constrained, high-stakes environments. Moreover, to demonstrate suitability for security-oriented tasks, JADEDO was applied to an attack-response optimization scenario, efficiently identifying cost-effective, low-risk countermeasures under stringent time constraints. These collective findings highlight JADEDO as a robust, flexible, and high-performing framework capable of tackling both benchmark-oriented and practical optimization challenges.

Accurate short-horizon downlink throughput prediction is essential for automation in high-density 5G deployments (e.g., stadiums and events), where user load, scheduling decisions, and interference conditions change rapidly and produce highly variable user-perceived rates. This paper benchmarks lightweight regression models for per-user throughput prediction from readily available radio access network (RAN) key performance indicators (KPIs) and studies a risk-aware extension that augments point forecasts with calibrated uncertainty and an abstention (deferral) rule. Experiments use a strictly time-ordered train/calibration/test protocol on the Liverpool 5G High-Density Demand (L5GHDD) dataset. The target is strongly zero-inflated (about 62% of samples at 0 Mbps) and heavy-tailed, creating regimes where average-error optimization can mask rare but operationally important bursts. In the point-prediction benchmark, the best model is a tuned two-stage support vector regressor with a mean absolute error (MAE) of $0.452$ Mbps, while the strongest single-stage model attains a weighted mean absolute percentage error (WMAPE) of $56.200\%$. For uncertainty quantification, we compare standard split conformal prediction against two input-adaptive alternatives. Constant-width split conformal attains $88.900\%$ marginal coverage for a nominal 90% target with an average interval width of $2.288$ Mbps, but width-based deferral is degenerate because all intervals have the same size. Variable-length conformal intervals preserve near-nominal coverage ($91.100\%$) while producing informative width variation: normalized conformal reduces the average width to $1.344$ Mbps, and conformalized quantile regression reduces it to $0.641$ Mbps. At a deferral threshold of $1.500$ Mbps, constant-width conformal defers all samples, whereas normalized conformal still acts on $61.200\%$ of samples with selective MAE $0.219$ Mbps. These results show that input-adaptive uncertainty is necessary for meaningful selective prediction in heteroscedastic 5G throughput dynamics.

Machine-learning intrusion detection on multi-attack Internet-of-Things (IoT) datasets is often reported through aggregate metrics that can hide near-failure on rare attack classes, and through single hand-picked feature subsets whose stability is seldom examined. Using the real TON_IoT Network dataset (211,043 flows, 10 classes, with man-in-the-middle (MITM) traffic at 0.5% prevalence), this study contributes a transparent empirical mapping rather than a new detector. A strong full-feature Random Forest reaches 0.969 accuracy and 0.945 macro-averaged F1 (macro-F1) yet recalls only 0.776 of the rare MITM class, and a mutual-information filter raises aggregate macro-F1 to 0.948 while leaving worst-class recall unchanged at 0.776. A self-coded non-dominated sorting genetic algorithm (NSGA-II) is then used to map the trade-off between macro-F1 and worst-class recall as the selected subset varies, alongside a matched-size random-subset control, standard imbalance remedies, and two additional classifiers. Three findings are robust. First, imbalance remedies on the full feature set—class weighting and, especially, the synthetic minority over-sampling technique (SMOTE)—raise MITM recall to 0.818 and 0.904, matching or exceeding the wrapper, so feature selection is not the most effective tool for the rare class. Second, re-adding a single destination-port feature inflates ransomware recall by 0.094 and cross-site-scripting (XSS) recall by 0.066 while doing nothing for MITM, showing that several apparently detected classes are separable by an identifier artifact whereas the genuinely hard class is not. Third, selected subsets are unstable (Nogueira index 0.21) despite reproducible objective values. The wrapper’s own rare-class advantage is small and does not reach significance under a cross-validation-appropriate corrected test. The two transferable findings replicate on the NSL-KDD benchmark. We argue that IoT intrusion-detection studies should report the aggregate–worst-class frontier, an identifier ablation, a selection-stability index, and cross-validation-appropriate significance tests.