Random Algorithm Research Articles

Enhancing currency prediction in international e-commerce: Bayesian-optimized random forest approach using the Klarna dataset

In the ever-evolving landscape of global commerce, marked by the convergence of digital transformation and borderless markets, this research addresses the intricate challenges of currency exchange and risk management. Leveraging Bayesian optimization, the study fine-tunes the random forest algorithm using the extensive Klarna E-commerce dataset. Through systematic analysis, the research uncovers insights into managing currency prediction amid dynamic global markets. Emphasizing the role of Bayesian optimization parameters, the study reveals nuanced trade-offs in model performance. Notably, the optimal simulation, conducted with 14 iterations, 1 job, and a random state set to 684, exhibits a standout performance, showcasing a negative mean squared error (MSE) of approximately -0.9891 and an accuracy rate of 74.63%. The primary objective is to assess the impact of Bayesian optimization in enhancing the random forest algorithm's predictive capabilities, particularly in currency prediction within international e-commerce. These findings offer refined strategies for businesses navigating the intricate landscape of global finance, empowering decision-making through a comprehensive understanding of data, algorithms, and challenges in international commerce.

Enhancing aerial image registration: outlier filtering through feature classification

In the context of feature-based image registration, the crucial task of outlier removal plays a pivotal role in achieving precise registration accuracy. This research introduces an innovative binary classifier founded on an adaptive approach for effectively identifying and eliminating outliers. The methodology begins with the utilization of the scale invariant feature transform (SIFT) to extract features from two images, initially matched using the Euclidian distance metrics. Subsequently, a classification procedure is executed to segregate the feature points into two categories: genuine matches (inliers) and spurious matches (outliers), which is accomplished through the brute-force matcher (BFM) technique. To enhance this process further, a novel classifier rooted in the random forest algorithm is introduced. This classifier is trained and tested using a comprehensive dataset curated for this study. The newly proposed classifier plays a pivotal role in attenuating the influence of outliers, ultimately leading to refined image registration process characterized by enhanced accuracy. The effectiveness of this outlier removal approach is assessed through a meticulous analysis of positional and classification accuracy. Additionally, we offer comparative insights by evaluating the performance of selected algorithm on our dataset.

Factor reallocation path for low-carbon transformation: A perspective of manufacturing industry ecosystem

The incomplete production factor market induces more factors flowing to industries with lower carbon productivity, highlighting the need for effective factor reallocation to promote low-carbon transformation. Current research is constrained by the dimension of factor endowments, with insufficient attention given to the problem of diminishing marginal benefit of factor input in factor reallocation. By combining the factor flow effect induced by productivity difference with the factor allocation equilibrium effect under diminishing marginal benefit, the factor reallocation path for low-carbon transformation is analyzed using random forest algorithm within a high-dimensional data environment from the perspective of manufacturing industry ecosystem. The results indicate that effective factor reallocation for low-carbon transformation depends on breaking the dual segmentation of factors between regions and industries to facilitate flow of factors from shrub layer industries to canopy layer industries. Adjusting key factor combinations to fall within the optimal equilibrium interval of S-shaped or anti S-shaped partial dependence curves maximizes the marginal benefits of low-carbon transformation. Industries must upgrade their factor endowment structure to accumulate essential carrying capacities that support advanced manufacturing, leveraging the low-carbon transformation dividends of advanced factors inflow and technology spillover. Otherwise, aligning with regional factor comparative advantage to cultivate differentiated characteristic manufacturing industries offers a promising alternative, capitalizing on the low-carbon efficiency premium derived from specialized division of labor.

A genetic algorithm-based feature selection approach for diabetes prediction

<p>Genetic algorithms have emerged as a powerful optimization technique for feature selection due to their ability to search through a vast feature space efficiently. This study discusses the importance of feature selection for prediction in healthcare and prominently focuses on diabetes mellitus. Feature selection is essential for improving the performance of prediction models, by finding significant features and removing unnecessary among them. The study aims to identify the most informative subset of features. Diabetes is a chronic metabolic disorder that poses significant health challenges worldwide. For the experiment, two datasets related to diabetes were downloaded from Kaggle and the results of both (datasets) with and without feature selection using the genetic algorithm were compared. Machine learning classifiers and genetic algorithms were combined to increase the precision of diabetes risk prediction. In the preprocessing phase, feature selection, machine learning classifiers, and performance metrics methods were applied to make this study feasible. The results of the experiment showed that genetic algorithm + logistic regression i.e., 80% (accuracy) works better for PIMA diabetes, and for Germany diabetes dataset genetic algorithm + random forest and genetic algorithm + K-Nearest Neighbor i.e., 98.5% performed better than other chosen classifiers. The researchers can better comprehend the importance of feature selection in healthcare through this study.</p>

Camera-IMU extrinsic calibration method based on intermittent sampling and RANSAC optimization

Accurately calibrate the extrinsic parameters between a camera and an Inertial Measurement Unit (IMU) is a prerequisite for achieving sensor data fusion in visual-inertial navigation systems. Due to delays introduced by triggering, transmission, and other factors, the sampled times of the camera and IMU do not align with the system timestamps, leading to a decrease in the accuracy of extrinsic calibration. Existing calibration methods are unable to avoid the temporal misalignment between the camera and IMU. This paper introduces a novel intermittent calibration sampling method to address temporal misalignment challenges. The approach entails detecting motion-stillness transition points in the IMU data as keyframe segmentation criteria, thereby decoupling data acquisition from system timestamps and aligning it with the sensor’s motion process. Subsequently, the robot hand-eye calibration method is applied to the extrinsic calibration of the Camera-IMU, combining the Random Sample Consensus algorithm to filter the poses of the camera and IMU. This process achieves precise calibration of extrinsic parameters. Experimental validation shows that the algorithm proposed in this paper enhances calibration accuracy when compared to the widely used Kalibr Camera-IMU calibration toolbox. The rotational extrinsic parameter accuracy increased by 181.25%, and the translational extrinsic parameter accuracy improved by 168.33%. These findings provide effective technical means for precise measurement of Camera-IMU extrinsic parameters.

Drill tools sticking prediction based on adaptive long short-term memory

As one of the most severe disasters in deep coal mining, rockburst can be prevented through drill-hole pressure relief. However, the coal mine is characterized by high crustal stress and changeable mechanical properties of surrounding rock, which will cause drill rod deflection phenomenon, then lead to rod-deflection sticking accidents. This paper proposes a prediction method based on adaptive long short-term memory (ALSTM) for rod-deflection sticking accidents to improve drilling efficiency and reduce sticking accidents. Firstly, the sticking data is collected through the intelligent drilling condition simulation experimental platform, and then the sticking features are extracted based on the sticking data. Secondly, the sticking factor is constructed, and the sticking critical line is set. Thirdly, the good-point set and the proposed random perturbation algorithm are employed to improve the spotted hyena optimizer (SHO) to obtain the improved SHO (ISHO). Finally, we use the ISHO to optimize the hyperparameters of the long short-term memory and then establish the sticking prediction model based on ALSTM. The experimental results show that the proposed prediction model meets the demands for sticking prediction very well.

Machine learning provides insights for spatially explicit pest management strategies by integrating information on population connectivity and habitat use in a key agricultural pest.

Insect pests have garnered increasing interest because of anthropogenic global change, and their sustainable management requires knowledge of population habitat use and spread patterns. To enhance this knowledge for the prevalent tea pest Empoasca onukii, we utilized a random forest algorithm and a bivariate map to develop and integrate models of its habitat suitability and genetic connectivity across China. Our modeling revealed heterogeneous spatial patterns in suitability and connectivity despite the common key environmental predictor of isothermality. Analyses indicated that tea cultivation in areas surrounding the Tibetan Plateau and the southern tip of China may be at low risk of population outbreaks because of their predicted low suitability and connectivity. However, regions along the middle and lower reaches of the Yangtze River should consider the high abundance and high recolonization potential of E. onukii, and thus the importance of control measures. Our results also emphasized the need to prevent dispersal from outside regions in the areas north of the Yangtze River and highlighted the effectiveness of internal management efforts in southwestern China and along the southeastern coast. Further projections under future conditions suggested the potential for increased abundance and spread in regions north of the Yangtze River and the southern tip of China, and indicated the importance of long-term monitoring efforts in these areas. These findings highlighted the significance of combining information on habitat use and spread patterns for spatially explicit pest management planning. In addition, the approaches we used have potential applications in the management of other pest systems and the conservation of endangered biological resources. © 2024 Society of Chemical Industry.

A fundamental study on postmortem submersion interval estimation by metabolomics analyzing of gastrocnemius muscle from submersed rat models in freshwater.

In forensic practice, determining the postmortem submersion interval (PMSI) and cause-of-death of cadavers in aquatic ecosystems has always been challenging task. Traditional approaches are not yet able to address these issues effectively and adequately. Our previous study proposed novel models to predict the PMSI and cause-of-death based on metabolites of blood from rats immersed in freshwater. However, with the advance of putrefaction, it is hardly to obtain blood samples beyond 3 days postmortem. To further assess the feasibility of PMSI estimation and drowning diagnosis in the later postmortem phase, gastrocnemius, the more degradation-resistant tissue, was collected from drowned rats and postmortem submersion model in freshwater immediately after death, and at 1day, 3 days, 5 days, 7 days, and 10 days postmortem respectively. Then the samples were analyzed with liquid chromatography-tandem mass spectrometry (LC-MS/MS) to investigate the dynamic changes of the metabolites. A total of 924 metabolites were identified. Similar chronological changes of gastrocnemius metabolites were observed in the drowning and postmortem submersion groups. The difference in metabolic profiles between drowning and postmortem submersion groups was only evident in the initial 1day postmortem, which was faded as the PMSI extension. Nineteen metabolites representing temporally-dynamic patterns were selected as biomarkers for PMSI estimation. A regression model was built based on these biomarkers with random forest algorithm, which yielded a mean absolute error (± SE) of 5.856 (± 1.296) h on validation samples from an independent experiment. These findings added to our knowledge of chronological changes in muscle metabolites from submerged vertebrate remains during decomposition, which provided a new perspective for PMSI estimation.

Using Multiple Machine Learning Models to Predict the Strength of UHPC Mixes with Various FA Percentages

Ultra High-Performance Concrete (UHPC) has shown extraordinary performance in terms of strength and durability. However, having a cost-effective and sustainable UHPC mix design is a challenge in the construction sector. This study aims on building a predictable model that can help in determining the compressive strength of UHPC. The research focuses on applying multiple machine learning (ML) models and evaluating their performance in predicting the strength prediction of UHPC. Two reliable metrics are used to evaluate the performance of the model which are the coefficient of determination (R2) and mean squared error (MSE). The parameters that are affecting the compressive strength of UHPC are fly ash percentage levels (FA%), superplasticizer content, water to binder ratio (w/b), and curing period. A total of 54 ML models were used, consisting of Linear Regression, Support Vector Machines (SVM), Neural Networks, and Random forests algorithms. Among these models, Random Forest proved to be the most effective in capturing the relationships in UHPC’s behaviour with an R squared score of 0.8857. The Random Forest ML model is also used in this paper to conduct a parametric study that will help in obtaining the compressive strength of UHPC with higher content of FA%, which is not sufficiently studied in the literature.

Diagnostic biomarkers for ST-segment elevation myocardial infarction using RNA methylation regulators

AimsAdditional evidence has indicated a correlation between N6-methyladenosine (m6A) RNA methylation and cardiovascular disease. Nevertheless, the alterations in RNA methylation modification and the expression of numerous genes remains unclear. This study aimed to identify the role of m6A in ST-segment elevation myocardial infarction (STEMI).MethodsTwo microarray datasets (GSE123342 and GSE59867) were downloaded from the GEO database. After merging the data and batch normalization, differentially expressed regulators were identified using the limma package. Subtyping consistency analysis was performed to group samples. The random forest algorithm and support vector machine were used to identify diagnostic biomarkers. Immune infiltration and inflammation levels among the subtypes were assessed using a single-sample gene set enrichment analysis.ResultsA total of 15 key differential m6A regulators (RBM15B, ELAVL1, ALKBH5, METTL16, ZC3H13, RBM15, YTHDC1, YTHDC2, YTHDF3, HNRNPC, FMR1, LRPPRC, HNRNPA2B1, RBMX, FTO) were identified using the random forest classifier and were found to be highly correlated by PPI analysis. Two distinct RNA modification patterns (cluster A and B) were validated based on the expression levels of the 15 key m6A regulators. GO and KEGG annotations showed that immunity and inflammation pathways were enriched. Immune infiltration analysis revealed that cluster 2 had higher immune activation than cluster 1. Further analysis showed that cluster 2 had a higher inflammation level, with IL-4 and IL-33 showing differential expression (p < 0.05).ConclusionA set of 15 m6A RNA methylation regulators could alter the STEMI microenvironment to improve risk stratification and clinical treatment.

Differentiating Microplastics from Natural Particles in Aqueous Suspensions Using Flow Cytometry with Machine Learning.

Microplastics (MPs) in natural waters are heterogeneously mixed with other natural particles including algal cells and suspended sediments. An easy-to-use and rapid method for directly measuring and distinguishing MPs from other naturally present colloids in the environment would expedite analytical workflows. Here, we established a database of MP scattering and fluorescence properties, either alone or in mixtures with natural particles, by stain-free flow cytometry. The resulting high-dimensional data were analyzed using machine learning approaches, either unsupervised (e.g., viSNE) or supervised (e.g., random forest algorithms). We assessed our approach in identifying and quantifying model MPs of diverse sizes, morphologies, and polymer compositions in various suspensions including phototrophic microorganisms, suspended biofilms, mineral particles, and sediment. We could precisely quantify MPs in microbial phototrophs and natural sediments with high organic carbon by both machine learning models (identification accuracies over 93%), although it was not possible to distinguish between different MP sizes or polymer compositions. By testing the resulting method in environmental samples through spiking MPs into freshwater samples, we further highlight the applicability of the method to be used as a rapid screening tool for MPs. Collectively, this workflow can be easily applied to a diverse set of samples to assess the presence of MPs in a time-efficient manner.

Comparison of Logistic Regression, Random Forest and Adaboost Algorithms for Diabetes Mellitus Classification

Diabetes mellitus is a chronic disease that affects the way the body regulates sugar (glucose). High blood sugar levels can lead to health complications including heart problems, eye disorders, nerve damage, kidney and blood vessel disorders. It is important for early detection of diabetes by utilizing data mining technology. Data mining has various classification models that can be used to detect diabetes, including logistic regression, random forest and adaboost. The comparison of the three algorithms aims to find out which algorithm is most appropriate in the classification of diabetes. From the results obtained, the random forest algorithm has the best performance in the classification of diabetes mellitus compared to other algorithms.

Intelligent Intrusion Detection System in Internet of Vehicles using Random Forest, Logistic Regression and Decision Tree Algorithm

This study introduces an Intelligent Intrusion Detection System (IDS) for Internet of Vehicles (IoV) security. Using Random Forest, Logistic Regression, and Decision Tree algorithms, it detects and prevents intrusions in real-time. By combining these algorithms, the IDS aims to improve accuracy and adapt to evolving threats. Evaluation metrics include accuracy (93.6%), precision (94.4%), recall (95.3%), and F1 score (96.7%). Analysis of false positives/negatives informs refinement for real-world use. The IDS enhances cybersecurity for connected vehicles. This IDS provides a robust defense against cyber threats in the IoV ecosystem, ensuring user safety and privacy. Its multi-algorithmic approach and high-performance metrics make it a reliable solution for safeguarding connected vehicles.

Fundus image segmentation based on random collision whale optimization algorithm

Medical image segmentation is an important technical tool, OTSU algorithm is a common method in threshold segmentation, but with the increase of the number of threshold segmentation, the selection of its threshold is a big problem, and the segmentation effect is difficult to be guaranteed. In order to solve this problem, this paper proposes a random collision whale optimization algorithm to optimize OTSU for reliable image segmentation. The algorithm is called RCWOA for short. Firstly, the Halton sequence is used to uniformly initialize the population to make the population position distribution uniform, and then the dimensional Opposition-based learning of small-hole imaging is introduced to update the whale position and find out the missing feasible solution. Finally, the random collision theory is used to update the position of the optimal individual to improve the quality of the solution, At the same time, it also improves the search ability of the algorithm. In 12 test functions, RCWOA was compared with 6 other algorithms, demonstrating the feasibility and novelty of RCWOA. In 8 experiments of fundus image segmentation, RCWOA was compared with 9 other algorithms. The results showed that RCWOA had a Friedman test composite ranking of 1.3516, ranking at the forefront, and exhibited significantly improved segmentation quality.

Multi-decadal floodplain classification and trend analysis in the Upper Columbia River valley, British Columbia

Abstract. Floodplain wetland ecosystems experience significant seasonal water fluctuation over the year, resulting in a dynamic hydroperiod, with a range of vegetation community responses. This paper assesses trends and changes in land cover and hydroclimatological variables, including air temperature, river discharge, and water level in the Upper Columbia River Wetlands (UCRW), British Columbia, Canada. A land cover classification time series from 1984 to 2022 was generated from the Landsat image archive using a random forest algorithm. Peak river flow timing, duration, and anomalies were examined to evaluate temporal coincidence with observed land cover trends. The land cover classifier used to segment changes in wetland area and open water performed well (kappa of 0.82). Over the last 4 decades, observed river discharge and air temperature have increased, precipitation has decreased, the timing of peak flow is earlier, and the flow duration has been reduced. The frequency of both high-discharge events and dry years have increased, indicating a shift towards more extreme floodplain flow behavior. These hydrometeorological changes are associated with a shift in the timing of snowmelt, from April to mid-May, and with seasonal changes in the vegetative communities over the 39-year period. Thus, woody shrubs (+6 % to +12 %) have expanded as they gradually replaced marsh and wet-meadow land covers with a reduction in open-water area. This suggests that increasing temperatures have already impacted the regional hydrology, wetland hydroperiod, and floodplain land cover in the Upper Columbia River valley. Overall, there is substantial variation in seasonal and annual land cover, reflecting the dynamic nature of floodplain wetlands, but the results show that the wetlands are drying out with increasing areas of woody/shrub habitat and loss of aquatic habitat. The results suggest that floodplain wetlands, particularly marsh and open-water habitats, are vulnerable to climatic and hydrological changes that could further reduce their areal extent in the future.

Analyzing inflow to Faye reservoir sensitivity to climate change using CMIP6 and random forest algorithm

ABSTRACT In the era of Climate Change and Climate Variability (CC and CV), renewable energy sources such as Hydropower (HP) have a significant role to play in mitigation. However, inflow to reservoir which is the key fuel for HP generation is vulnerable to CC and CV. Thus, there is a need to investigate the potential impacts of CC and CV on HP systems in the future. This study attempts to assess the potential impacts of CC and CV on the Faye reservoir inflow using the Random Forest (RF) algorithm. For this purpose, bias-adjusted precipitation and temperature data of thirteen climate model outputs and their ensemble mean from Coupled Model Inter-comparison Project Phase 6 (CMIP6) under three Shared Socioeconomic Pathways scenarios (SSP1-2.6; SSP2-4.5 and SSP5-8.5) were used as predictors. The potential changes in reservoir inflows were evaluated in the near (2025–2049), mid (2050–2074) and far (2075–2099) futures relative to the reference period (1990–2014). The results show the good performance of the RF algorithm in simulating reservoir inflows with Cor > 0.6 for all models. The annual inflows to the Faye reservoir are noted to increase in the future compared to the reference period despite the potential decrease in future precipitation probably due to land use/cover change. For the ensemble mean of models, this projected increase is estimated to around 16%, 23% and 10%, respectively under the SSP1-2.6, SSP2-4.5 and SSP5-8.5 scenarios for all projection periods. The largest annual increase is noted under the SSP2-4.5 scenario while the lowest increase is noted under the SSP5-8.5 scenario for all projection periods. This study could help the small dam managers better consider the implications of CC and CV on inflow management.

COVID-19 mortality prediction in Hungarian ICU settings implementing random forest algorithm

The emergence of newer SARS-CoV-2 variants of concern (VOCs) profoundly changed the ICU demography; this shift in the virus’s genotype and its correlation to lethality in the ICUs is still not fully investigated. We aimed to survey ICU patients’ clinical and laboratory parameters in correlation with SARS-CoV-2 variant genotypes to lethality. 503 COVID-19 ICU patients were included in our study beginning in January 2021 through November 2022 in Hungary. Furthermore, we implemented random forest (RF) as a potential predictor regarding SARS-CoV-2 lethality among 649 ICU patients in two ICU centers. Survival analysis and comparison of hypertension (HT), diabetes mellitus (DM), and vaccination effects were conducted. Logistic regression identified DM as a significant mortality risk factor (OR: 1.55, 95% CI 1.06–2.29, p = 0.025), while HT showed marginal significance. Additionally, vaccination demonstrated protection against mortality (p = 0.028). RF detected lethality with 81.42% accuracy (95% CI 73.01–88.11%, [AUC]: 91.6%), key predictors being PaO2/FiO2 ratio, lymphocyte count, and chest Computed Tomography Severity Score (CTSS). Although a smaller number of patients require ICU treatment among Omicron cases, the likelihood of survival has not proportionately increased for those who are admitted to the ICU. In conclusion, our RF model supports more effective clinical decision-making among ICU COVID-19 patients.

DERIVATION AND VALIDATION OF A MACHINE LEARNING MODEL FOR THE PREVENTION OF UNPLANNED DIALYSIS.

Approximately half of all patients with advanced chronic kidney disease (CKD) who progress to kidney failure initiate dialysis in an unplanned fashion which is associated with high morbidity, mortality, and healthcare costs. A novel prediction model designed to identify advanced CKD patients who are at high risk for developing kidney failure over short time frames (6-12 months) may help reduce the rates of unplanned dialysis and improve the quality of transitions from CKD to kidney failure. We performed a retrospective study employing machine learning random forest algorithms incorporating routinely collected age and sex data along with time-varying trends in laboratory measurements to derive and validate six- and 12-month kidney failure risk prediction models in the advanced CKD population. The models were comprehensively characterized in three independent cohorts in Ontario, Canada - derived in a cohort of 1,849 consecutive advanced CKD patients (mean [standard deviation] age 66 [15] years, eGFR 19 [7] mL/min/1.73m2), and validated in two external advanced CKD cohorts (n=1,356; age 69 [14] years, eGFR 22 [7] mL/min/1.73m2). Across all cohorts, 55% of patients experienced kidney failure, of which 35% involved unplanned dialysis. The six- and 12-month models demonstrated excellent discrimination with area under the receiver operating characteristic curve of 0.88 (95%CI: 0.87-0.89) and 0.87 (95%CI: 0.86-0.87) along with high probabilistic accuracy with Brier scores of 0.10 (95%CI 0.09-0.10) and 0.14 (95%CI 0.13-0.14), respectively. The models were also well-calibrated and delivered timely alerts on a significant number of patients who ultimately initiated dialysis in an unplanned fashion. Similar results were found upon external validation testing. These machine-learning models using routinely collected patient data accurately predict near-future kidney failure risk among the advanced CKD population, and retrospectively deliver advanced warning on a substantial proportion of unplanned dialysis events. Optimal implementation strategies still need to be elucidated.

A hybrid feature weighted attention based deep learning approach for an intrusion detection system using the random forest algorithm.

Due to the recent advances in the Internet and communication technologies, network systems and data have evolved rapidly. The emergence of new attacks jeopardizes network security and make it really challenging to detect intrusions. Multiple network attacks by an intruder are unavoidable. Our research targets the critical issue of class imbalance in intrusion detection, a reflection of the real-world scenario where legitimate network activities significantly out number malicious ones. This imbalance can adversely affect the learning process of predictive models, often resulting in high false-negative rates, a major concern in Intrusion Detection Systems (IDS). By focusing on datasets with this imbalance, we aim to develop and refine advanced algorithms and techniques, such as anomaly detection, cost-sensitive learning, and oversampling methods, to effectively handle such disparities. The primary goal is to create models that are highly sensitive to intrusions while minimizing false alarms, an essential aspect of effective IDS. This approach is not only practical for real-world applications but also enhances the theoretical understanding of managing class imbalance in machine learning. Our research, by addressing these significant challenges, is positioned to make substantial contributions to cybersecurity, providing valuable insights and applicable solutions in the fight against digital threats and ensuring robustness and relevance in IDS development. An intrusion detection system (IDS) checks network traffic for security, availability, and being non-shared. Despite the efforts of many researchers, contemporary IDSs still need to further improve detection accuracy, reduce false alarms, and detect new intrusions. The mean convolutional layer (MCL), feature-weighted attention (FWA) learning, a bidirectional long short-term memory (BILSTM) network, and the random forest algorithm are all parts of our unique hybrid model called MCL-FWA-BILSTM. The CNN-MCL layer for feature extraction receives data after preprocessing. After convolution, pooling, and flattening phases, feature vectors are obtained. The BI-LSTM and self-attention feature weights are used in the suggested method to mitigate the effects of class imbalance. The attention layer and the BI-LSTM features are concatenated to create mapped features before feeding them to the random forest algorithm for classification. Our methodology and model performance were validated using NSL-KDD and UNSW-NB-15, two widely available IDS datasets. The suggested model's accuracies on binary and multi-class classification tasks using the NSL-KDD dataset are 99.67% and 99.88%, respectively. The model's binary and multi-class classification accuracies on the UNSW-NB15 dataset are 99.56% and 99.45%, respectively. Further, we compared the suggested approach with other previous machine learning and deep learning models and found it to outperform them in detection rate, FPR, and F-score. For both binary and multiclass classifications, the proposed method reduces false positives while increasing the number of true positives. The model proficiently identifies diverse network intrusions on computer networks and accomplishes its intended purpose. The suggested model will be helpful in a variety of network security research fields and applications.

Key nodes of misinformation source inference: A message-passing-based approach

The misinformation spreading in social networks causes unpredictable damage to the networked system, thus inferring the misinformation source is an important research topic in the field of network science and security. Many source inference algorithms have been proposed to find the most likely propagation source through observable snapshot. However, under limited observable conditions, observing different nodes states markedly affects the algorithm’s effectiveness. Yet, we still lack relevant research on which nodes can more accurately assist us in completing source inference. Here, we propose the heuristic message-passing-based algorithm to find the key nodes that can maximize the accuracy of source inference, which uses the average rank of the source in the message-passing method as a measure and performs continuous annealing on this basis to update the set. As a comparison, we propose random selection algorithm as the basic, high-eigenvalue algorithm and high-degree algorithm focused on centrality, and basic message-passing-based algorithm from the perspective of energy entropy in message passing. Through extensive numerical simulation on artificial and real-world networks, compared with other four algorithms, our heuristic message-passing-based algorithm finds the optimal key node set that can more accurately complete source inference. Moreover, it has over 8% higher inference accuracy than other methods in low visibility situations especially.