Improved Machine Learning Research Articles

Calculating Pairwise Similarity of Polymer Ensembles via Earth Mover's Distance.

Synthetic polymers, in contrast to small molecules and deterministic biomacromolecules, are typically ensembles composed of polymer chains with varying numbers, lengths, sequences, chemistry, and topologies. While numerous approaches exist for measuring pairwise similarity among small molecules and sequence-defined biomacromolecules, accurately determining the pairwise similarity between two polymer ensembles remains challenging. This work proposes the earth mover's distance (EMD) metric to calculate the pairwise similarity score between two polymer ensembles. EMD offers a greater resolution of chemical differences between polymer ensembles than the averaging method and provides a quantitative numeric value representing the pairwise similarity between polymer ensembles in alignment with chemical intuition. The EMD approach for assessing polymer similarity enhances the development of accurate chemical search algorithms within polymer databases and can improve machine learning techniques for polymer design, optimization, and property prediction.

Fractal feature selection model for enhancing high-dimensional biological problems

The integration of biology, computer science, and statistics has given rise to the interdisciplinary field of bioinformatics, which aims to decode biological intricacies. It produces extensive and diverse features, presenting an enormous challenge in classifying bioinformatic problems. Therefore, an intelligent bioinformatics classification system must select the most relevant features to enhance machine learning performance. This paper proposes a feature selection model based on the fractal concept to improve the performance of intelligent systems in classifying high-dimensional biological problems. The proposed fractal feature selection (FFS) model divides features into blocks, measures the similarity between blocks using root mean square error (RMSE), and determines the importance of features based on low RMSE. The proposed FFS is tested and evaluated over ten high-dimensional bioinformatics datasets. The experiment results showed that the model significantly improved machine learning accuracy. The average accuracy rate was 79% with full features in machine learning algorithms, while FFS delivered promising results with an accuracy rate of 94%.

Prediction of postoperative infectious complications in elderly patients with colorectal cancer: a study based on improved machine learning

BackgroundInfectious complications after colorectal cancer (CRC) surgery increase perioperative mortality and are significantly associated with poor prognosis. We aimed to develop a model for predicting infectious complications after colorectal cancer surgery in elderly patients based on improved machine learning (ML) using inflammatory and nutritional indicators.MethodsThe data of 512 elderly patients with colorectal cancer in the Third Affiliated Hospital of Anhui Medical University from March 2018 to April 2022 were retrospectively collected and randomly divided into a training set and validation set. The optimal cutoff values of NLR (3.80), PLR (238.50), PNI (48.48), LCR (0.52), and LMR (2.46) were determined by receiver operating characteristic (ROC) curve; Six conventional machine learning models were constructed using patient data in the training set: Linear Regression, Random Forest, Support Vector Machine (SVM), BP Neural Network (BP), Light Gradient Boosting Machine (LGBM), Extreme Gradient Boosting (XGBoost) and an improved moderately greedy XGBoost (MGA-XGBoost) model. The performance of the seven models was evaluated by area under the receiver operator characteristic curve, accuracy (ACC), precision, recall, and F1-score of the validation set.ResultsFive hundred twelve cases were included in this study; 125 cases (24%) had postoperative infectious complications. Postoperative infectious complications were notably associated with 10 items features: American Society of Anesthesiologists scores (ASA), operation time, diabetes, presence of stomy, tumor location, NLR, PLR, PNI, LCR, and LMR. MGA-XGBoost reached the highest AUC (0.862) on the validation set, which was the best model for predicting postoperative infectious complications in elderly patients with colorectal cancer. Among the importance of the internal characteristics of the model, LCR accounted for the highest proportion. Conclusions: This study demonstrates for the first time that the MGA-XGBoost model with 10 risk factors might predict postoperative infectious complications in elderly CRC patients.

Changes in oscillatory patterns of microstate sequence in patients with first-episode psychosis

We aimed to utilize chaos game representation (CGR) for the investigation of microstate sequences and explore its potential as neurobiomarkers for psychiatric disorders. We applied our proposed method to a public dataset including 82 patients with first-episode psychosis (FEP) and 61 control subjects. Two time series were constructed: one using the microstate spacing distance in CGR and the other using complex numbers representing the microstate coordinates in CGR. Power spectral features of both time series and frequency matrix CGR (FCGR) were compared between groups and employed in a machine learning application. The four canonical microstates (A, B, C, and D) were identified using both shared and separate templates. Our results showed the microstate oscillatory pattern exhibited alterations in the FEP group. Using oscillatory features improved machine learning performance compared with classical features and FCGR. This study opens up new avenues for exploring the use of CGR in analyzing EEG microstate sequences. Features derived from microstate sequence CGR offer fine-grained neurobiomarkers for psychiatric disorders.

Correction to “Application of Machine Learning and Reaction Optimization for the Iterative Improvement of Enantioselectivity of Cinchona-Derived Phase Transfer Catalysts”

Correction to “Application of Machine Learning and Reaction Optimization for the Iterative Improvement of Enantioselectivity of Cinchona-Derived Phase Transfer Catalysts”

Cybersecurity and Risk Prediction Based on Machine Learning Algorithms

Abstract This paper focuses on the application of machine learning in cybersecurity and risk prediction. Considering the complexity and dynamics of the cyber environment, the role of plain Bayesian algorithms in improving the accuracy of cybersecurity prediction is explored. First, the limitations of traditional linear regression machine learning algorithms in cybersecurity are analyzed, pointing out that it is difficult for these algorithms to capture complex nonlinear relationships. For this reason, plain Bayes is introduced to improve machine learning for more accurate prediction and assessment of cybersecurity risks by capturing functions and probabilistic agent models. In terms of methodology, this paper combines various technical tools such as statistical analysis, feature extraction, and risk assessment. The research results show that the improved algorithm outperforms the traditional method in simulation experiments, which is reflected in the 10% improvement in risk assessment accuracy and 20% enhancement of detection sensitivity to network intrusion. In addition, through big data analysis, the algorithm successfully identifies key risk points in network security and effectively predicts potential security threats. The improved machine learning algorithm with plain Bayes has significant advantages in network security and risk prediction and can effectively enhance network security protection capability and prediction accuracy.

Takween: A Comprehensive Arabic Collection and Annotation Tool

In the rapidly evolving field of Natural Language Processing (NLP), the availability of labeled data is crucial for developing and improving machine learning models. Labeled data provides the essential context and structure required for training algorithms to accurately interpret and process human language. However, obtaining labeled data, especially for languages with complex structures like Arabic, presents significant challenges. This paper introduces ”Takween,” a novel tool designed to meet the need for high-quality labeled data in Arabic NLP. Takween leverages social media and Wikipedia as data sources to facilitate the collection of large-scale Arabic text data. The tool enables users to annotate and label the collected data, with functionalities for reviewing annotations and visualizing the labeled data to ensure quality. Additionally, Takween incorporates active learning techniques, which allow for automatically selecting data samples for annotation based on model uncertainty. By integrating data collection, annotation, review, and visualization features with active learning capabilities, Takween offers a comprehensive solution for acquiring high-quality labeled data for Arabic NLP tasks.

Fuzzy Similarity Analysis of Effective Training Samples to Improve Machine Learning Estimations of Water Quality Parameters Using Sentinel-2 Remote Sensing Data

Fuzzy Similarity Analysis of Effective Training Samples to Improve Machine Learning Estimations of Water Quality Parameters Using Sentinel-2 Remote Sensing Data

Improving Machine Learning Based PM2.5 Prediction by Segregating Biomass Emission Factor from Chemical Transport Model

Located in the heart of Mainland Southeast Asia, Thailand is associated with high biomass burning (BB) activities from local and neighbouring countries. The seasonal pattern of BB manifests itself as a potential predictor for PM2.5 concentration. Consequently, we enhanced machine learning based PM2.5 prediction by segregating BB factor from the Community Multiscale Air Quality (CMAQ). Two Light Gradient Boosting Machine (LightGBM) models with different CMAQ predictors were developed: the BB-integrated model, which incorporated CMAQ-simulated PM2.5 from all emission sources and the BB-segregated model, which incorporated CMAQ-simulated PM2.5 from sources other than BB (CMAQ_PM25_Othr) and CMAQ-simulated PM2.5 from BB emissions (CMAQ_PM25_BB). The two models had shared control predictors, which included simulated meteorological variables from WRF model, population, elevation, and land-use variables, and they were evaluated using a crossvalidation (CV). The BB-segregated model outperformed the BB-integrated model, achieving overall-CV R2 values of 0.86 and 0.82, respectively. The analysis of feature importance for the BB-segregated model indicates that CMAQ_PM25_Othr and CMAQ_PM25_BB are the two most significant predictors. These findings emphasize the importance of considering BB emissions when predicting PM2.5 concentrations, particularly in regions with high BB activities.

Advancing energy efficiency: Harnessing machine learning for smart grid management

The concept of Smart Grids (SG) has emerged as a solution to address challenges in traditional power systems, including resource inefficiency, reliability issues, and instability. Since its inception in the early 21st century, Smart Grid technology has undergone significant development, integrating advanced information communication and automation technologies with conventional power infrastructure. This integration enhances efficiency, reliability, and sustainability, while enabling the integration of renewable energy sources and optimizing energy distribution and consumption. Machine learning algorithms play a pivotal role in the development of Smart Grids, facilitating energy consumption prediction, optimization, anomaly detection, and fault diagnosis. This paper explores methodologies for developing and improving machine learning algorithms for efficient energy consumption prediction and management within Smart Grids. It discusses the application of deep learning techniques, reinforcement learning, and integration with the Internet of Things (IoT) to enhance energy management systems. The study highlights the potential impact of deep convolutional neural networks (CNNs) on energy consumption regulation and emphasizes the need for further research to address challenges associated with model complexity and data requirements in Smart Grid contexts.

Safeguarding Industry 4.0: A Machine Learning Approach for Cyber-Physical Systems Security and Sustainability

The proliferation of connected objects, with over 8 billion IoT devices currently and a projected increase to 41 billion by 2027, signifies the widespread integration of technology in sectors like Smart City, Industry 4.0, e-commerce, and e-health. This study focuses on the security assessment of Cyber-Physical Systems (CPS) in manufacturing processes, utilizing six supervised algorithms on a dataset with 61 features. The results not only offer valuable insights into security but also contribute to the optimization of machine learning models. This research implicitly addresses the sustainability aspect by acknowledging the broader impact of CPS technologies. Cyber-Physical Systems (CPS) optimization of machine learning models not only fits in with the industry 4.0 framework’s overarching goal of promoting environmentally friendly practices, but it also creates a vital connection between sustainability and the security paradigm that these complex systems are built upon. This mutually beneficial relationship highlights how improving machine learning algorithms with the goal of reducing environmental impact also helps to strengthen the security infrastructure of CPS. Industry 4.0 prioritizes environmental responsibility by emphasizing the development and application of eco-conscious practices. It also acknowledges the interdependence of sustainability and security within the framework of dynamic cyber-physical ecosystems.

An Improved Machine Learning Based Improved GOOSE Protocol for Fault Detection

An Improved Machine Learning Based Improved GOOSE Protocol for Fault Detection

Generating Synthetic Light-Adapted Electroretinogram Waveforms Using Artificial Intelligence to Improve Classification of Retinal Conditions in Under-Represented Populations.

Visual electrophysiology is often used clinically to determine the functional changes associated with retinal or neurological conditions. The full-field flash electroretinogram (ERG) assesses the global contribution of the outer and inner retinal layers initiated by the rods and cone pathways depending on the state of retinal adaptation. Within clinical centers, reference normative data are used to compare clinical cases that may be rare or underpowered within a specific demographic. To bolster either the reference dataset or the case dataset, the application of synthetic ERG waveforms may offer benefits to disease classification and case-control studies. In this study and as a proof of concept, artificial intelligence (AI) to generate synthetic signals using generative adversarial networks is deployed to upscale male participants within an ISCEV reference dataset containing 68 participants, with waveforms from the right and left eye. Random forest classifiers further improved classification for sex within the group from a balanced accuracy of 0.72-0.83 with the added synthetic male waveforms. This is the first study to demonstrate the generation of synthetic ERG waveforms to improve machine learning classification modelling with electroretinogram waveforms.

Data Encoding with Generative AI: Towards Improved Machine Learning Performance

Data Encoding with Generative AI: Towards Improved Machine Learning Performance

Transformed Image Features Can Improve Machine Learning Performance for Detecting Benign-Malignant of Breast Cancer

Transformed Image Features Can Improve Machine Learning Performance for Detecting Benign-Malignant of Breast Cancer

Segmenting and classifying skin lesions using a fruit fly optimization algorithm with a machine learning framework

The deadliest forms of skin cancer, melanomas have a large fatality rate. In the United States of America, 196,060 new cases of melanoma are anticipated in 2020. In the past, many automated methods for diagnosing skin lesions have been proposed, but they have not yet proven to be very accurate. Based on skin cells’ exposure to sunlight, aberrant skin cell development frequently results in skin cancer. Ultraviolet radiation, viruses, bacteria, chemicals, and fungi are the main contributors to skin conditions. The creation of a precise computer-aided system for diagnosing breast cancer is of tremendous clinical importance. An improved machine learning framework has been developed in this research to detect skin lesions or skin cancer. Hence it is important to segment and classify the skin lesion. The research utilizes the fruit fly optimization algorithm and machine learning framework to segment and classifies skin disease and cancer. This platform's central idea is to use the fruit fly optimization algorithm (FOA) to improve two crucial SVM variables and create an FOA-based SVM (FOA-SVM) for the diagnosis of skin cancer. The integrative approach not only improves accuracy but also provides important data for more accurate classification.

PSOGSA: A parallel implementation model for data clustering using new hybrid swarm intelligence and improved machine learning technique

With the digitization of the entire world and huge requirements of understanding unknown patterns from the data, clustering becomes an important research area. The quick and accurate division of large datasets with a range of properties or features becomes challenging. The parallel implementation of clustering algorithms must satisfy stringent computational requirements to handle large amounts of data. This can be achieved by designing a GPU based optimal computational model with a heuristic approach. Swarm Intelligence (SI), a family of bio-inspired algorithms, that has been effectively applied to a number of real-world clustering problems. The Gravitational Search Algorithm (GSA) is a heuristic search optimization approach based on Newton's Law of Gravitation and mass interactions. Although it has a slow searching rate in the last iterations, this strategy has been proved to be capable of discovering the global optimum. This paper presents GPU based hybrid parallel algorithms for data clustering. A newly developed, hybrid Particle Swarm Optimization (PSO) and Gravitational Search Algorithm (GSA) i.e., PSOGSA achieves the global optima. PSOGSA utilizes novel training methods for enhanced Neural Networks (NN) in order to examine the efficiency of algorithms and resolves the challenges of trapping in local minima. This also shows the sluggish convergence rate of standard evolutionary learning algorithms. The Nearest Neighbour Partition (Partitioning of the Neighbourhood) algorithm can be used to improve the performance of NN. A parallel version of Hybrid PSOGSA with NN is implemented to achieve optimal results with better computational time. Compared to the CPU-based regular PSO, the suggested Hybrid PSOGSA with NN achieved optimal clustering with 71% improved computational time.

Steady flow properties and spectral absorption potential of supercritical carbon dioxide nanofluids: Experimental comparison and machine learning optimization

To improve the performance of the power cycle and accelerate the realization of the carbon neutrality goal, the development and application of nanofluids are imminent. In this research, we established a supercritical nanofluid spectral absorbance prototype and prepared seven supercritical carbon dioxide nanofluids for various valuable works. The experiments indicated that titanium carbide/supercritical carbon dioxide had the most outstanding stability with the shortest stabilization time (3.2 min) and the smallest fluctuation amplitude in spectral irradiance (±6.33%). Graphene/supercritical carbon dioxide emerged as the most competitive choice in terms of deposition characteristics and spectral absorption capacity, boasting a deposition density and spectral absorbance of 8.146% and 86.39%. Titanium was the most suitable raw material for constructing the prototype. Then, a performance factor was introduced to quantify the contribution of stability and deposition characteristics, and hydroxylated multi-walled carbon nanotube/supercritical carbon dioxide was found to have the most outstanding comprehensive performance with promising applications. Finally, we combined particle swarm optimization and artificial neural networks to develop an improved machine learning optimization model. Compared to the original prediction model, the root mean square error and mean absolute error of this model were reduced by 64.02% and 66.8%. Significantly, the spectral irradiance through the supercritical carbon dioxide nanofluid was reduced by 68.21% compared to the base working fluid, providing guidelines for improving the performance of the power cycle.

Heart Disease Prediction Using Outlier Removal based Max Voting Ensemble Method

Heart disease has emerged as a serious health concern for many individuals due to its high death rate around the world. The routine clinical data analysis has a significant difficulty in the early diagnosis of cardiac disease. The identification of cardiac disease may benefit from the use of machine learning. To improve machine learning models, several studies have previously been conducted. The suggested study uses the maximum voting ensemble technique of classification to effectively identify heart disease. The suggested classifier is a more reliable and accurate approach. To identify and eliminate outliers, conduct Inter quartile range outlier removal and min-max normalization during preprocessing. Accuracy, Precision, Recall, and F1 Score are calculated and evaluated against various models. For the heart disease dataset collected from the Kaggle, the suggested max voting ensemble classifier has an accuracy of 99.22%.

PASTEL

Federated Learning (FL) aims to improve machine learning privacy by allowing several data owners in edge and ubiquitous computing systems to collaboratively train a model, while preserving their local training data private, and sharing only model training parameters. However, FL systems remain vulnerable to privacy attacks, and in particular, to membership inference attacks that allow adversaries to determine whether a given data sample belongs to participants' training data, thus, raising a significant threat in sensitive ubiquitous computing systems. Indeed, membership inference attacks are based on a binary classifier that is able to differentiate between member data samples used to train a model and non-member data samples not used for training. In this context, several defense mechanisms, including differential privacy, have been proposed to counter such privacy attacks. However, the main drawback of these methods is that they may reduce model accuracy while incurring non-negligible computational costs. In this paper, we precisely address this problem with PASTEL, a FL privacy-preserving mechanism that is based on a novel multi-objective learning function. On the one hand, PASTEL decreases the generalization gap to reduce the difference between member data and non-member data, and on the other hand, PASTEL reduces model loss and leverages adaptive gradient descent optimization for preserving high model accuracy. Our experimental evaluations conducted on eight widely used datasets and five model architectures show that PASTEL significantly reduces membership inference attack success rates by up to -28%, reaching optimal privacy protection in most cases, with low to no perceptible impact on model accuracy.