Machine Learning Classification Methods Research Articles

Geographical Origin Traceability of Atractylodes macrocephala Koidz. Using Mass Spectrometry Data Fusion and Ensemble Learning

The use of data fusion based with multiple analytical techniques was investigated to provide the accurate geographical origin identification of Atractylodes macrocephala Koidz. (AMK). Inductively coupled plasma - mass spectrometry (ICP-MS), gas chromatography - mass spectrometry (GC-MS) and liquid chromatography - mass spectrometry (LC-MS) were used to characterize Hubei, Zhejiang, and Hunan production regions. After the implementation of data fusion, the ensemble learning method multi-forest joint network (MFJN) and classic machine learning methods were used to identify the AMK production regions. The MFJN based upon high-level data fusion distinguished AMK samples from different regions with the highest accuracy. The classification accurate rate of AMK in the prediction set was 95%, which was significantly better than the results obtained using twenty-five mineral element or nine bioactive component data sets. The results showed that mass spectrometry data fusion in combination with MFJN is suitable for the geographic origin determination of AMK and has potential to ensure this product’s fair trade.

Analyzing the effect of user‐generated content on studio performance: A combined approach

AbstractThis research explores how user‐generated content (UGC) influences the performance of studios. Our dataset included both structured and unstructured UGC, analyzed through a hybrid of statistical techniques and machine learning algorithms. The findings revealed a positive relationship between UGC and both box office revenues and stock market performance. Additionally, classical machine learning methods demonstrated exceptional capabilities in accurately classifying the data. The insights from this study offer valuable marketing strategies for studio marketing managers and studio executives in leveraging social networks effectively.

Enhancing HLA-B27 antigen detection: Leveraging machine learning algorithms for flow cytometric analysis.

As the association of human leukocyte antigen B27 (HLA-B27) with spondylarthropathies is widely known, HLA-B27 antigen expression is frequently identified using flow cytometric or other techniques. Because of the possibility of cross-reaction with off target antigens, such as HLA-B7, each flow cytometric technique applies a "gray zone" reserved for equivocal findings. Our aim was to use machine learning (ML) methods to classify such equivocal data as positive or negative. Equivocal samples (n = 99) were selected from samples submitted to our institution for clinical evaluation by HLA-B27 antigen testing. Samples were analyzed by flow cytometry and polymerase chain reaction. Features of histograms generated by flow cytometry were used to train and validate ML methods for classification as logistic regression (LR), decision tree (DT), random forest (RF) and light gradient boost method (GBM). All evaluated ML algorithms performed well, with high accuracy, sensitivity, specificity, as well as negative and positive predictive values. Although, gradient boost approaches are proposed as high performance methods; nevertheless, their effectiveness may be lower for smaller sample sizes. On our relatively smaller sample set, the random forest algorithm performed best (AUC: 0.92), but there was no statistically significant difference between the ML algorithms used. AUC values for light GBM, DT, and LR were 0.88, 0.89, 0.89, respectively. Implementing these algorithms into the process of HLA-B27 testing can reduce the number of uncertain, false negative or false positive cases, especially in laboratories where no genetic testing is available.

MetDIT: Transforming and Analyzing Clinical Metabolomics Data with Convolutional Neural Networks.

Clinical metabolomics is growing as an essential tool for precision medicine. However, classical machine learning algorithms struggle to comprehensively encode and analyze the metabolomics data due to their high dimensionality and complex intercorrelations. This article introduces a new method called MetDIT, designed to analyze intricate metabolomics data effectively using deep convolutional neural networks (CNN). MetDIT comprises two components: TransOmics and NetOmics. Since CNN models have difficulty in processing one-dimensional (1D) sequence data efficiently, we developed TransOmics, a framework that transforms sequence data into two-dimensional (2D) images while maintaining a one-to-one correspondence between the sequences and images. NetOmics, the second component, leverages a CNN architecture to extract more discriminative representations from the transformed samples. To overcome the overfitting due to the small sample size and class imbalance, we introduced a feature augmentation module (FAM) and a loss function to improve the model performance. Furthermore, we systematically optimized the model backbone and image resolution to balance the model parameters and computational costs. To demonstrate the performance of the proposed MetDIT, we conducted extensive experiments using three different clinical metabolomics data sets and achieved better classification performance than classical machine learning methods used in metabolomics, including Random Forest, SVM, XGBoost, and LightGBM. The source code is available at the GitHub repository at https://github.com/Li-OmicsLab/MetDIT, and the WebApp can be found at http://metdit.bioinformatics.vip/.

Comparing programming languages for data analytics: Accuracy of estimation in Python and R

AbstractSeveral open‐source programming languages, particularly R and Python, are utilized in industry and academia for statistical data analysis, data mining, and machine learning. While most commercial software programs and programming languages provide a single way to deliver a statistical procedure, open‐source programming languages have multiple libraries and packages offering many ways to complete the same analysis, often with varying results. Applying the same statistical method across these different libraries and packages can lead to entirely different solutions due to the differences in their implementations. Therefore, reliability and accuracy should be essential considerations when making library and package usage decisions while conducting statistical analysis using open source programming languages. Instead, most users take this for granted, assuming that their chosen libraries and packages produce accurate results for their statistical analysis. To this extent, this study assesses the estimation accuracy and reliability of Python and R's various libraries and packages by evaluating the univariate summary statistics, analysis of variance (ANOVA), and linear regression procedures using benchmarking data from the National Institutes of Standards and Technology (NIST). Further, experimental results are presented comparing machine learning methods for classification and regression. The libraries and packages assessed in this study include the stats package in R and Pandas, Statistics, NumPy, statsmodels, SciPy, statsmodels, scikit‐learn, and pingouin in Python. The results show that the stats package in R and statsmodels library in Python are reliable for univariate summary statistics. In contrast, Python's scikit‐learn library produces the most accurate results and is recommended for ANOVA. Among the libraries and packages assessed for linear regression, the results demonstrated that the stats package in R is more reliable, accurate, and flexible; thus, it is recommended for linear regression analysis. Further, we present results and recommendations for machine learning using R and Python.This article is categorized under: Algorithmic Development > Statistics Application Areas > Data Mining Software Tools

Can We Boost N-Glycopeptide Identification Confidence? Smart Collision Energy Choice Taking into Account Structure and Search Engine.

High confidence and reproducibility are still challenges in bottom-up mass spectrometric N-glycopeptide identification. The collision energy used in the MS/MS measurements and the database search engine used to identify the species are perhaps the two most decisive factors. We investigated how the structural features of N-glycopeptides and the choice of the search engine influence the optimal collision energy, delivering the highest identification confidence. We carried out LC-MS/MS measurements using a series of collision energies on a large set of N-glycopeptides with both the glycan and peptide part varied and studied the behavior of Byonic, pGlyco, and GlycoQuest scores. We found that search engines show a range of behavior between peptide-centric and glycan-centric, which manifests itself already in the dependence of optimal collision energy on m/z. Using classical statistical and machine learning methods, we revealed that peptide hydrophobicity, glycan and peptide masses, and the number of mobile protons also have significant and search-engine-dependent influence, as opposed to a series of other parameters we probed. We envisioned an MS/MS workflow making a smart collision energy choice based on online available features such as the hydrophobicity (described by retention time) and glycan mass (potentially available from a scout MS/MS). Our assessment suggests that this workflow can lead to a significant gain (up to 100%) in the identification confidence, particularly for low-scoring hits close to the filtering limit, which has the potential to enhance reproducibility of N-glycopeptide analyses. Data are available via MassIVE (MSV000093110).

A Comprehensive Systematic Review for Cardiovascular Disease using Machine Learning Techniques

The global upswing in cardiovascular disease (CVD) cases presents a critical challenge. While the ultimate goal remains elusive, improving CVD prediction accuracy is vital. Machine learning and deep learning are crucial for decoding complex health data, enhancing cardiac imaging, and predicting disease outcomes in clinical practice. This systematic literature review meticulously analyses CVD using machine learning techniques, with a particular emphasis on algorithms for classification and prediction. The metaanalysis covers 343 references from 2020 to November 2023, preceding a thorough examination of 65 selected references. Acknowledging current hurdles in CVD classification methods that impede practical use, this systematic literature review (SLR) is conducted. The study provides valuable insights for researchers and healthcare professionals, facilitating the integration of clinical applications in machine learning settings related to CVD. It also aids in promptly identifying potential threats and implementing precautionary measures. The study also recognizes prevalent classical machine learning methods, emphasizing their clinically relevant diagnostic outcomes. Deliberating on current trends, algorithms, and potential areas for future research offers a comprehensive insight into the present state of affairs.

Applying machine learning to predict reproductive condition in fish

Knowledge of reproductive traits in exploited marine populations is crucial for their management and conservation. The maturity status in fish is usually assigned by traditional methods such as macroscopy and histology. Macroscopic analysis is the assessing of maturity stages by naked eye and usually introduces large amount of error. In contrast, histology is the most accurate method for maturity staging but is expensive and unavailable for many stocks worldwide. Here, we use the Random Forest (RF) machine learning method for classification of reproductive condition in fish, using the extensive data from Chilean hake (Merluccius gayi gayi). Gonads randomly collected from commercial industrial and acoustic surveys were classified as immature, mature-active and mature-inactive. A classifier for these three maturity classes was fitted using RFs, with the continuous covariates total length (TL), gonadosomatic index (GSI), condition factor (Krel), latitude, longitude, and depth, along with month as a factor variable. The RF model showed high accuracy (>82%) and high proportion of agreement (>71%) compared to histology, with an OOB error rate lower than 15%. GSI and TL were the most important variables for predicting the reproductive condition in Chilean hake, and to lesser extent, depth when using survey data. The application of the RF shows a promising tool for assigning maturity stages in fishes when covariates are available, and also to improve the accuracy of maturity classification when only macroscopic staging is available.

Electron-vibrational renormalization in fullerenes through ab initio and machine learning methods.

The effect of nuclear vibrations on the electronic eigenvalues and the HOMO-LUMO gap is known for several kinds of carbon-based materials, like diamond, diamondoids, carbon nanoclusters, carbon nanotubes and others, like hydrogen-terminated oligoynes and polyyne. However, it has not been widely analysed in another remarkable kind which presents both theoretical and technological interest: fullerenes. In this article we present the study of the HOMO, LUMO and gap renormalizations due to zero-point motion of a relatively large number (163) of fullerenes and fullerene derivatives. We have calculated this renormalization using density-functional theory with the frozen-phonon method, finding that it is non-negligible (above 0.1 eV) for systems with relevant technological applications in photovoltaics and that the strength of the renormalization increases with the size of the gap. In addition, we have applied machine learning methods for classification and regression of the renormalizations, finding that they can be approximately predicted using the output of a computationally cheap ground state calculation. Our conclusions are supported by recent research in other systems.

Analysis of uncertainty of neural fingerprint-based models.

Machine learning has gained popularity for predicting molecular properties based on molecular structure. This study explores the uncertainty estimates of neural fingerprint-based models by comparing pure graph neural networks (GNN) to classical machine learning algorithms combined with neural fingerprints. We investigate the advantage of extracting the neural fingerprint from the GNN and integrating it into a method known for producing better-calibrated probability estimates. Comparisons are made using three classical machine learning methods and the Chemprop model, considering different molecular representations and calibration techniques. We utilize 19 datasets from Toxcast, reflecting real-world scenarios with balanced accuracies ranging from 0.6 to 0.8. Results demonstrate that neural fingerprints combined with classical machine learning methods exhibit a slight decrease in prediction performance compared to the native Chemprop model. However, these models provide significantly improved uncertainty estimates. Notably, uncertainty estimates of neural fingerprint-based methods remain relatively robust for molecules dissimilar to the training set. This suggests that methods like random forest with neural fingerprints can deliver strong prediction performance and reliable uncertainty estimates. When considering both performance and uncertainty, the calibrated Chemprop model and the combination of neural fingerprints with random forest or support vector classifier (SVC) yield comparable results. Surprisingly, the SVC method shows promising performance when combined with neural or count fingerprints. These findings are particularly relevant in real-world industrial projects where accurate predictions and reliable uncertainty estimates are crucial.

Synthetic data generation methods for training neural networks in the task of segmenting the level of crop nitrogen status on UAV images of agricultural fields

This study is devoted to the automatization of the image masks’ construction of large sized agricultural objects in precision farming tasks for training neural network methods for crop’s nitrogen status analysis using georeferenced images. The scientific direction is extremely relevant because it allows to automate and replace the manual process of data labeling, significantly reducing the cost of preparing training samples. In the paper, four new synthetic data generation methods are proposed for training neural networks aimed at UAV image segmentation by the level of crop nitrogen supply on an agricultural field. In particular, the paper gives a description of synthetic data generation algorithms based on nitrogen covering with lines, parabolas, and areas. Experiments were carried out to test and evaluate the quality of these algorithms using eight modern image segmentation methods: two classical machine learning methods (Random Forest and XGBoost), four convolutional neural network methods based on U-Net architecture, and two transformers (TransUnet and UnetR). The results showed that two algorithms based on areas gave the best accuracy for convolutional neural networks and transformers — 98–100 %. Classical machine learning methods showed very low values for all quality metrics — 27–44 %.

Insulator Defect Recognition Based on Vision Big‐Model Transfer Learning and Stochastic Configuration Network

Insulator faults are an important factor in causing outages and accidents in power transmission lines. In response to problems related to inefficient insulator positioning, limited robustness of insulator defect feature extraction methods, and the scarcity of defective insulator samples leading to poor classifier generalization, a method for insulator defect detection and recognition based on vision big‐model transfer learning and a stochastic configuration network (SCN) is proposed. First, data augmentation methods, such as Mosaic and Mixup, are employed to mitigate overfitting in the YOLOv7 network. Second, StyleGanv3 adversarial generative networks are used to augment the dataset of defective insulators, which enhances dataset diversity. Third, a vision big‐model transfer learning method based on DINOv2 is introduced to extract features from insulator images. Finally, an SCN classifier is used to determine the status of insulators. Experimental results demonstrate that the applied data augmentation methods effectively mitigate overfitting. YOLOv7 accurately detects insulator positions, and the use of the DINOv2 feature extraction method increases the accuracy of insulator defect recognition by 28.6%. Compared with machine learning classification methods, the SCN classifier achieves the highest accuracy improvement of 17.4%. The proposed method effectively detects insulator positions and recognizes insulator defects.

Implementation of an intelligence-based framework for anomaly detection on the demand-side of sustainable compressed air systems

The implementation of intelligent techniques produces good results in automating fault finding and predicting future outcomes. These approaches have been on the increase in the past years, especially so to detect faults within Compressed Air Systems (CASs). With the use of intelligent techniques, one could minimise the manual and time-consuming aspect of CAS maintenance, improve the environmental impact of the system, while minimising downtime. This paper proposes a general framework for the implementation of intelligent analysis techniques within a real-world system. Such an approach has been implemented on the demand-side of a CAS. In literature, no open datasets are available for use by artificial intelligence models. Hence, as part of this research, a fault generating and monitoring system has been connected to an existing production machine in a manufacturing site to collect the required data. Two classification machine learning methods were implemented and compared across a number of performance metrics. Both the general framework, and its implementation, provide a stepping stone in integrating smart systems with real-time intelligent data analytics for the demand-side of a CAS. These systems would provide a sustainable CAS operation through the effective detection of anomalies and their timely repair.

Deep learning outperforms classical machine learning methods in pediatric brain tumor classification through mass spectra

Pediatric brain tumors are the most common cause of death among all childhood cancers and surgical resection usually is the first step in disease management. During surgery, it is important to perform safe gross resection of tumors, retaining as much brain tissue as possible. Therefore, appropriate resection margin delineation is extremely relevant.Currently available methods for tissue analysis have limited precision, are time-consuming, and often require multiple invasive procedures. Our main goal is to test whether machine learning techniques are capable of classifying the pediatric brain tissue chemical profile generated by DESI-MSI, which is mainly lipidic, into normal or abnormal tissue and into low- and high-grade malignancy subareas within each sample.Our experiments show that deep learning methods outperform classical machine learning methods in the task of classifying brain tissue from DESI-MSI mass spectra, both in normal versus abnormal tissue, and, for malignant tissues, in low-grade versus high-grade malignancy.Our conclusion are based on the analysis of 34,870 annotated spectra, obtained from the neoplastic and non-neoplastic microanatomical stratification of individual samples from 116 pediatric patients who underwent brain tumor surgical resection at the Boldrini Children’s Center between 2000 and 2020. Support Vector Machines, Random, Forests, and Least Absolute Shrinkage and Selection Operator (LASSO) were among the classical machine learning techniques evaluated.

Detection of cardiovascular disease using explainable artificial intelligence and gut microbiota data

Purpose:Gut microbiota are defined as the microbial population of the intestines. They include various types of bacteria which can influence and predict the existence or onset of some specific diseases. Therefore, it is a common practice in medicine to analyze the gut microbiota for diagnostic purposes by analyzing certain measurable biochemical features associated with the disease under investigation. However, the evaluation of all the data collected from the gut microbiota is a labor-intensive process. Artificial Intelligence (AI) may be a helpful tool to identify the hidden patterns in gut microbiota for the detection of disease and other classification problems. Methods:In this study, we propose a deep neural model based on a one-dimensional convolutional neural network (1D-CNN) to detect cardiovascular disease using bacterial taxonomy and OTU (Operational Taxonomic Unit) table data. The developed AI method is compared to classical machine learning algorithms, regression, boosting algorithms and a deep model, Tabular Network (TabNet), developed for tabular data and obtained outperforming classification results. Results:According to AUC (Area Under Curve) values, boosting and regression methods outperformed the classical machine learning methods. However, the highest value of 97.09 AUC was obtained with the developed 1D-CNN model by using bacterial taxonomy data even with less then expected number of samples. Using explainable AI, nine bacteria were identified which the models find important for classification. Conclusion:The proposed method is robust and well adapted to taxonomy data in tabular form. It can be easily adapted to detect other diseases by using taxonomy data. The study also investigated the effect on barcode sequence for the classification, but the result showed that barcode sequences do not contribute to the bacterial taxonomy data for the estimation of CVD disease.

A model for identifying project sprint tasks based on their description

The subject of research in this article is the identification of project sprint tasks. The purpose of the article is to find approaches to reducing the risks of not fulfilling sprint tasks. The article solves the following tasks: analyzing research on the classification and visualization of project tasks, developing an algorithm that can automatically classify text descriptions of sprint tasks, collecting and preparing a training sample of text descriptions of sprint tasks for training and testing the classification model, applying natural language processing methods to improve classification and ensure the accuracy of the results, validating the model on real data to assess the efficiency and accuracy of classification, and analyzing the results. The following methods have been used: machine learning methods for classification, text vectorization methods, methods for classifying text descriptions, natural language processing methods, methods for semantic analysis of task description text, methods for processing expert opinions. The following results were obtained: a comprehensive approach to using machine learning algorithms, including the collection and processing of textual descriptions of tasks, for classification and involvement of expert opinions to improve the quality of task perception by the project team. Text expressions were classified based on the Bayesian classifier and neural classifiers. A visual representation of the data was implemented. Semantic analysis of the text of the description and title of the tasks was performed. Data markup was obtained to classify the quality of the wording, which was performed by a team of experts. To measure the reliability of the obtained expert assessments, we calculated Cohen's kappa coefficient for each pair of markers. According to the experimental results, the accuracy of the Bayesian classifier is 70%. For the classifier based on deep learning, a neural network for binary classification based on the transformer architecture was selected. The neural network was trained using the Python programming language and deep learning frameworks. The result is a classifier that gives an accuracy score of 83% on a test dataset, which is a good result for a small dataset and data with conflicting labels. Conclusions: The analysis of textual data confirms that the existing data in the tracking system is incomplete and contains abbreviations, conventions, and slang. The results show that the assessment of the quality of the wording is determined by the level of expert knowledge of the specifics and context of the project, while increasing the number of experts has almost no effect on the result. In further research, it is recommended to test the hypothesis that the effectiveness of the classifier depends on the specific project and the use of unsupervised learning methods for the task of identifying the quality of formulations.

Machine Learning Algorithm: Texture Analysis in CNO and Application in Distinguishing CNO and Bone Marrow Growth-Related Changes on Whole-Body MRI.

The purpose of this study is to analyze the texture characteristics of chronic non-bacterial osteomyelitis (CNO) bone lesions, identified as areas of altered signal intensity on short tau inversion recovery (STIR) sequences, and to distinguish them from bone marrow growth-related changes through Machine Learning (ML) and Deep Learning (DL) analysis. We included a group of 66 patients with confirmed diagnosis of CNO and a group of 28 patients with suspected extra-skeletal systemic disease. All examinations were performed on a 1.5 T MRI scanner. Using the opensource 3D Slicer software version 4.10.2, the ROIs on CNO lesions and on the red bone marrow were sampled. Texture analysis (TA) was carried out using Pyradiomics. We applied an optimization search grid algorithm on nine classic ML classifiers and a Deep Learning (DL) Neural Network (NN). The model's performance was evaluated using Accuracy (ACC), AUC-ROC curves, F1-score, Positive Predictive Value (PPV), Mean Absolute Error (MAE) and Root-Mean-Square Error (RMSE). Furthermore, we used Shapley additive explanations to gain insight into the behavior of the prediction model. Most predictive characteristics were selected by Boruta algorithm for each combination of ROI sequences for the characterization and classification of the two types of signal hyperintensity. The overall best classification result was obtained by the NN with ACC = 0.91, AUC = 0.93 with 95% CI 0.91-0.94, F1-score = 0.94 and PPV = 93.8%. Between classic ML methods, ensemble learners showed high model performance; specifically, the best-performing classifier was the Stack (ST) with ACC = 0.85, AUC = 0.81 with 95% CI 0.8-0.84, F1-score = 0.9, PPV = 90%. Our results show the potential of ML methods in discerning edema-like lesions, in particular by distinguishing CNO lesions from hematopoietic bone marrow changes in a pediatric population. The Neural Network showed the overall best results, while a Stacking classifier, based on Gradient Boosting and Random Forest as principal estimators and Logistic Regressor as final estimator, achieved the best results between the other ML methods.

Hybrid machine learning method for classification and recommendation of vector-borne disease

<p>Vector-borne diseases (VBD) are a class of infectious illnesses that are transmitted to humans and animals through the bites of arthropod vectors, such as mosquitoes, ticks, and fleas. These diseases are caused by a variety of pathogens, including bacteria, viruses, and parasites, and are a significant global public health concern. Vector-borne diseases are prevalent in many parts of the world, particularly in tropical and subtropical regions, where the vectors thrive. This research has contributed by constructing a hybrid machine learning based prediction model, which helps to discover patients who are infected by vector-borne disease at an earlier stage and also helps with the categorization and diagnosis of severe vector-borne disease. The model that has been proposed is made up of units: data conversion, data preprocessing, normalization, extraction of feature, splitting of dataset, and classification and prediction unit. The fact that the suggested prediction model is capable of identifying vector-borne disease in its early phases as well as categorizing the kind of disease using the medical report of a sufferer is one of the innovative aspects of the model. The 7 distinct conventional machine learning and single hybrid machine learning (HML) are applied for classification and Recurrent Neural Network (RNN) based reinforcement learning are utilized for recommendation. In order to evaluate the effectiveness of the system that’s been proposed, a number of tests were carried out. A dataset consisting of 1539 different cases of a disease transmitted by vectors has been collected. The 11 common vector-borne diseases namely malaria, dengue, Japanese encephalitis, kala-azar and chikungunya were taken for experimental evaluation. The performance accuracy of the proposed prediction model has been measured at 98.76%, which assists the healthcare team in making decisions on a timely basis and ultimately helps to save the patient’s lives. The final phase system provides the recommendation for those classifiers resulting in four different classes such as normal, mild, moderate and severe respectively. The recommendation is also demonstrating future direction for cure of vector borne disease.</p>

Analysis of Fault Ride through the Improvement of PV Power Plant Based on Capacitive Bridge Fault Current Limiter Using Machine Learning

Photovoltaic power plant (PVPP) has increased importance among renewable energy sources due to their ability to be connected more easily to a modern power grid. However, the reliability and stability operation of a grid-connected PVPP system is very important to ensure even during grid faults. In this study, a capacitive bridge fault current limiter (CBFCL) using a machine learning (ML) method is applied to enhance the fault ride-through (FRT) capability of a grid-connected PVPP system. Three different protection methods called DC chopper, CBFCL, and DC chopper + CBFCL are designed to prevent the harmful effects of overcurrent that occurs during grid faults to protect the grid-connected PVPP system. The ML algorithm can be trained on historical data to predict optimum control parameters based on real-time conditions such as normal and fault operations of the grid-connected PVPP system. An ensemble classification algorithm has the best results among the four classification algorithms in machine learning methods. The ensemble classification algorithm is separately implemented into the control systems of three protection strategies. Bagged Trees and Subspace KNN classifiers in ensemble classification methods have obtained an impressive accuracy of 98% in ML classification methods. The simulation results illustrate that the DC chopper + CBFCL based ensemble provides the best protection for the grid-connected PVPP system compared to other protection systems.

Optimasi Gaussian Naïve Bayes dengan Hyperparameter Tuning dan Univariate Feature Selection dalam Prediksi Cuaca

The importance of conducting weather prediction research is due to the significant influence of weather changes on daily life. The purpose of this study is to apply an optimal machine-learning classification method for weather prediction. The method used is the Gaussian Naïve Bayes model, which has been optimized using Univariate Feature Selection ANOVA-f test and Hyperparameter Tuning GridsearchCV techniques. The data used consists of 6454 daily weather data in Palembang City. There are 5 tests on the Gaussian Naïve Bayes model before and after optimization. The research results show that the optimization of the model successfully improves the performance in weather prediction. The highest accuracy result after optimization reaches 98.33% with 644 test data, an improvement from the pre-optimization accuracy of only 96.95%. Before optimization, the predictions for weather conditions such as sunny, cloudy/rainy, light rain, and heavy rain match the actual data. However, there were 20 prediction errors when dealing with data that should represent very heavy rain conditions. After optimization, the number of prediction errors for the very heavy rain data was reduced to seven. The optimization approach used in this research helps find the most suitable parameter combinations and eliminates irrelevant features, allowing the model to consider only significant features in weather p