F-score Values Research Articles

Unsupervised defect detection for solar photovoltaic cells based on convolutional autoencoder

ABSTRACT Solar photovoltaic (PV) cells are inevitably subject to defects during the production process, affecting their power generation efficiency and life. Electroluminescence (EL) imaging is the mainstream non-destructive method for PV cell defect detection. Aiming at PV cell EL images, an unsupervised defect detection method was proposed. Specifically, an unsupervised convolutional autoencoder (CAE), the scale structure perception convolutional autoencoder (SSP_CAE), was constructed, whose Squeeze-and-Excitation Attention (SE Attention) and skip connections avoid the blurring of image structure information and the loss of pixel-level details in the encoding and decoding process. Furthermore, to balance the global and local information of the image, a scale perception loss function called SP_SSIM was proposed for model training. The defect segmentation was achieved by using Otsu thresholding method to binarize the obtained Mean Absolute Error (MSE) residual heat image in the testing stage of the model. Finally, the experiments were performed on the test dataset and the experimental results showed that the proposed SSP_CAE can effectively detect PV cell defects. The experimentally obtained defect detection performance metrics Precision, Recall, IoU, F1-score and AUROC values were 0.739, 0.886, 0.723, 0.764 and 0.841, respectively. Compared with other classical methods, the proposed SSP_CAE had a better comprehensive performance for defect detection.

Multi-parametric MRI combined with radiomics for the diagnosis and grading of endometrial fibrosis.

To evaluate the application of multi-parametric MRI (MP-MRI) combined with radiomics in diagnosing and grading endometrial fibrosis (EF). A total of 74 patients with severe endometrial fibrosis (SEF), 41 patients with mild to moderate fibrosis (MMEF) confirmed by hysteroscopy, and 40 healthy women of reproductive age were prospectively enrolled. The enrolled data were randomly stratified and divided into a train set (108 cases: 28 healthy women, 29 with MMEF, and 51 with SEF) and a test set (47 cases: 12 healthy women, 12 MMEF and 23 SEF) at a ratio of 7:3. All participants underwent T2 and DWI sequence scans. By freely delineating the volume of interest (VOI) of the endometrium in three subgroups, radiomic features were extracted and selected. Two feature selection methods and four machine learning (ML) classifiers were combined in pairs to establish five prediction models [model1 (T2 + ADC + clinical data), model2 (T2 + ADC), model3 (T2), model4 (ADC), and model5 (clinical data)], resulting in a total of 40 classification models. The predictive performance of all models was evaluated using the area under the curve (AUC), F1-score, and accuracy (ACC). The "UFS-LR" model, which combined unsupervised feature selection (UFS) with the logistic regression (LR) classifier, performed the best, with an average AUC of 0.92 on the test set. Among the five models constructed via UFS-LR, model1 exhibited the best performance, with average AUC, F1-score, and ACC values of 0.92, 0.80, and 0.81, respectively. T2-related features were the most significant in distinguishing fibrosis levels, with T2_wavelet-LLL_gldm_DependenceVariance being the most important characteristic among them. MP-MRI radiomics analysis using ML has excellent performance in grading EF. This approach is non-invasive and has the potential to reduce the reliance on hysteroscopy.

PENERAPAN ALGORITMA NAIVE BAYES DENGAN CHI-SQUARE UNTUK KLASIFIKASI SPAM EMAIL BERBASIS KATA DAN FREKUENSI

Email has become an essential communication tool in everyday life. However, the ease of its use is also exploited by irresponsible parties to spread spam. This research aims to implement the Naive Bayes algorithm with Chi-square in classifying spam emails based on words and frequency. The dataset used in this research consists of 153 data. This data was processed using the classification method using the Naive Bayes algorithm with Chi-square through the Knowledge Discovery in Databases (KDD) process. The results show that the accuracy value is 81.00%, the precision value is 100%, the recall value is 65%, and the F1-score value is 79% using Naive Bayes with Chi-square. Furthermore, the evaluation results using the ROC curve show that the AUC value reaches 0.91, which is categorized as very good. This research shows that the Naive Bayes algorithm with Chi-square is successful in classifying spam emails based on words and frequency.

Automatic Vertical Root Fracture Detection on Intraoral Periapical Radiographs With Artificial Intelligence-Based Image Enhancement.

To explore transfer learning (TL) techniques for enhancing vertical root fracture (VRF) diagnosis accuracy and to assess the impact of artificial intelligence (AI) on image enhancement for VRF detection on both extracted teeth images and intraoral images taken from patients. A dataset of 378 intraoral periapical radiographs comprising 195 teeth with fractures and 183 teeth without fractures serving as controls was included. DenseNet, ConvNext, Inception121, and MobileNetV2 were employed with model fusion. Prior to evaluation, Particle Swarm Optimization (PSO) and Deep Learning (DL) image enhancement were applied. Performance assessment included accuracy rate, precision, recall, F1-score, AUC, and kappa values. Intra- and inter-observer agreement, according to the Gold Standard (GS), were assessed using ICC and t-tests. Statistical significance was set at p < 0.05. The DenseNet + Inception fusion model achieved the highest accuracy rate of 0.80, with commendable recall, F1-score, and AUC values, supported by precision (0.81) and kappa (0.60) values. Molar tooth examination yielded an accuracy rate, precision, recall, and F1-score of 0.80, with an AUC of 0.84 and kappa of 0.60. For premolar teeth, the fusion network showed an accuracy rate of 0.78, an AUC of 0.78, and notable metrics, including F1-score (0.80), recall (0.85), precision (0.71), and kappa (0.55). ICC results demonstrated acceptable agreement (≥ 0.57 for molars, ≥ 0.52 for premolars). TL methods have demonstrated significant potential in enhancing diagnostic accuracy for VRFs in radiographic imaging. TL is emerging as a valuable tool in the development of robust, automated diagnostic systems for VRF identification, ultimately supporting clinicians in delivering more accurate diagnoses.

Research on Innovative Apple Grading Technology Driven by Intelligent Vision and Machine Learning.

In the domain of food science, apple grading holds significant research value and application potential. Currently, apple grading predominantly relies on manual methods, which present challenges such as low production efficiency and high subjectivity. This study marks the first integration of advanced computer vision, image processing, and machine learning technologies to design an innovative automated apple grading system. The system aims to reduce human interference and enhance grading efficiency and accuracy. A lightweight detection algorithm, FDNet-p, was developed to capture stem features, and a strategy for auxiliary positioning was designed for image acquisition. An improved DPC-AWKNN segmentation algorithm is proposed for segmenting the apple body. Image processing techniques are employed to extract apple features, such as color, shape, and diameter, culminating in the development of an intelligent apple grading model using the GBDT algorithm. Experimental results demonstrate that, in stem detection tasks, the lightweight FDNet-p model exhibits superior performance compared to various detection models, achieving an mAP@0.5 of 96.6%, with a GFLOPs of 3.4 and a model size of just 2.5 MB. In apple grading experiments, the GBDT grading model achieved the best comprehensive performance among classification models, with weighted Jacard Score, Precision, Recall, and F1 Score values of 0.9506, 0.9196, 0.9683, and 0.9513, respectively. The proposed stem detection and apple body classification models provide innovative solutions for detection and classification tasks in automated fruit grading, offering a comprehensive and replicable research framework for standardizing image processing and feature extraction for apples and similar spherical fruit bodies.

Simulation Research on Data Acquisition and Supervisory Control System of Distribution Network Based on Vector Regression Model

With the gradual maturity of deep learning and the Internet of Things, a smart grid is urgently needed to be established. For the intellectualization of the power grid, non-intrusive load monitoring (NILM) technology is the key step of intellectualization. The traditional load monitoring method is an invasive scheme, in which monitoring sensors are usually installed at the output end of each grid load. This method requires a lot of manpower and material resources in terms of equipment installation and maintenance, which is difficult to sustain. Therefore, monitoring electrical appliances are installed at the entrance of the distribution network to decompose the classification and operation status of the individual electrical appliances in the grid. Aiming at the problem of long-term multi-state electrical apparatus monitoring, this paper mainly optimizes the two optimal deep learning models Seq2Seq and WindowGRU based on activation function optimization, regular function optimization and neural network structure optimization. PReLU, Leaky-ReLU and ELU are used and tested. For the optimization based on regular functions, the value of F1-score is 0.9700, which is 0.1534 more than the optimization algorithm Seq2SeqLR in this paper. Meanwhile, it can also reach 0.8100 for other devices, which is 0.2373 more than the nilmTCN designed in this paper. The Recall value of the identification is improved to 0.6673, which is significantly higher than that of other models. It is proved that the research contents of this paper can improve the effective analysis of load energy consumption data, and can minimize unnecessary energy consumption to achieve the purpose of saving electricity.

Integrating machine learning and nano-QSAR models to predict the oxidative stress potential caused by single and mixed carbon nanomaterials in algal cells.

In silico methods are increasingly important in predicting the ecotoxicity of engineered nanomaterials (ENMs), encompassing both individual and mixture toxicity predictions. It is widely recognized that ENMs trigger oxidative stress effects by generating intracellular reactive oxygen species (ROS), serving as a key mechanism in their cytotoxicity studies. However, existing in silico methods still face significant challenges in predicting the oxidative stress effects induced by ENMs. Herein, we utilized laboratory-derived toxicity data and machine learning methods to develop quantitative nanostructure-activity relationship (nano-QSAR) classification and regression models, aiming to predict the oxidative stress effects of five carbon nanomaterials (fullerene, graphene, graphene oxide, single-walled carbon nanotubes, and multi-walled carbon nanotubes) and their binary mixtures on Scenedesmus obliquus cells. We constructed five nano-QSAR classification models by combining zeta potential (ζP) with the C4.5 decision tree, support vector machine, artificial neural network, naive Bayes, and K-nearest neighbor algorithms. Moreover, we constructed three classification models by integrating the features including ζP, hydrodynamic diameter (DH), and specific surface area (SSA) with the logistic regression, random forest, and Adaboost algorithms. The Accuracy, Recall, Precision and harmonic mean of Precision and Recall (F1-score) values of these models were all higher than 0.600, indicating an excellent performance in distinguishing whether CNMs have the potential to generate ROS. In addition, using the ζP, DH, and SSA descriptors, we combined decision tree regression, random forest regression, gradient boosting, and the Adaboost algorithm, and successfully constructed four nano-QSAR regression models with applicable application domains (all training and testing data points lie within 95% confidence intervals), goodness-of-fit (Rtrain2 ≥ 0.850), and robustness (cross-validation R2 ≥ 0.650) as well as predictive power (Rtest2 ≥ 0.610). The method developed would establish a fundamental basis for more precise evaluations of ecological risks posed by these materials from a mechanistic standpoint.

E-mail Classifications Based on Deep Learning Techniques

Email types sorting is one of the most important tasks in current information systems with the purpose to improve the security of messages, allowing for their sorting into different types. This paper aims at studying the Convolution Neural Network and Long Short-Term Memory (CNN-LSTM), Convolution Neural Network and Gated Recurrent Unit (CNN-GRU) and Long Short-Term Memory (LSTM) deep learning models for the classification of emails into categories such as “Normal”, “Fraudulent”, “Harassment” and “Suspicious”. The architecture of each model is discussed and the results of the models’ performance by testing on labelled emails are presented. Evaluation outcomes show substantial gains in precision and throughput to conventional approaches hence inferring to the efficiency of these proposed models for automated email filtration and content evaluation. Last but not the least, the performance of the classification algorithms is evaluated with the help of parameters like Accuracy, precision, recall and F1-Score. From the experiment, the models found out that CNN-LSTM, together with the Term Frequency and Inverse Document Frequency (TF-IDF) feature extraction yielded the highest accuracy. The accuracy, precision, recall and f1-score values are 99. 348%, 99. 5%, 99. 3%, and 99. 2%, respectively.

Uncovering Import Document Fraud: Leveraging the Deep Learning Approach

The scrutiny, identification, and verification of handwritten signatures on Certificates of Origin (C/O) is a critical task for customs authorities in preventing trade fraud. This task remains a significant challenge due to limitations in manpower and the need for manual verification amidst a vast volume of documents. Deep learning (DL) algorithms offer a valuable solution to address this issue. This paper deploys a Siamese Neural Network (SNN) model to assist customs officials in identifying and verifying handwritten signatures on C/O. The results demonstrate the superior performance of the SNN model over conventional Convolutional Neural Network (CNN) models and Machine Learning (ML) models, with accuracy, precision, recall, F1-score, and AUC values of 0.943, 0.912, 0.899, 0.905, and 0.919, respectively. The paper also provides in-depth analyses of omission cases and suggests applications of the model to support the work of customs officials.

Enhancing Multiclass Network Intrusion Detection Systems Using Continuous Wavelet Transform on Network Traffic

Network systems are susceptible to cyberattacks, which motivates attackers to exploit their vulnerabilities. Scanning network traffic to identify malicious activity is becoming a trend in the cybersecurity domain to mitigate the negative effects of intruders. Network intrusion detection systems (NIDS) are widely recognized as essential tools against cyberattacks. However, there is a need to go beyond designing traditional NIDS, which are preferred to be used with binary classification, towards designing multiclass network intrusion detection systems (MNIDS) to predict the cyberattack category. This, indeed, assists in understanding cyberattack behavior, which mitigates their effects quickly. Machine learning models, including conventional and deep learning, have been widely employed in the design of MNIDS. However, MNIDS based on machine learning can face challenges in predicting the category of cyberattack, especially with complex data that has a large number of categories. Thus, this paper proposes an enhanced MNIDS by exploiting the power of integrating continuous wavelet transform (CWT) with machine learning models to increase the accuracy of predicting cyberattacks in network traffic. This is due to the fact that CWT is considered as an effective method for feature extraction. The experimental results emphasize that using CWT with machine learning models improves the classification performance of MNIDS by up to 3.36% in overall accuracy. Additionally, it enhances the F1-score value in up to 40% of the total classes using the proposed model.

Improve unsupervised Learning-based landslides detection by band ratio processing of RGB optical images: a case study on rainfall-induced landslide clusters

Automatic detection of the heavy rainfall-induced landslide clusters based on optical remote sensing images had low accuracy due to the interference of many features. This work proposes a method to improve the accuracy of landslide identification based on change detection of pre- and post-disaster satellite images. Firstly, by band ratio preprocessing, the interference features on landslide detection are eliminated. Then, by further optimizing the relevant parameters, the threshold-based Image Difference and Principal Component Analysis-based K-means unsupervised learning classification methods are used to perform change detection and landslide cluster identification. Finally, the causes of missing and errors in landslide detection by using the two methods are analyzed and discussed. The results show that band preprocessing of remote sensing images can significantly weaken the interference features of human activities on landslide detection. The verification metrics of the two methods for detecting landslides, except for a slight decrease in precision, recall, F1-score and Kappa values have all increased significantly. After band ratio preprocessing and parameters optimization, both change detection methods have significantly improved the performance of the rainfall-induced landslide clusters detection, with effects close to those of deep learning methods. These findings can provide a reference for landslide detection using optical remote-sensing images.

Implementasi Convolutional Neural Network untuk Klasifikasi Ekspresi Wajah Penonton Stand Up Comedy

Stand Up Comedy is one of the growing entertainment industries in Indonesia. The quality of the joke of the stand up comedian can be seen from the audience's response in the form of laughter and facial expressions. But it is very difficult to determine the audience's laughter level manually. Therefore, a tool is needed to measure the audience's laughter level more objectively. In this study, a machine learning model was designed and developed using a convolutional neural network to analyze the facial expressions of the stand-up comedian audience. The research stages include dataset collection, data preprocessing, model design, model training and model testing. The model was tested using artificial data from crowd emotion and live data from video crowd viewers. The results of this study show that the model can classify facial expressions in image and video data well. The model has an overall validation accuracy value of 72.42% with precision, recall, and F1-score values of 0.7341, 0.7309, and 0.7294, respectively, and a testing accuracy of 83.28% with precision, recall, and F1-score values of 0.8477, 0.8329, and 0.8297, respectively.

An Adaptive Denoising Method for Photon-Counting LiDAR Point Clouds: Application in Intertidal Zones

The intertidal zone, as a dynamic ecosystem at the interface of land and sea, plays a critical role in environmental protection and disaster mitigation. The Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) is equipped with the Advanced Topographic Laser Altimeter System (ATLAS) with the ability to penetrate the water bodies, enabling its use for bathymetric measurements. However, the complex land cover types and frequent environmental changes in intertidal zones pose significant challenges for precise measurement and dynamic monitoring. In an effort to address the denoising challenges of ICESat-2 photon point cloud data in such complex environments, this study proposes an adaptive photon denoising method that is capable of dynamically adjusting the denoising strategy for different types of photon data. ATL03 data from four typical intertidal zones were selected for denoising experiments. The results indicated that the proposed adaptive denoising method achieved average recall, precision, and F-score values of 0.9885, 0.9927, and 0.9906, respectively, demonstrating excellent denoising performance and stability. This method provides an effective data processing approach for high-precision monitoring of intertidal zone topography.

Building Damage Detection Using Deep Learning Architecture with Satellite Images: The Case of the 6 February 2023 Kahramanmaraş Earthquake

Turkey is located in a region with a high density of fault lines, which makes it susceptible to a significant earthquake risk. The Kahramanmaraş earthquake on February 6, 2023, was one of the most devastating in recent years, causing extensive damage and loss. This study aims to support post-disaster rapid response and rescue operations by using deep learning techniques to detect and classify damaged and intact buildings from satellite images. Satellite images of the Kahramanmaraş and Antakya regions, with a resolution of 8192x4537, were obtained via Google Earth Pro. The images were labeled as damaged or undamaged using the Labelme editor, which generated JSON format files for the labeled images. Using Google Colab, the JSON files and unlabeled images were merged, and buildings were cropped and categorized into two classes: damaged and undamaged. As a preprocessing step, interpolation was applied, resulting in 2211 images with a size of 128x128. A Convolutional Neural Network [2] algorithm was created using TensorFlow, a Python library, via Google Colab. The performance metrics, including accuracy, loss, F1 score, ROC curve, precision, recall, and confusion matrix values, were compared based on the experiments.

Classifying Alzheimer's Disease Using a Finite Basis PhysicsNeural Network.

The disease amyloid plaques, neurofibrillary tangles, synaptic dysfunction, and neuronal death gradually accumulate throughout Alzheimer's disease (AD), resulting in cognitive decline and functional disability. The challenges of dataset quality, interpretability, ethical integration, population variety, and picture standardization must be addressed using deep learning for the functional magnetic resonance imaging (MRI) classification of AD in order to guarantee a trustworthy and practical therapeutic application. In this manuscript Classifying AD using a finite basis physics neural network (CAD-FBPINN) is proposed. Initially, images are collected from AD Neuroimaging Initiative (ADNI) dataset. The images are fed to Pre-processing segment. During the preprocessing phase the reverse lognormal Kalman filter (RLKF) is used to enhance the input images. Then the preprocessed images are given to the feature extraction process. Feature extraction is done by Newton-time-extracting wavelet transform (NTEWT), which is used to extract the statistical features such as the mean, kurtosis, and skewness. Finally the features extracted are given to FBPINNs for Classifying AD such as early mild cognitive impairment (EMCI), AD, mild cognitive impairment (MCI), late mild cognitive impairment (LMCI), normal control (NC), and subjective memory complaints (SMCs). In General, FBPINN does not express adapting optimization strategies to determine optimal factors to ensure correct AD classification. Hence, sea-horse optimization algorithm (SHOA) to optimize FBPINN, which accurately classifies AD. The proposed technique implemented in python and efficacy of the CAD-FBPINN technique is assessed with support of numerous performances like accuracy, precision, Recall, F1-score, specificity and negative predictive value (NPV) is analyzed. Proposed CAD-FBPINN method attain 30.53%, 23.34%, and 32.64% higher accuracy; 20.53%, 25.34%, and 29.64% higher precision; 20.53%, 25.34%, and 29.64% higher NP values analyzed with the existing for Classifying AD Stages through Brain Modifications using FBPINNs Optimized with sea-horse optimizer. Then, the effectiveness of the CAD-FBPINN technique is compared to other methods that are currently in use, such as AD diagnosis and classification using a convolution neural network algorithm (DC-AD-AlexNet), Predicting diagnosis 4 years before Alzheimer's disease incident (PDP-ADI-GCNN), and Using the DC-AD-AlexNet convolution neural network algorithm, diagnose and classify AD.

User Behavior Analysis and Prediction Based on Differential Evolution Algorithm Optimized Transformer Combined with Bidirectional Long Short-Term Memory Neural Network

This paper discusses a method combining differential evolution algorithm and bidirectional long short-term memory neural network (BiLSTM) to optimize Transformer model, aiming to improve the accuracy of bank customer credit analysis and prediction. By optimizing the parameters of Transformer model through differential evolution algorithm and combining with the powerful time series analysis capability of BiLSTM, an efficient credit prediction model is constructed. In the process of model training, with the increase of the number of iterations, the correct rate of the model steadily improves and eventually becomes stable. Meanwhile, the value of the loss function also gradually decreases, indicating that the performance of the model on the training data is gradually enhanced. In addition, through confusion matrix analysis, we found that the prediction accuracy of the training set reached 86.15%, and that of the test set reached 82.91%, which showed that the model not only performed well on the training data, but also had strong generalization ability on the unseen data. In addition, the F1 score and DAUC values of the model are both positive, which further confirms the superiority of the model's prediction effect. The research results of this paper show that the Transformer model optimized by differential evolution algorithm and BiLSTM has high accuracy and good generalization ability in the analysis and prediction of bank customer credit. This optimization method not only improves the prediction performance of the model, but also enhances the stability and reliability of the model in practical application by reducing overfitting. Through this optimization strategy, banks can more accurately assess the credit risk of customers and make more informed credit decisions. This research provides a new technical means for credit risk assessment in the financial field, which is helpful to improve the efficiency and security of banking business.

K-Nearest Neighbor and Random Forest Algorithms in Loan Approval Prediction

Loan approval prediction is an important task in the financial sector, which helps banking institutions and lenders make informed decisions regarding loan applications. This research compares the performance of two machine learning algorithms, namely K-Nearest Neighbor (KNN) and Random Forest (RF), in the context of loan approval prediction. The research methodology includes data collection, pre-processing, modeling, and evaluation. The analysis results showed that the Random Forest model performed better overall than KNN, with more true positives and true negatives, and fewer false positives and false negatives. In addition, Random Forest recorded higher accuracy, precision, recall, and F1-score values. These findings provide valuable insights for financial institutions in improving credit risk management strategies and decision-making regarding loan applications.

EVALUATING THE EFFECTIVENESS OF MACHINE LEARNING ALGORITHMS IN PREDICTING CRYPTOCURRENCY PRICES UNDER MARKET VOLATILITY: A STUDY BASED ON THE USA FINANCIAL MARKET

The cryptocurrency market is one of the most dynamic and volatile markets in the world's financial ecosystem, and investment landscapes in the US financial market have changed so much. In slightly over a decade, cryptocurrencies have moved from niche digital assets to mainstream investment opportunities such as Bitcoin, Ethereum, and many others. The prime objective of this research project was to investigate the effectiveness of various machine learning algorithms in the prediction of cryptocurrency prices within the volatile US financial market. This research pinpointed which Machine Learning techniques provide the most accurate and reliable predictions under different market conditions, with a full understanding of their strengths and limitations. The dataset gathered for analyzing and forecasting cryptocurrency prices entailed diverse and extensive data points, affirming a well-rounded foundation for machine learning algorithms. Particularly, current and historic price data from cryptocurrency exchanges such as Binance, Coinbase, and Kraken, together with trading metrics important for the definition of market dynamics. Aggregated data from financial databases such as Coin-Market-Cap, Crypto-Compare, and Yahoo Finance comes in structured form and presents historical consistency, hence perfectly fitting for machine learning applications. Models considered for the study ranged from simple, linear methods to complex ensemble and gradient-boosting algorithms. Precise performance evaluation is a proxy of its reliability and correctness of effectiveness in price predictions in a cryptocurrency market. Several measures of the effectiveness of prediction have been used here for assessing the different properties of models' performance: Precision, Recall, and F1-Score. Additional performance metrics were applied to evaluate the models in this study including Mean Absolute Error, Root Mean Squared Error, and R-squared. The gradient Boosting model did an excellent job as compared to other algorithms, as the values of accuracy, precision, recall, and F1-score for both classes were quite high. All three models have quite a relatively low MAE and RMSE, which means that each model is remarkably good at predicting the target variable. The application of machine learning models in the sphere of cryptocurrency price prediction might finally give very important implications to investors and stakeholders of the financial market in the USA, especially since recently, cryptocurrencies have been made integral parts of both individual and institutional investors' portfolios and trading strategies. To investors, it may provide indications of the entry and exit points, diversification of portfolios, and risk management by using machine learning models. Consolidation with the financial system will indeed mark a strategic shift toward data-driven decision-making in investment management and trading by integrating machine learning models into the financial systems.

Sustainable Air Quality Detection Using Sequential Forward Selection-Based ML Algorithms

Air pollution has exceeded the anticipated safety limit and addressing this issue is crucial for sustainability, particularly in countries with high pollution levels. So, monitoring and forecasting air quality is essential for sustainable urban development. Therefore, this paper presents multiclass classification using two feature selection techniques, namely Sequential Forward Selection (SFS) and filtering, both with different machine learning and ensemble techniques, to predict air quality and make sure that the most relevant features are included in datasets for air quality determination. The results of the considered framework reveal that the SFS technique provides superior performance compared to filter feature selection (FFS) with different ML methods, including the AdaBoost Classifier, the Extra Tree Classifier, Random Forest (RF), and the Bagging Classifier, for efficiently determining the Air Quality Index (AQI). These models’ performances are assessed using predetermined performance metrics. The AdaBoost Classifier model with FFS has the lowest accuracy, while the RF model with SFS achieves the highest accuracy, at 78.4% and 99.99%, respectively. Based on the raw dataset, it was noted that the F1-score, recall, and precision values of the RF model with SFS are 99.96%, 99.97%, and 99.98%, respectively. Therefore, the experimental results undoubtedly show the supremacy, reliability, and robustness of the proposed approach in determining the AQI effectively.

Automated Detection of Spine Deformities: Advancing Orthopedic Care with Convolutional Neural Networks

This paper proposes Spine-CNN, a deep learning model for the detection of spinal deformities that can assist orthopedic doctors as a reliable tool for diagnosis. This technology promises to dramatically simplify the diagnostic process, freeing valuable time, and resources for healthcare professionals. To achieve this objective, a dataset of spine deformity X-ray images was curated from the PhysioNet database. The Spine-CNN was specially designed for detecting the spine deformity by incorporating features to leverage its ability to extract intricate features from radiographic images and by fine tuning the hyperparameters to properly train the model. Model performance was evaluated using standard metrics. Results from the Spine-CNN demonstrated promising performance in detecting spinal deformities. The model achieved an accuracy of 74%, with precision, recall, and F1-score values of 77%, 70%, and 73% respectively. Specifically, this research work introduces a Spine-CNN that underscore the potential of deep learning techniques to revolutionize diagnostic practices in orthopedic medicine, leading to improved treatment outcomes and patient care. Keywords: Computer-aided detection, Convolutional neural network, Image classification, Spine Deformation, X-ray imaging