Correlation-based Feature Selection Research Articles

Cross-Project Defect Prediction Based on Domain Adaptation and LSTM Optimization

Cross-project defect prediction (CPDP) aims to predict software defects in a target project domain by leveraging information from different source project domains, allowing testers to identify defective modules quickly. However, CPDP models often underperform due to different data distributions between source and target domains, class imbalances, and the presence of noisy and irrelevant instances in both source and target projects. Additionally, standard features often fail to capture sufficient semantic and contextual information from the source project, leading to poor prediction performance in the target project. To address these challenges, this research proposes Smote Correlation and Attention Gated recurrent unit based Long Short-Term Memory optimization (SCAG-LSTM), which first employs a novel hybrid technique that extends the synthetic minority over-sampling technique (SMOTE) with edited nearest neighbors (ENN) to rebalance class distributions and mitigate the issues caused by noisy and irrelevant instances in both source and target domains. Furthermore, correlation-based feature selection (CFS) with best-first search (BFS) is utilized to identify and select the most important features, aiming to reduce the differences in data distribution among projects. Additionally, SCAG-LSTM integrates bidirectional gated recurrent unit (Bi-GRU) and bidirectional long short-term memory (Bi-LSTM) networks to enhance the effectiveness of the long short-term memory (LSTM) model. These components efficiently capture semantic and contextual information as well as dependencies within the data, leading to more accurate predictions. Moreover, an attention mechanism is incorporated into the model to focus on key features, further improving prediction performance. Experiments are conducted on apache_lucene, equinox, eclipse_jdt_core, eclipse_pde_ui, and mylyn (AEEEM) and predictor models in software engineering (PROMISE) datasets and compared with active learning-based method (ALTRA), multi-source-based cross-project defect prediction method (MSCPDP), the two-phase feature importance amplification method (TFIA) on AEEEM and the two-phase transfer learning method (TPTL), domain adaptive kernel twin support vector machines method (DA-KTSVMO), and generative adversarial long-short term memory neural networks method (GB-CPDP) on PROMISE datasets. The results demonstrate that the proposed SCAG-LSTM model enhances the baseline models by 33.03%, 29.15% and 1.48% in terms of F1-measure and by 16.32%, 34.41% and 3.59% in terms of Area Under the Curve (AUC) on the AEEEM dataset, while on the PROMISE dataset it enhances the baseline models’ F1-measure by 42.60%, 32.00% and 25.10% and AUC by 34.90%, 27.80% and 12.96%. These findings suggest that the proposed model exhibits strong predictive performance.

Improving with Hybrid Feature Selection in Software Defect Prediction

Software defect prediction (SDP) is used to identify defects in software modules that can be a challenge in software development. This research focuses on the problems that occur in Particle Swarm Optimization (PSO), such as the problem of noisy attributes, high-dimensional data, and premature convergence. So this research focuses on improving PSO performance by using feature selection methods with hybrid techniques to overcome these problems. The feature selection techniques used are Filter and Wrapper. The methods used are Chi-Square (CS), Correlation-Based Feature Selection (CFS), and Forward Selection (FS) because feature selection methods have been proven to overcome data dimensionality problems and eliminate noisy attributes. Feature selection is often used by some researchers to overcome these problems, because these methods have an important function in the process of reducing data dimensions and eliminating uncorrelated attributes that can cause noisy. Naive Bayes algorithm is used to support the process of determining the most optimal class. Performance evaluation will use AUC with an alpha value of 0.050. This hybrid feature selection technique brings significant improvement to PSO performance with a much lower AUC value of 0.00342. Comparison of the significance of AUC with other combinations shows the value of FS PSO of 0.02535, CFS FS PSO of 0.00180, and CS FS PSO of 0.01186. The method in this study contributes to improving PSO in the SDP domain by significantly increasing the AUC value. Therefore, this study highlights the potential of feature selection with hybrid techniques to improve PSO performance in SDP.

Predicting DoS and DDoS attacks in network security scenarios using a hybrid deep learning model

Abstract Network security faces increasing threats from denial of service (DoS) and distributed denial of service (DDoS) attacks. The current solutions have not been able to predict and mitigate these threats with enough accuracy. A novel and effective solution for predicting DoS and DDoS attacks in network security scenarios is presented in this work by employing an effective model, called CNN-LSTM-XGBoost, which is an innovative hybrid approach designed for intrusion detection in network security. The system is applied and analyzed to three datasets: CICIDS-001, CIC-IDS2017, and CIC-IDS2018. We preprocess the data by removing null and duplicate data, handling imbalanced data, and selecting the most relevant features using correlation-based feature selection. The system is evaluated using accuracy, precision, F1 score, and recall. The system achieves a higher accuracy of 98.3% for CICIDS-001, 99.2% for CICIDS2017, and 99.3% for CIC-ID2018, compared to other existing algorithms. The system also reduces the overfitting of the model using the most important features. This study shows that the proposed system is an effective and efficient solution for network attack detection and classification.

Correlation-based feature selection and parallel spatiotemporal networks for efficient passenger flow forecasting in metro systems

This paper presents a novel framework for predicting metro passenger flow that is both interpretable and computationally efficient. The proposed method first uses a correlation-based spatiotemporal feature selection strategy (Cor-STFS) to identify the optimal input scheme for the prediction model, effectively reducing unnecessary interference. The framework then introduces a new multivariate passenger flow prediction architecture called STA-PTCN-BiGRU, which combines a spatiotemporal attention (STA) mechanism, parallel temporal convolutional networks (PTCN), and bidirectional gated recurrent units (BiGRU) to capture the dynamic internal patterns of passenger flow. By utilising parallel computing, this architecture significantly reduces resource consumption. The effectiveness of the proposed approach is evaluated using four datasets from the Shanghai Metro. Experimental results show that the new method outperforms baseline approaches in terms of root mean square error (RMSE), mean absolute error (MAE), and symmetric mean absolute percentage error (SMAPE), achieving average reductions of 9.98%, 8.08%, and 13.29% in these metrics, respectively.

A distributed feature selection pipeline for survival analysis using radiomics in non-small cell lung cancer patients

Predictive modelling of cancer outcomes using radiomics faces dimensionality problems and data limitations, as radiomics features often number in the hundreds, and multi-institutional data sharing is ()often unfeasible. Federated learning (FL) and feature selection (FS) techniques combined can help overcome these issues, as one provides the means of training models without exchanging sensitive data, while the other identifies the most informative features, reduces overfitting, and improves model interpretability. Our proposed FS pipeline based on FL principles targets data-driven radiomics FS in a multivariate survival study of non-small cell lung cancer patients. The pipeline was run across datasets from three institutions without patient-level data exchange. It includes two FS techniques, Correlation-based Feature Selection and LASSO regularization, and Cox Proportional-Hazard regression with Overall Survival as endpoint. Trained and validated on 828 patients overall, our pipeline yielded a radiomic signature comprising "intensity-based energy" and "mean discretised intensity". Validation resulted in a mean Harrell C-index of 0.59, showcasing fair efficacy in risk stratification. In conclusion, we suggest a distributed radiomics approach that incorporates preliminary feature selection to systematically decrease the feature set based on data-driven considerations. This aims to address dimensionality challenges beyond those associated with data constraints and interpretability concerns.

Shortest-length and coarsest-granularity constructs vs. reducts: An experimental evaluation

In the domain of rough set theory, super-reducts represent subsets of attributes possessing the same discriminative power as the complete set of attributes when it comes to distinguishing objects across distinct classes in supervised classification problems. Within the realm of super-reducts, the concept of reducts holds significance, denoting subsets that are irreducible.Contrastingly, constructs, while serving the purpose of distinguishing objects across different classes, also exhibit the capability to preserve certain shared characteristics among objects within the same class. In essence, constructs represent a subtype of super-reducts that integrates information both inter-classes and intra-classes. Despite their potential, constructs have garnered comparatively less attention than reducts.Both reducts and constructs find application in the reduction of data dimensionality. This paper exposes key concepts related to constructs and reducts, providing insights into their roles. Additionally, it conducts an experimental comparative study between optimal reducts and constructs, considering specific criteria such as shortest length and coarsest granularity, and evaluates their performance using classical classifiers.The outcomes derived from employing seven classifiers on sixteen datasets lead us to propose that both coarsest granularity reducts and constructs prove to be effective choices for dimensionality reduction in supervised classification problems. Notably, when considering the optimality criterion of the shortest length, constructs exhibit clear superiority over reducts, which are found to be less favorable.Moreover, a comparative analysis was conducted between the results obtained using the coarsest granularity constructs and a technique from outside of rough set theory, specifically correlation-based feature selection. The former demonstrated statistically superior performance, providing further evidence of its efficacy in comparison.

Comparative Analysis of LSTM and BiLSTM in Image Detection Processing

Tuberculosis is an infectious disease and requires serious treatment. Extrapulmonary Tuberculosis is detected using a microscope. Currently it will take a long time because the fluid preparations are viewed in a microscope one by one carefully and in the fluid preparations there are 150 fields of vision. Examination for Extra Pulmonary Tuberculosis by culture takes between 1-2 weeks or even more. Examination by biopsy will take a long time because the fluid preparations are looked at carefully under the microscope one by one. The image of Tuberculosis is expressed if in the image there is a bacillus object in red, and it turns out that apart from the bacillus object there are other objects also in red. So that examinations for tuberculosis can be more efficient, examinations using computer technology are needed. This research aims to compare the Long Short-Term Memory (LSTM) and Bidirectional Long Short-Term Memory (BiLSTM) classification methods in the detection of extra-pulmonary tuberculosis disease to obtain better accuracy results. This research carried out HSI color space transformation, segmentation using global thresholding, feature extraction using 13 features based on shape and texture using the Correlation Based Feature Selection (CFS) feature selection method. The results show that BiLSTM has the best accuracy with a value of 88.40% at the number of features = 3, namely Short Run High Gray-Level Emphasis, Run Length Nonuniformity, Minor axis length, while LSTM produces an accuracy of 63.19% at the number of features = 5. BiLSTM is capable of detecting opposite features, meaning that BiLSTM can detect opposite features in data sequences and BiLSTM's ability to understand multiple contexts, so it tends to provide more accurate results in some data classification tasks.

LSFSR: Local label correlation-based sparse multilabel feature selection with feature redundancy

In recent studies, existing multilabel feature selection models have focused on either considering the relationship between labels or the redundancy between features. Furthermore, they only use simple sparsity constraints to process high-dimensional data without the intrinsic relationships between features and labels. These issues can have a great impact on the classification effectiveness of feature selection. To address these limitations, this article describes a new local label correlation-based sparse multilabel feature selection approach with feature redundancy. First, a new loss function is established among the matrices of samples, label coefficients, and labels. Then, the Frobenius norm is imposed to investigate the potential relationships between features and labels. The weight matrix is sparsified by the l2,1 norm to ensure that the new loss function has high interpretability. Second, a manifold constraint is employed to capture the local geometric structure between labels and to delve deeper into the latent information among the local labels. Then manifold constraints and Laplacian scores are combined for embedding feature selection to guide the exploration of hidden latent label. Finally, by considering the differences between the feature scores and the redundancy between the samples, feature redundancy is analyzed via the modified cosine similarity, and a candidate feature subset with low redundancy is generated. The l2 norm is used to select features with low redundancy while preserving sparsity, and a novel objective function is developed to optimize this solution. Thus, a sparse feature selection algorithm via local label correlation and feature redundancy is designed, and has demonstrated remarkable classification effectiveness in comparative experiments conducted on 21 multilabel datasets.

Point biserial correlation symbiotic organism search nanoengineering based drug delivery for tumor diagnosis

Nanoparticulate systems have the prospect of accounting for a new making of drug delivery systems. Nanotechnology is manifested to traverse the hurdle of both physical and biological sciences by implementing nanostructures indistinct fields of science, particularly in nano-based drug delivery. The low delivery efficiency of nanoparticles is a critical obstacle in the field of tumor diagnosis. Several nano-based drug delivery studies are focused on for tumor diagnosis. But, the nano-based drug delivery efficiency was not increased for tumor diagnosis. This work proposes a method called point biserial correlation symbiotic organism search nanoengineering-based drug delivery (PBC-SOSN). The objective and aim of the PBC-SOSN method is to achieve higher drug delivery efficiency and lesser drug delivery time for tumor diagnosis. The contribution of the PBC-SOSN is to optimized nanonengineering-based drug delivery with higher r drug delivery detection rate and smaller drug delivery error detection rate. Initially, raw data acquired from the nano-tumor dataset, and nano-drugs for glioblastoma dataset, overhead improved preprocessed samples are evolved using nano variational model decomposition-based preprocessing. After that, the preprocessed samples as input are subjected to variance analysis and point biserial correlation-based feature selection model. Finally, the preprocessed samples and features selected are subjected to symbiotic organism search nanoengineering (SOSN) to corroborate the objective. Based on these findings, point biserial correlation-based feature selection and a symbiotic organism search nanoengineering were tested for their modeling performance with a nano-tumor dataset and nano-drugs for glioblastoma dataset, finding the latter the better algorithm. Incorporated into the method is the potential to adjust the drug delivery detection rate and drug delivery error detection rate of the learned method based on selected features determined by nano variational model decomposition for efficient drug delivery.

Flood mapping based on novel ensemble modeling involving the deep learning, Harris Hawk optimization algorithm and stacking based machine learning

Among the various natural disasters that take place around the world, flood is considered to be the most extensive. There have been several floods in Buzău river basin, and as a result of this, the area has been chosen as the study area. For the purpose of this research, we applied deep learning and machine learning benchmarks in order to prepare flood potential maps at the basin scale. In this regard 12 flood predictors, 205 flood and 205 non-flood locations were used as input data into the following 3 complex models: Deep Learning Neural Network-Harris Hawk Optimization-Index of Entropy (DLNN-HHO-IOE), Multilayer Perceptron-Harris Hawk Optimization-Index of Entropy (MLP-HHO-IOE) and Stacking ensemble-Harris Hawk Optimization-Index of Entropy (Stacking-HHO-IOE). The flood sample was divided into training (70%) and validating (30%) sample, meanwhile the prediction ability of flood conditioning factors was tested through the Correlation-based Feature Selection method. ROC Curve and statistical metrics were involved in the results validation. The modeling process through the stated algorithms showed that the most important flood predictors are represented by: slope (importance ≈ 20%), distance from river (importance ≈ 17.5%), land use (importance ≈ 12%) and TPI (importance ≈ 10%). The importance values were used to compute the flood susceptibility, while Natural Breaks method was used to classify the results. The high and very high flood susceptibility is spread on approximately 35–40% of the study zone. The ROC Curve, in terms of Success, Rate shows that the highest performance was achieved FPIDLNN-HHO-IOE (AUC = 0.97), followed by FPIStacking-HHO-IOE (AUC = 0.966) and FPIMLP-HHO-IOE (AUC = 0.953), while the Prediction Rate indicates the FPIStacking-HHO-IOE as being the most performant model with an AUC of 0.977, followed by FPIDLNN-HHO-IOE (AUC = 0.97) and FPIMLP-HHO-IOE (AUC = 0.924).

Integrating Machine Learning Ensembles for Landslide Susceptibility Mapping in Northern Pakistan

Natural disasters, notably landslides, pose significant threats to communities and infrastructure. Landslide susceptibility mapping (LSM) has been globally deemed as an effective tool to mitigate such threats. In this regard, this study considers the northern region of Pakistan, which is primarily susceptible to landslides amid rugged topography, frequent seismic events, and seasonal rainfall, to carry out LSM. To achieve this goal, this study pioneered the fusion of baseline models (logistic regression (LR), K-nearest neighbors (KNN), and support vector machine (SVM)) with ensembled algorithms (Cascade Generalization (CG), random forest (RF), Light Gradient-Boosting Machine (LightGBM), AdaBoost, Dagging, and XGBoost). With a dataset comprising 228 landslide inventory maps, this study employed a random forest classifier and a correlation-based feature selection (CFS) approach to identify the twelve most significant parameters instigating landslides. The evaluated parameters included slope angle, elevation, aspect, geological features, and proximity to faults, roads, and streams, and slope was revealed as the primary factor influencing landslide distribution, followed by aspect and rainfall with a minute margin. The models, validated with an AUC of 0.784, ACC of 0.912, and K of 0.394 for logistic regression (LR), as well as an AUC of 0.907, ACC of 0.927, and K of 0.620 for XGBoost, highlight the practical effectiveness and potency of LSM. The results revealed the superior performance of LR among the baseline models and XGBoost among the ensembles, which contributed to the development of precise LSM for the study area. LSM may serve as a valuable tool for guiding precise risk-mitigation strategies and policies in geohazard-prone regions at national and global scales.

The multi-objective optimization and mix parameter evaluation of one-part alkali-activated grouting material

To obtain a green and cost-effective one-part alkali-activated grouting material (OAGM) with good comprehensive performance for subgrade reinforcement and determine its proper mix parameters. In this study, using response surface methodology (RSM), multiple objective particle swarm optimization (MOPSO) algorithm and the technique for order preference by similarity to an ideal solution (TOPSIS) algorithm coupled with entropy weight method, two multi-objective optimization schemes for OAGM were performed, one only considering performance and another considering performance, embodied CO2 emission (ECO2e) and cost. Then, the correlation-based feature selection (CFS) algorithm was used to evaluate the importance of various molar ratios as mix parameters on the performance of OAGM. Finally, the effects of key mix parameters on the microstructure and reaction products of OAGM were studied. RSM results show that W/S (water to solid) ratio, borax (BR) content and UCG content chiefly govern the flowability, initial setting time and mechanical strength of OAGM respectively, and the interactions between some factors also significantly affect these properties. MOPSO and entropy-TOPSIS algorithm can be used to optimize OAGM objectively. Two optimal OAGM pastes both have better comprehensive performance and much lower ECO2e than ordinary Portland cement (OPC) paste. And the optimal OAGM considering performance, ECO2e and cost even has 7.68% lower cost than OPC paste. But another optimal OAGM has the best comprehensive performance. CFS results show that the molar ratios of Si/Al, Al/Ca, Na2O/H2O, B2O3/CaO and B2O3/Al2O3 are the key mix parameters of OAGM. Al/Ca rather than Si/Al is the most important mix parameter to the 28-day compressive strength. And higher Si/Al and Al/Ca deteriorates the microstructure and hinders the formation of C-A-S-H gel, while higher Na2O/H2O has the opposite effect. This study provides guidance for objective optimization and the determination of proper mix parameters of OAGM, and promotes the market-oriented application of OAGM.

CFS-AE: Correlation-based Feature Selection and Autoencoder for Improved Intrusion Detection System Performance

The major problem computer network users face concerning data – whether in storage, in transit, or being processed - is unauthorized access. This unauthorized access typically leads to the loss of confidentiality, integrity, and availability of data. Consequently, it is essential to implement an accurate Intrusion Detection System (IDS) for every information system. Many researchers have proposed machine learning and deep learning models, such as autoencoders, to enhance existing IDS. However, the accuracy of these models remains a significant research challenge. This paper proposes a Correlation-Based Feature Selection and Autoencoder (CFS-AE) to enhance detection accuracy and reduce the false alarms associated with the current anomaly-based IDS. The first step involves feature selection for the NSL-KDD and CIC-IDS2017 datasets which are used to train and test our model. Subsequently, an autoencoder is employed as a classifier to categorize data traffic into attack and normal categories. The results from our experimental study revealed an accuracy of 94.32% and 97.71% for the NSL-KDD and CIC-IDS2017 datasets, respectively. These results demonstrate improved performance over existing IDS systems.

A continuous cuffless blood pressure measurement from optimal PPG characteristic features using machine learning algorithms

Background and objectiveHypertension is a potentially dangerous health condition that can be detected by measuring blood pressure (BP). Blood pressure monitoring and measurement are essential for preventing and treating cardiovascular diseases. Cuff-based devices, on the other hand, are uncomfortable and prevent continuous BP measurement. MethodsIn this study, a new non-invasive and cuff-less method for estimating Systolic Blood Pressure (SBP), Mean Arterial Pressure (MAP), and Diastolic Blood Pressure (DBP) has been proposed using characteristic features of photoplethysmogram (PPG) signals and nonlinear regression algorithms. PPG signals were collected from 219 participants, which were then subjected to preprocessing and feature extraction steps. Analyzing PPG and its derivative signals, a total of 46 time, frequency, and time-frequency domain features were extracted. In addition, the age and gender of each subject were also included as features. Further, correlation-based feature selection (CFS) and Relief F feature selection (ReliefF) techniques were used to select the relevant features and reduce the possibility of over-fitting the models. Finally, support vector regression (SVR), K-nearest neighbour regression (KNR), decision tree regression (DTR), and random forest regression (RFR) were established to develop the BP estimation model. Regression models were trained and evaluated on all features as well as selected features. The best regression models for SBP, MAP, and DBP estimations were selected separately. ResultsThe SVR model, along with the ReliefF-based feature selection algorithm, outperforms other algorithms in estimating the SBP, MAP, and DBP with the mean absolute error of 2.49, 1.62 and 1.43 mmHg, respectively. The proposed method meets the Advancement of Medical Instrumentation standard for BP estimations. Based on the British Hypertension Society standard, the results also fall within Grade A for SBP, MAP, and DBP. ConclusionThe findings show that the method can be used to estimate blood pressure non-invasively, without using a cuff or calibration, and only by utilizing the PPG signal characteristic features.

The Hybrid Features and Supervised Learning for Batik Pattern Classification

Several countries have traditional textiles as a piece of their cultural heritage. Indonesia has a traditional textile called batik. Central Java is one of the regions producing batik known for its variety of distinctive themes. It has unique designs and several motifs that emphasize the beauty of historic sites. Since the diversity of central java batik motifs and the lack of knowledge from the surrounding community, only a select group of people, especially the batik craftsmen themselves, can recognize these motifs. Consequently, the method to identify the batik according to the primary ornament pattern is required. Therefore, this study proposes a computer vision-based method for classifying batik patterns. The proposed method required discriminating appropriate features to produce optimal results. The discriminating features were constructed based on color, shape, and texture. Those features were derived using the method of Color Moments, Area Based Invariant Moments, Gray Level Co-occurrence Matrix (GLCM), and Local Binary Pattern (LBP). This study’s proposed hybrid features were formed based on the most discriminating and appropriate features. These were yielded by the Correlation-based feature selection (CFS) method. The hybrid features were then fed into several classifiers to determine the batik pattern. The pattern consists of ten classes: Asem Arang, Asem Sinom, Asem Warak, Blekok, Blekok Warak, Gambang Semarangan, Kembang Sepatu, Semarangan, Tugu Muda, and Warak Beras Utah. Based on the experimental results, the most optimal predicted class of the batik pattern was generated using the Artificial Neural Network (ANN) classifier. It was indicated by achieving an accuracy value of 99.76% based on the 3,000 images (each class consists of 300 images) with cross-validation using a k-fold value of 10. This study has proved that the hybrid features incorporated with ANN can be selected as a suitable model to classify the batik patterns.

A260 ADVANCING INFLAMMATORY BOWEL DISEASE DIAGNOSIS THROUGH STOOL PROTEOMIC SIGNATURES OBTAINED VIA DIA-MASS SPECTROMETRY AND MACHINE LEARNING

Abstract Background Resembling symptoms between flare-up Inflammatory bowel disease (IBD) and other gastrointestinal diseases and conditions, make the initial diagnosis and accurate treatment of this disease challenging. Currently, the most common clinical methods used for diagnosing and monitoring IBD are colonoscopy and biopsy, which are invasive and uncomfortable procedures, and the fecal calprotectin test, which is not sufficiently accurate. Therefore, it is necessary to develop an alternative method. Aims In this study, our aim was to develop a robust predictive model for a noninvasive and accurate test to distinguish active IBD from non-IBD controls. Methods A total of 120 samples were collected, with 78 samples separated for retrospective analysis and model training, while 42 samples were set aside for prospective validation. We employed SWATH mass spectrometry for the identification and quantification of the stool proteome. The obtained data underwent multiple stages, including data preprocessing, feature selection, model training, and performance evaluation. We optimized data processing procedures through advanced bioinformatics, selecting an appropriate pipeline that included data normalization, batch effect correction, and missing value imputation. Subsequently, we assessed various machine learning algorithms to determine the most effective classifier for predicting IBD based on the selected proteins. Results After data preprocessing, we identified 48 differentially abundant proteins (DAPs). To eliminate redundant proteins, we employed Correlation-based Feature Selection (CFS), resulting in selecting 7 proteins. To identify the most suitable predictive model for our dataset, we assessed five popular machine learning methods: Support Vector Machines (SVM), Random Forests, Logistic Regression, k-nearest neighbors (KNN), and Naive Bayes. Among these, SVM exhibited the highest performance. Finally, we evaluated the model's performance by applying the selected algorithm to 42 prospective blind samples. The results revealed a sensitivity of 96% and a specificity of 76%, highlighting its strength and effectiveness. Conclusions In conclusion, this study offers a proof of concept for the application of SWATH for precise IBD diagnosis using stool proteomics and showcases the effectiveness of our data processing and machine learning approach. Additionally, it highlights the potential of this method for classifying Crohn's Disease (CD) vs. Ulcerative Colitis (UC) and distinguishing active IBD from remission. Workflow of study Funding Agencies CIHRCrohn's and Colitis Canada , Studentship from CRMUS,

Monitoring Soil Salinity Classes through Remote Sensing-Based Ensemble Learning Concept: Considering Scale Effects

Remote sensing (RS) technology can rapidly obtain spatial distribution information on soil salinization. However, (1) the scale effects resulting from the mismatch between ground-based “point” salinity data and remote sensing pixel-based “spatial” data often limit the accuracy of remote sensing monitoring of soil salinity, and (2) the same salinity RS monitoring model usually provides inconsistent or sometimes conflicting explanations for different data. Therefore, based on Landsat 8 imagery and synchronously collected ground-sampling data of two typical study regions (denoted as N and S, respectively) of the Yichang Irrigation Area in the Hetao Irrigation District for May 2013, this study used geostatistical methods to obtain “relative truth values” of salinity corresponding to the Landsat 8 pixel scale. Additionally, based on Landsat 8 multispectral data, 14 salinity indices were constructed. Subsequently, the Correlation-based Feature Selection (CFS) method was used to select sensitive features, and a strategy similar to the concept of ensemble learning (EL) was adopted to integrate the single-feature-sensitive Bayesian classification (BC) model in order to construct an RS monitoring model for soil salinization (Nonsaline, Slightly saline, Moderately saline, Strongly saline, and Solonchak). The research results indicated that (1) soil salinity exhibits moderate to strong variability within a 30 m scale, and the spatial heterogeneity of soil salinity needs to be considered when developing remote sensing models; (2) the theoretical models of salinity variance functions in the N and S regions conform to the exponential model and the spherical model, with R2 values of 0.817 and 0.967, respectively, indicating a good fit for the variance characteristics of salinity and suitability for Kriging interpolation; and (3) compared to a single-feature BC model, the soil salinization identification model constructed using the concept of EL demonstrated better potential for robustness and effectiveness.

Application of SWATH Mass Spectrometry and Machine Learning in the Diagnosis of Inflammatory Bowel Disease Based on the Stool Proteome.

Inflammatory bowel disease (IBD) flare-ups exhibit symptoms that are similar to other diseases and conditions, making diagnosis and treatment complicated. Currently, the gold standard for diagnosing and monitoring IBD is colonoscopy and biopsy, which are invasive and uncomfortable procedures, and the fecal calprotectin test, which is not sufficiently accurate. Therefore, it is necessary to develop an alternative method. In this study, our aim was to provide proof of concept for the application of Sequential Window Acquisition of All Theoretical Mass Spectra-Mass spectrometry (SWATH-MS) and machine learning to develop a non-invasive and accurate predictive model using the stool proteome to distinguish between active IBD patients and symptomatic non-IBD patients. Proteome profiles of 123 samples were obtained and data processing procedures were optimized to select an appropriate pipeline. The differentially abundant analysis identified 48 proteins. Utilizing correlation-based feature selection (Cfs), 7 proteins were selected for proceeding steps. To identify the most appropriate predictive machine learning model, five of the most popular methods, including support vector machines (SVMs), random forests, logistic regression, naive Bayes, and k-nearest neighbors (KNN), were assessed. The generated model was validated by implementing the algorithm on 45 prospective unseen datasets; the results showed a sensitivity of 96% and a specificity of 76%, indicating its performance. In conclusion, this study illustrates the effectiveness of utilizing the stool proteome obtained through SWATH-MS in accurately diagnosing active IBD via a machine learning model.

Anomaly Detection IDS for Detecting DoS Attacks in IoT Networks Based on Machine Learning Algorithms.

Widespread and ever-increasing cybersecurity attacks against Internet of Things (IoT) systems are causing a wide range of problems for individuals and organizations. The IoT is self-configuring and open, making it vulnerable to insider and outsider attacks. In the IoT, devices are designed to self-configure, enabling them to connect to networks autonomously without extensive manual configuration. By using various protocols, technologies, and automated processes, self-configuring IoT devices are able to seamlessly connect to networks, discover services, and adapt their configurations without requiring manual intervention or setup. Users' security and privacy may be compromised by attackers seeking to obtain access to their personal information, create monetary losses, and spy on them. A Denial of Service (DoS) attack is one of the most devastating attacks against IoT systems because it prevents legitimate users from accessing services. A cyberattack of this type can significantly damage IoT services and smart environment applications in an IoT network. As a result, securing IoT systems has become an increasingly significant concern. Therefore, in this study, we propose an IDS defense mechanism to improve the security of IoT networks against DoS attacks using anomaly detection and machine learning (ML). Anomaly detection is used in the proposed IDS to continuously monitor network traffic for deviations from normal profiles. For that purpose, we used four types of supervised classifier algorithms, namely, Decision Tree (DT), Random Forest (RF), K Nearest Neighbor (kNN), and Support Vector Machine (SVM). In addition, we utilized two types of feature selection algorithms, the Correlation-based Feature Selection (CFS) algorithm and the Genetic Algorithm (GA) and compared their performances. We also utilized the IoTID20 dataset, one of the most recent for detecting anomalous activity in IoT networks, to train our model. The best performances were obtained with DT and RF classifiers when they were trained with features selected by GA. However, other metrics, such as training and testing times, showed that DT was superior.

Prediction of Student Entrepreneurship Future Work based on Entrepreneurship Course using the Naïve Bayes Classifier Model

Entrepreneurs are critical to a country's economic progress and job creation. Few people felt schools have much to offer with business a generation ago. Students are expected to be an entrepreneur as the outcome of the course. The goal of this study is building a model to predict students' future employment, particularly in the field of entrepreneurship, using big data analysis and data mining. Various educational institutions can use data mining methodologies to identify hidden patterns in data contained in databases. The feature selection technique was utilised in this study to select and assess the significance of each element. The model was built using the final parameters determined by the feature selection technique (Correlation Based Feature Selection). Using the 10-fold cross validations for training and testing dataset distribution, the Naïve Bayes classifier was used to forecast the students' future of work. The dataset for the study was gathered from a student's performance report at Universitas Negeri Medan's engineering department. The effectiveness of using feature selection algorithms was compared to the effectiveness of not using feature selection algorithms, and the results are discussed. According to the findings of this study, the accuracy of Naïve Bayes with Correlation Based Feature Selection is 87.4%, which is higher than the model that did not use any feature selection. It was also discovered that the overall accuracy of the Correlation Based Feature Selection and Naïve Bayes Classifier models appears to be higher than that of the other treatments.