Support Vector Machine Tree Research Articles

How Can Machine Learning Contribute to the Credit Scoring Process in SME Lending?

This research investigates the potential of machine learning to enhance the credit scoring process in small and medium-sized enterprise (SME) lending. SMEs are vital to global economies, accounting for over 90\% of businesses and more than half of total employment worldwide, according to The World Bank (2023). In emerging economies, SME lending is particularly crucial, providing access to capital that supports economic growth and development. However, assessing the credit risk of SMEs poses unique challenges due to limited financial histories and variable cash flows, making accurate credit scoring essential for lenders. Machine learning offers promising solutions for the risk assessment process in SME lending, leveraging algorithms and large datasets to identify patterns and improve decision-making. By applying techniques such as artificial neural networks, decision tree learning, gradient boosting, logistic regression, random forest, and support vector machines, machine learning can enhance the accuracy and efficiency of credit scoring models. This research explores how these machine learning methods can reduce biases, improve risk prediction, and ultimately support sustainable lending practices by enabling lenders to make more informed credit decisions. The findings aim to illustrate the transformative potential of machine learning in SME credit scoring, offering a framework for financial institutions seeking to integrate advanced data-driven technologies in the credit assessment process.

Deep learning for prediction and classifying the dynamical behaviour of piecewise-smooth maps

This paper explores novel ways of predicting and classifying the dynamics of piecewise smooth maps using various deep learning models. Moreover, we have used machine learning models such as Decision Tree Classifier, Logistic Regression, K-Nearest Neighbor, Random Forest, and Support Vector Machine for predicting the border collision bifurcation in the 1D normal form map and the 1D tent map. The decision tree classifier best predicts the border collision bifurcation for the 1D normal form map, the random forest, and the 1D tent map. This study introduces a novel application of deep learning models to cobweb diagrams and phase portraits, which provides a new perspective for classifying regular and chaotic behaviour. Further, we classified the regular and chaotic behaviour of the 1D tent map and the 2D Lozi map using deep learning models like Convolutional Neural Network (CNN), ResNet50, and ConvLSTM via cobweb diagram and phase portraits, where CNN exhibits better performance than other models. We also classified the chaotic and hyperchaotic behaviour of the 3D piecewise smooth map using deep learning models such as the Feed Forward Neural Network (FNN), Long Short-Term Memory (LSTM), and Recurrent Neural Network (RNN). We have shown that LSTM performs best for classifying chaotic and hyperchaotic behaviour. Additionally, LSTM outperforms other models in accuracy and computational efficiency, making it highly effective for real-time analysis. Finally, deep learning models such as Long Short Term Memory (LSTM) and Recurrent Neural Network (RNN) are used for reconstructing the two-parameter bifurcation charts of 2D normal form map, in which LSTM is more precise than RNN in reconstructing the two-parametric charts.

Application of Improved Classification Algorithm with Binary Tree Support Vector Machine in Power Grid Prediction

Abstract To further improve the accuracy of the power grid prediction, the paper proposes an improved binary tree classification algorithm with the SVM and predicts the real time power grid. Based on the basic principles of SVM, the common multi-classifier classification algorithm and its characteristics are summarized. Incorporating the advantages of existing classification algorithms, an improved binary tree classification algorithm was used, addressing the limitations of conventional methods. From the simulation results, it can be seen that the improved binary tree algorithm can shorten the test time from the original 0.285 seconds to 0.267 seconds, and the measured category can also be increased from the original 94.42 to 95.27. This shows that the improved binary tree can not only improve the practicability of the support vector machine algorithm, but also better improve the ability of real-time prediction of power grid.

Optimizing Prehospital Stroke Diagnosis: Integrating Machine Learning with the FAST Scoring System

Strokes are a leading cause of disability and death worldwide, and timely diagnosis is critical for effective treatment. Prehospital stroke diagnosis by emergency medical services (EMS) is essential for ensuring prompt care. The FAST (Face, Arms, Speech, and Time) scoring system is a common tool used to identify stroke symptoms quickly. However, the traditional FAST system has limitations, including reliance on subjective assessments and potential misdiagnoses due to symptom variability. This study aims to optimize the FAST scoring system and compare its performance with various machine learning models to enhance stroke prediction accuracy. A dataset from an ambulance service, including prehospital data such as age, sex, and FAST test results, was used. Data preprocessing involved handling missing values, encoding categorical variables, and applying the Synthetic Minority Over-sampling Technique (SMOTE) to balance the dataset. A grid search was conducted to test different weighting schemes for the FAST system. Machine learning models, including Logistic Regression, Random Forest, Support Vector Machine (SVM), XGBoost, and Decision Tree, were trained and evaluated using cross-validation on the training set and tested on a separate test set. The traditional FAST system showed an accuracy of 62.95%. Optimizing the FAST system through grid search marginally increased accuracy to 64.36%. However, machine learning models, particularly Random Forest, significantly outperformed the FAST system, achieving a test set accuracy of 88.54%. XGBoost also demonstrated strong performance with an accuracy of 77.07%, while Logistic Regression, SVM, and Decision Tree showed lower accuracies. These findings suggest that integrating machine learning models into EMS protocols could substantially improve the accuracy of prehospital stroke diagnosis, leading to quicker and more accurate identification and treatment of stroke patients. Although the optimized FAST system offers slight improvements, machine learning models present a promising avenue for enhancing stroke prediction and improving patient outcomes in emergency settings while acknowledging practicality issues.

A novel infrared thermography image analysis for transformer condition monitoring

Electrical systems are deeply ingrained in most industrial facilities, their maintenance is increasingly becoming a critical and significant component of economic policy. Condition monitoring of electrical transformers is essential for improving their dependability and availability, averting costly maintenance and additional significant breakdowns. This research adopts the approach based on infrared thermography techniques (IRT) to keep eye on electrical transformers and detect their defects. Thermal images of the transformer were captured at two distinct operating states with an infrared camera. These images were then compiled into a dataset for further analysis. This method uses infrared imaging (IRT), along with feature analysis and machine learning, to identify issues in electrical transformers in a new way. To find the best performing machine learning model, different techniques are compared here in terms of their accuracy and stability. Two approaches are investigated for identifying features in thermography images. Approach-1 employed five common machine learning algorithms, such as Support Vector Machine (SVM), K-Nearest Neighbours (KNN), Decision Tree (DT), Logistic Regression (LR), and Least Squares Support Vector Machine (LS-SVM). Approach-2 utilized four deep learning techniques, such as MobileNetV2 (MNV2), InceptionV3 (InV3), DenseNet121(DN121), and our proposed modified VGG-16. Among all evaluated methods, the modified VGG-16 architecture achieved the highest level of dependability, demonstrating exceptional efficiency and accuracy in transformer condition monitoring and fault diagnosis.

Hepatitis C prediction using SVM, logistic regression and decision tree

Hepatitis C is an infection of the liver brought on by the HCV virus. In this condition, early diagnosis is challenging because of the delayed onset of symptoms. Predicting well enough can spare patients from permeant liver damage. The primary goal of this work is to use several machine learning methods to forecast this disease based on widely available and reasonably priced blood test data in order to diagnose and treat patients early on. Three machine learning techniques support vector machine (SVM), logistic regression, decision tree, has been applied on one dataset in this work. To find a suitable approach for illness prediction, the confusion matrix, precision, recall, F1 score, accuracy, receiver operating characteristics (ROC), performances of different strategies have been assessed. The SVM model's overall accuracy is 0.92, the highest among the three models.

PREDICTIVE MODELING FOR LOAN APPROVAL: A MACHINE LEARNING APPROACH

Exploring machine learning approaches to enhance the effectiveness and precision of procedures related to bank loan approval. This investigation encompasses various methods such as logistic regression, decision trees, linear regression, as well as GaussianNB, Random Forest, and SVM. Utilizing a substantial dataset containing past loan applications and diverse applicant attributes like demographics, credit scores, income levels, and employment histories. The research endeavors to evaluate the recall, accuracy, precision, and F1-score metrics of various algorithms. Additionally, it investigates the interpretability and transparency of machine learning models to offer further insight into the variables affecting decisions on loan acceptance. The study emphasizes the efficacy of logistic regression, which outperformed SVM (77%), GaussianNB (78%), random forests (78%), and decision trees (69%), achieving the highest accuracy of 80% in loan approval. By implementing this model, we can enhance ML-driven loan approval processes within the banking industry, thereby elevating decision-making standards and enhancing consumer satisfaction. KEYWORDS— Machine Learning Algorithms, Loan Approval, LogisticRegression, DecisionTree, Linear Regression, GaussianNB, RandomForest, SupportVectorMachine (SVM), Decision-Making,

Optimising air quality prediction in smart cities with hybrid particle swarm optimization‐long‐short term memory‐recurrent neural network model

AbstractIn smart cities, air pollution is a critical issue that affects individual health and harms the environment. The air pollution prediction can supply important information to all relevant parties to take appropriate initiatives. Air quality prediction is a hot area of research. The existing research encounters several challenges that is, poor accuracy and incorrect real‐time updates. This research presents a hybrid model based on long‐short term memory (LSTM), recurrent neural network (RNN), and Curiosity‐based Motivation method. The proposed model extracts a feature set from the training dataset using an RNN layer and achieves sequencing learning by applying an LSTM layer. Also, to deal with the overfitting issues in LSTM, the proposed model utilises a dropout strategy. In the proposed model, input and recurrent connections can be dropped from activation and weight updates using the dropout regularisation approach, and it utilises a Curiosity‐based Motivation model to construct a novel motivational model, which helps in the reconstruction of long short‐term memory recurrent neural network. To minimise the prediction error, particle swarm optimisation is implemented to optimise the LSTM neural network's weights. The authors utilise an online Air Pollution Monitoring dataset from Salt Lake City, USA with five air quality indicators for comparison, that is, SO2, CO, O3, and NO2, to predict air quality. The proposed model is compared with existing Gradient Boosted Tree Regression, Existing LSTM, and Support Vector Machine based Regression Model. Experimental analysis shows that the proposed method has 0.0184 (Root Mean Square Error (RMSE)), 0.0082 (Mean Absolute Error), 2002*109 (Mean Absolute Percentage Error), and 0.122 (R2‐Score). The experimental findings demonstrate that the proposed LSTM model had RMSE performance in the prescribed dataset and statistically significant superior outcomes compared to existing methods.

Prediction Models of Growth Characteristics and Yield for Chinese Winter Wheat Based on Machine Learning

In order to eliminate the limitations of traditional winter wheat yield prediction methods, the prediction models based on machine learning are used to improve the accuracy of winter wheat yield prediction. In this study, by collecting a large amount of domestic literature about wheat growth characteristics, the irrigation amount, fertilization amount, soil nutrient status, planting density, maximum leaf area index (LAImax), maximum aboveground dry matter accumulation (Dmax) and yield (Y) were chosen to develop the learning models. Using the data of the irrigation amount, fertilization amount, soil nutrient status and planting density as the training set, the regression prediction models (Gaussian process regression mode, linear regression model, regression tree mode and support vector machine model) were used to train and learn the data of the LAImax, Dmax and Y, respectively. The results show that the Gaussian regression model has the best precision compared to the other models. The coefficients of determination (R2) of the learning results of the Gaussian regression model for the LAImax, Dmax and Y are 0.9, 0.93 and 0.86, and the root mean square error (RMSE) is 0.57, 1125.1 and 640.41. Based on the data of the irrigation amount, nitrogen application amount, potassium application amount, phosphorus application amount, organic matter content, total nitrogen content, alkali-hydrolyzable nitrogen content, available phosphorus content, available potassium content and planting density, the method proposed in this paper can reliably predict the LAImax, the Dmax and Y of winter wheat. The results also have certain reference significance for the yield prediction of other crops.

Fabric tearing performance state perception and classification driven by multi-source data.

The tear strength of textiles is a crucial characteristic of product quality. However, during the laboratory testing of this indicator, factors such as equipment operation, human intervention, and test environment can significantly influence the results. Currently, there is a lack of traceable records for the influencing factors during the testing process, and effective classification of testing activities is not achieved. Therefore, this study proposes a state-awareness and classification approach for fabric tear performance testing based on multi-source data. A systematic design is employed for fabric tear performance testing activities, which can real-time monitor electrical parameters, operational environment, and operator behavior. The data are collected, preprocessed, and a Decision Tree Support Vector Machine (DTSVM) is utilized for classifying various working states, and introducing ten-fold cross-validation to enhance the performance of the classifier, forming a comprehensive awareness of the testing activities. Experimental results demonstrate that the system effectively perceives fabric tear performance testing processes, exhibiting high accuracy in the classification of different fabric testing states, surpassing 98.73%. The widespread application of this system contributes to continuous improvement in the workflow and traceability of fabric tear performance testing processes.

Machine Learning and Texture Analysis of [18F]FDG PET/CT Images for the Prediction of Distant Metastases in Non-Small-Cell Lung Cancer Patients.

The aim of our study was to predict the occurrence of distant metastases in non-small-cell lung cancer (NSCLC) patients using machine learning methods and texture analysis of 18F-labeled 2-deoxy-d-glucose Positron Emission Tomography/Computed Tomography {[18F]FDG PET/CT} images. In this retrospective and single-center study, we evaluated 79 patients with advanced NSCLC who had undergone [18F]FDG PET/CT scan at diagnosis before any therapy. Patients were divided into two independent training (n = 44) and final testing (n = 35) cohorts. Texture features of primary tumors and lymph node metastases were extracted from [18F]FDG PET/CT images using the LIFEx program. Six machine learning methods were applied to the training dataset using the entire panel of features. Dedicated selection methods were used to generate different combinations of five features. The performance of selected machine learning methods applied to the different combinations of features was determined using accuracy, the confusion matrix, receiver operating characteristic (ROC) curves, and area under the curve (AUC). A total of 104 and 78 lesions were analyzed in the training and final testing cohorts, respectively. The support vector machine (SVM) and decision tree methods showed the highest accuracy in the training cohort. Seven combinations of five features were obtained and introduced in the models and subsequently applied to the training and final testing cohorts using the SVM and decision tree. The accuracy and the AUC of the decision tree method were higher than those obtained with the SVM in the final testing cohort. The best combination of features included shape sphericity, gray level run length matrix_run length non-uniformity (GLRLM_RLNU), Total Lesion Glycolysis (TLG), Metabolic Tumor Volume (MTV), and shape compacity. The combination of these features with the decision tree method could predict the occurrence of distant metastases with an accuracy of 74.4% and an AUC of 0.63 in NSCLC patients.

Analysis of Machine Learning Classifiers for Speaker Identification: A Study on SVM, Random Forest, KNN, and Decision Tree

This study investigates the performance of machine learning classifiers in the domain of speaker identification, a pivotal component of modern digital security systems. With the burgeoning integration of voice-activated interfaces in technology, the demand for accurate and reliable speaker identification is paramount. This research provides a comprehensive comparison of four widely used classifiers: Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbors (KNN), and Decision Tree (DT). Utilizing the LibriSpeech dataset, known for its diversity of speakers and recording conditions, we extracted Mel-frequency cepstral coefficients (MFCCs) to serve as features for training and evaluating the classifiers. Each model's performance was assessed based on precision, recall, F1-score, and accuracy. The results revealed that RF outperformed all other classifiers, achieving near-perfect metrics, indicative of its robustness and generalizability for speaker identification tasks. KNN also demonstrated high performance, suggesting its suitability for applications where rapid execution and interpretability are critical. Conversely, SVM and DT, while yielding moderate and lower performances respectively, highlighted the necessity for further optimization. These findings underscore the effectiveness of ensemble and distance-based classifiers in handling complex patterns for speaker differentiation. The study not only guides the selection of appropriate classifiers for speaker identification but also sets the stage for future research, which could explore hybrid models and the impact of dataset variability on performance. The insights from this analysis contribute significantly to the field, providing a benchmark for developing advanced speaker identification systems

Improvement Path of Legal System Related to ChatGPT Application Combined with Decision Tree Algorithm

Abstract Artificial Intelligence (AI) has been widely used in the social and legal fields, and ChatGPT, after AI painting, has once again set off a wave of discussion on whether AI and its generated works can obtain legal protection. Starting from the theoretical orientation of the origin of ChatGPT legal governance, this paper proposes the legal positioning and layered governance framework of ChatGPT application. It explores the role mechanism of ChatGPT empowering legal modernization and combs through the realistic dilemmas of ChatGPT-generated content data compliance legalization. To effectively analyze legal risks in the process of the ChatGPT application, data crawling technology and SMOTE oversampling technology are utilized to obtain ChatGPT application data and produce datasets. The Stacking integration strategy is introduced to combine the Random Forest in the Decision Tree Algorithm, GBDT algorithm, and Support Vector Machine to construct the legal risk prediction model of the ChatGPT application. For the effectiveness of the model, the ChatGPT application dataset is used to analyze the accuracy, ROC curve, and AUC value, which provides a reference for improving the legal system related to the ChatGPT application. The results show that the accuracy of the SVM classifier reaches 0.839, the correctness of the GBDT model is 0.947, and the AUC value of ChatGPT legal risk prediction based on the Stacking integration strategy is 0.947. Based on the inspiration of the decision tree algorithm, the improvement of the legal system related to the ChatGPT application should be improved in terms of generating content and allocating risk. Based on the insights of the decision tree algorithm, the improvement of the legal system related to the ChatGPT application should be made based on the dimensions of generation content and risk allocation.

Machine Learning-Based Classification of Small-Sized Wetlands Using Sentinel-2 Images

<abstract> <p>Wetlands are invaluable ecosystems, offering essential services such as carbon sequestration, water purification, flood control and habitat for countless aquatic species. However, these critical environments are under increasing threat from factors like industrialization and agricultural expansion. In this research, we focused on small-sized wetlands, typically less than 10 acres in size, due to their unique ecological roles and groundwater recharge contributions. To effectively protect and manage these wetlands, precise mapping and monitoring are essential. To achieve this, we exploited the capabilities of Sentinel-2 imagery and employ a range of machine learning algorithms, including Random Forest (RF), Classification and Regression Tree (CART), Gradient Tree Boost (GTB), Naive Bayes (NB), k-nearest neighbors (KNN) and Support Vector Machine (SVM). Our evaluation used variables, such as spectral bands, indices and image texture. We also utilized Google Earth Engine (GEE) for streamlined data processing and visualization. We found that Random Forest (RF) and Gradient Tree Boost (GTB) outperformed other classifiers according to the performance evaluation. The Normalized Difference Water Index (NDWI) came out to be one of the important predictors in mapping wetlands. By exploring the synergistic potential of these algorithms, we aim to address existing gaps and develop an optimized approach for accurate small-sized wetland mapping. Our findings will be useful in understanding the value of small wetlands and their conservation in the face of environmental challenges. They will also lay the framework for future wetland research and practical uses.</p> </abstract>

Research on objective evaluation methods of labor education teaching for college students in higher education

Abstract Labor education in colleges and universities in the new era is the fundamental task of implementing the principle of establishing morality and educating people, as well as an essential hand in cultivating college students with all-round development of morality, intelligence, physicality, aesthetics, and labor. This paper focuses on the objective evaluation method of college students’ labor education to conduct research. Using the improved binary tree support vector machine multi-class classification algorithm, an evaluation model for college labor education based on a binary tree support vector machine has been constructed, and an evaluation system for labor education has been developed. The results show that the majors that have compulsory labor education courses, elective labor education courses, and have added labor education modules in their major courses are all art and sports, accounting for more than 50% on average. The percentages of liberal arts, science, economics and management, art and sports, medicine and science, and agriculture, which did not offer labor education-related courses, were 26.35%, 22.16%, 19.91%, 13.59%, 11.03%, and 13.19%, respectively. It is evident that labor education courses are not distributed equally in colleges and universities, and there are still some majors that do not offer labor education courses. In terms of labor education, most of the labor education through Civics and Political Science classes is provided to students. Managers are those who choose the most specialized courses. Students also tend to participate in bicultural education and social practice activities, accounting for 29.02% and 43.01%, respectively. It shows that the algorithm in this paper can objectively and effectively evaluate labor education in colleges and universities and identify the deficiencies in education.

Machine-learning prediction of treatment response to stereotactic body radiation therapy in oligometastatic gynecological cancer: A multi-institutional study

Background and purposeWe aimed to develop and validate different machine-learning (ML) prediction models for the complete response of oligometastatic gynecological cancer after SBRT. Material and methodsOne hundred fifty-seven patients with 272 lesions from 14 different institutions and treated with SBRT with radical intent were included. Thirteen datasets including 222 lesions were combined for model training and internal validation purposes, with an 80:20 ratio. The external testing dataset was selected as the fourteenth Institution with 50 lesions. Lesions that achieved complete response (CR) were defined as responders. Prognostic clinical and dosimetric variables were selected using the LASSO algorithm. Six supervised ML models, including logistic regression (LR), classification and regression tree analysis (CART) and support vector machine (SVM) using four different kernels, were trained and tested to predict the complete response of uterine lesions after SBRT. The performance of models was assessed by receiver operating characteristic curves (ROC), area under the curve (AUC) and calibration curves. An explainable approach based on SHapley Additive exPlanations (SHAP) method was deployed to generate individual explanations of the model's decisions. Results63.6% of lesions had a complete response and were used as ground truth for the supervised models. LASSO strongly associated complete response with three variables, namely the lesion volume (PTV), the type of lesions (lymph-nodal versus parenchymal), and the biological effective dose (BED10), that were used as input for ML modeling. In the training set, the AUCs for complete response were 0.751 (95% CI: 0.716–0.786), 0.766 (95% CI: 0.729–0.802) and 0.800 (95% CI: 0.742–0.857) for the LR, CART and SVM with a radial basis function kernel, respectively. These models achieve AUC values of 0.727 (95% CI: 0.669–0.795), 0.734 (95% CI: 0.649–0.815) and 0.771 (95% CI: 0.717–0.824) in the external testing set, demonstrating excellent generalizability. ConclusionML models enable a reliable prediction of the treatment response of oligometastatic lesions receiving SBRT. This approach may assist radiation oncologists to tailor more individualized treatment plans for oligometastatic patients.

Comparative SVM and Decision Tree Algorithm in Identifying the Eligibility of KIP Scholarship Awardee

Scholarship selection process has specific rules, but if the number of applicants exceeds the quota, a selection process is needed. Based on the observation of a university in Sukabumi, the selection for KIP scholarship has not yet had a standard method. Several methods can be used to assist the selection process, such as classification based on historical data of applicants. The algorithms used for classification include Decision Tree (DT) and Support Vector Machine (SVM). The research process uses SEMMA (Sample, Explore, Modify, Model, Assess) method. Dataset for KIP scholarship awardee from 2021-2022 consist of 519 samples with 16 attributes. From the exploration results, the most important features for model modeling are Status DTKS, Status P3KE, Father's income, mother's income, combined income, and performance. These attributes are converted into numerical data to facilitate model fitting. The K-Fold Cross-Validation results for the Decision Tree model in the case of KIP Scholarship classification yield an accuracy of 78.44% for the entire test dataset, a precision of 0.73107, indicating that 73.11% of the predictions are true, a recall (sensitivity) of 78.45%, and an F1 score of 73.20%. The results for the SVM model are an accuracy of 80.17%, a precision of 84.44%, and a recall of 80.17%.

Crop Yield Prediction in Nigeria Using Machine Learning Techniques: A Case Study of Southern Part Nigeria

A key tool for digitalizing the agriculture sector and other industries is using big data and machine learning to predict farm produce. The inability of farmers to accurately predict yield is a great problem using previous farming experience. This study adopted three (3) machine learning approaches, including a decision tree classifier, random forest, and support vector machine, to model data from different zones and make predictions. The techniques adopted were tested using root mean square error to ensure the right prediction algorithm is adopted and the right values are obtained. Results show prediction from the South East is the best in terms of yields and accuracy when tested and evaluated, with 138.9 %.

Proteomics Landscape and Machine Learning Predictor of Response to Splenectomy in Immune Thrombocytopenia

Introduction Immune thrombocytopenia (ITP) is an auto-immune disease characterized by antibody-mediated platelet destruction. Splenectomy remains the effective curative treatment for patients with ITP who failure to respond to TPO-RAs and/or rituximab (Mageau et al, Am J Hematol. 2022). The effectiveness rate of splenectomy for ITP is reported to be roughly 80%, with a 5-year remission rate ranging from 50% to 75% (Chaturvedi et al, Blood. 2018). For subjects presenting with severely low platelet counts or active bleeding, splenectomy represents the primary approach for rapidly increasing platelet levels and potentially saving lives. To achieve personalized and precise treatment, the prediction efficacy on splenectomy remains an urgent problem. Here, we have characterized immune features of ITP patients with splenectomy, and developed a predictive model for assessing the response to splenectomy. Methods Bone marrow plasma samples before splenectomy were collected and analyzed by high-resolution liquid chromatography-tandem mass spectrometry. Weight protein co-expression network analysis (WPCNA) was applied to generate modules from all proteins and mapped them to clinical characteristics. We used least absolute shrinkage and selection operator (Lasso) regression, recursive feature elimination (RFE) for predictors selection. Next, multivariable logistic regression (LR), support vector machine (SVM), and decision tree algorithms were used to develop a predictive model. Results 60 patients with ITP undergoing splenectomy treatment were enrolled into our study, and divided into three groups according to the response [patients with CR (n=30), NR (n=7), and Relapse (n=23)]. A pathway enrichment analysis of WPCNA modules by ClueGO revealed that megakaryocyte differentiation, humoral immune response mediated by circulating immunoglobulin, Fc receptor signaling pathway, and cGMP-PKG signaling pathway were correspondingly enriched in MEtan, MEturquoise, MEblue, and MEpink modules, which were associated with treatments with TPO-RA, IVIG, rhTPO, and post-operation platelet level, respectively. We then performed KEGG pathway enrichment analyses on the different expression proteins (DEPs) to deciphering the immune landscape among each group. For the relapse group, DEPs were primarily enriched in ribosome, spliceosome, neutrophil extracellular trap formation, and necroptosis pathways, comparing with the CR group. For the NR group, most DEPs were enriched in natural killer cell mediated cytotoxicity, NF-kappa B signaling, Fc gamma R-mediated phagocytosis pathways, comparing with the CR group. The models were based on a multi-step predictor pipeline. Inside the pipeline, features were first filtered by lasso regression and RFE algorithm. We identified five proteomic features present in the bone marrow plasm that associated with relapsed to therapy, which included CD40, COL4A3, GALNT6, GSTM3, and UBE2A (Figure). Each ensemble consisted of three algorithms acting in parallel: LR, SVM and decision tree. The AUC values of LR, SVM, and decision tree were 0.962 (95%CI: 0.915-1), 0.957 (95%CI: 0.897-1), and 0.875 (95%CI: 0.782-0.967), respectively. The three algorithms were then tested for performance. The sensitivity, specificity, positive predictive value, and negative predictive value analysis revealed that LR model reached the best performance. Targeting CD40 represents a promising therapeutic approach for enhancing clinical outcomes in patients with ITP who have shown inadequate response following splenectomy. Conclusion Our analysis revealed that ITP patients with splenectomy showed distinct immune characteristics when achieving different response. Protein features extracted from bone marrow plasma can be selected as predictors of response to splenectomy.

Comparison of Machine Learning Methods for Satellite Image Classification: A Case Study of Casablanca Using Landsat Imagery and Google Earth Engine

Satellite image classification is crucial in various applications such as urban planning, environmental monitoring, and land use analysis. In this study, the authors present a comparative analysis of different supervised and unsupervised learning methods for satellite image classification, focusing on a case study in Casablanca using Landsat 8 imagery. This research aims to identify the most effective machine-learning approach for accurately classifying land cover in an urban environment. The methodology used consists of the pre-processing of Landsat imagery data from Casablanca city, the authors extract relevant features and partition them into training and test sets, and then use random forest (RF), SVM (support vector machine), classification, and regression tree (CART), gradient tree boost (GTB), decision tree (DT), and minimum distance (MD) algorithms. Through a series of experiments, the authors evaluate the performance of each machine learning method in terms of accuracy, and Kappa coefficient. This work shows that random forest is the best-performing algorithm, with an accuracy of 95.42% and 0.94 Kappa coefficient. The authors discuss the factors of their performance, including data characteristics, accurate selection, and model influencing.