Accuracy Of Classifier Research Articles

A terrain data collection sensor box towards a better analysis of terrains conditions

<p>Autonomous mobile robots are increasingly used across various applications, relying on multiple sensors for environmental awareness and efficient task execution. Given the unpredictability of human environments, versatility is crucial for these robots. Their performance is largely determined by how they perceive their surroundings. This paper introduces a machine learning (ML) approach focusing on land conditions to enhance a robot’s locomotion. The authors propose a method to classify terrains for data collection, involving the design of an apparatus to gather field data. This design is validated by correlating collected data with the output of a standard ML model for terrain classification. Experiments show that the data from this apparatus improves the accuracy of the ML classifier, highlighting the importance of including such data in the dataset.</p>

Novel maternal risk factors for preeclampsia prediction using machine learning algorithms

<div align="center"><table width="590" border="1" cellspacing="0" cellpadding="0"><tbody><tr><td valign="top" width="387"><p>Preeclampsia and eclampsia are the most common obstetric disorders associated with poor outcomes for women, fetuses, and newborns throughout the perinatal period. The study’s primary objective was to assess the accuracy of novel high-risk factors using MLA in predicting preeclampsia. The study included 400 pregnant women and used 27 novel high-risk factors to predict preeclampsia. The target variables for predicting preeclampsia are systolic and diastolic blood pressures. Various algorithms, including DT, RF, Gradient Boosting, SVM, K-Neighbors, LGBM, MLP, Ada Boost Classifier, and Extra Trees Classifier are used in the analysis. The accuracy and precision of the LGBM Classifier (0.85 and 0.9583 with F1 0.7188), Support Vector Classifier (0.8417 and 0.92 with F1 0.7077), Decision Tree (0.825 and 0.913 with F1 0.6667), and Extra Trees (0.8167 and 0.9091 with F1 0.6452) are found to be better algorithms for prediction of preeclampsia than those of the other models used in the study. According to the novel high-risk factors score, 17.5% of pregnant women were identified as being at high risk for preeclampsia during the first trimester, which increased to 18.7% in 3rd trimester; in addition, 16% of pregnant women had a BP of 140/90 mmHg and the above. Novel, high-risk scores and MLA can effectively predict preeclampsia at an early period.</p></td></tr></tbody></table></div>

Geological type recognition for shield machine using a semi-supervised variational auto-encoder-based adversarial method

In the process of tunneling, accurate and timely recognition of the geological type is significant to optimize the control parameters of the tunneling machine, improving tunneling efficiency and avoiding accidents. The shield machine operator in shield tunneling machine cannot directly observe the geological environment due to the closed working environment, so the soft method that can indirectly recognize the geological type by the machine parameters has become a research hotspot. However, most current soft methods use only a small amount of labeled data for supervised learning, and large amounts of unlabeled data is wasted. In order to use all data to improve the recognition performance of the classifier, a semi-supervised variational auto-encoder-based adversarial method (VAE-EMGAN) is proposed. Firstly, 50 parameters associated with geological types are selected and pre-processed, then the Variational Auto-Encoder (VAE) is trained by unlabeled data, and the generated part of VAE is added to the structure of Enhanced Multi-Classification Adversarial Generative Network (EMGAN) as a generator. Finally, the recognition accuracy of classifier is improved through adversarial training with labeled data, unlabeled data and generated data. We used data from upper and lower tunnels in Singapore to create two tasks to verify the validity and generalization performance of VEVE-EMGAN. The results show that the proposed model not only achieves high accuracy of all test sets on both tasks, but also has much better generalization performance than other models. Mean accuracy is 10.82%, 17.68%, 11.05%, 17.72%, 17.45%, 12.68% and 5.27% higher than SVM, KNN, RF, XGBoost, MLP, DNN and CNN respectively of test set 2 on task A; Mean accuracy is 13.06%, 12.80%, 7.64%, 18.31%, 8.74%, 7.94% and 4.05% higher than SVM, KNN, RF, XGBoost, MLP, DNN and CNN respectively of test set 2 on task B. In particular, the performance of the adversarial trained classifier is better than that has the same structure but separately trained classifier. Therefore, this method can use unlabeled data for adversarial training to improve the classification accuracy and generalization performance of the classifier, which has important implications for engineering practice.

Optimization of machine learning models for sentiment analysis in social media

The primary goal of this paper is to develop optimized classical machine learning models for sentiment analysis of social media posts. The methodology includes the application of k-fold cross-validation and feature selection using the χ2 independence test to prevent overfitting and improve classifier accuracy. Classical algorithms such as naïve Bayes, support vector machines, decision trees, and k-nearest neighbors are applied, along with preprocessing techniques, across five distinct datasets for model construction and evaluation. These optimized classical models are compared with recurrent neural network architectures, including LSTM and GRU, to evaluate the relative efficiency of both approaches. In total, 34 classification models were generated, with hyperparameter optimization for classical methods performed by grid search. The highest accuracy achieved was 82.45% for data without preprocessing and 78.83% for fully preprocessed data, both with the naïve Bayes algorithm. After hyperparameter optimization, some models achieved an accuracy greater than 90%. An analysis of variance indicated statistically significant differences among the models, confirming that feature selection and hyperparameter optimization are important factors in classifier performance. The proposed approach proved efficient for analyzing unstructured textual data, allowing the development of optimized models even with limited labeled data, and providing valuable insights into user opinions across different contexts.

Deep learning-based classifier for carcinoma of unknown primary using methylation quantitative trait loci.

Cancer of unknown primary (CUP) constitutes between 2% and 5% of human malignancies and is among the most common causes of cancer death in the United States. Brain metastases are often the first clinical presentation of CUP; despite extensive pathological and imaging studies, 20%-45% of CUP are never assigned a primary site. DNA methylation array profiling is a reliable method for tumor classification but tumor-type-specific classifier development requires many reference samples. This is difficult to accomplish for CUP as many cases are never assigned a specific diagnosis. Recent studies identified subsets of methylation quantitative trait loci (mQTLs) unique to specific organs, which could help increase classifier accuracy while requiring fewer samples. We performed a retrospective genome-wide methylation analysis of 759 carcinoma samples from formalin-fixed paraffin-embedded tissue samples using Illumina EPIC array. Utilizing mQTL specific for breast, lung, ovarian/gynecologic, colon, kidney, or testis (BLOCKT) (185k total probes), we developed a deep learning-based methylation classifier that achieved 93.12% average accuracy and 93.04% average F1-score across a 10-fold validation for BLOCKT organs. Our findings indicate that our organ-based DNA methylation classifier can assist pathologists in identifying the site of origin, providing oncologists insight on a diagnosis to administer appropriate therapy, improving patient outcomes.

Characterization of the Freshness of Pork by Near-Infrared Spectroscopy (NIRS) and Ensemble Learning

Pork is a perishable food and often needs to be stored in the refrigerator to maintain its quality as much as possible. Traditional methods for discriminating fresh and refrigerated pork are subjective, time-consuming, or destructive. The feasibility of using near-infrared (NIR) spectroscopy combined with chemometrics was explored to discriminate fresh and refrigerated pork. A total of 104 samples including 40 fresh and 64 refrigerated samples were first prepared and split into the training and test sets. Both partial least squares (PLS) and a subspace-based ensemble algorithm were used to establish classifiers. Also, both the number of learners and the size of subspace were optimized for ensemble modeling. On the independent test set, three measures, that is, the sensitivity, specificity, and total accuracy of the ensemble classifier were 95%, 93.8%, and 94.2%, respectively, each of which is superior to that of the PLS classifier. In addition, the influence of training set composition on classifier performance was also studied, indicating that ensemble modeling is robust. The results show that the NIR spectroscopy coupled with such an ensemble model can serve as a potential tool of discriminating fresh and refrigerated pork.

Health Big Data Classification Based on Collaborative Training Optimization Algorithm

In semisupervised learning, particularly in dealing with health big data classification problems, optimizing the performance of classifiers has always been a challenge. Accordingly, this study explores an optimization algorithm based on collaborative training to better handle health big data. First, the tri-training and decision tree classification models were selected for comparison. The average classification accuracy of the tri-training classification model was 4.20% higher than that of the decision tree classification model. Subsequently, the standard tri-training classifier was compared with these two classifiers. The classification accuracy of the standard tri-training classifier increased by 3.88% and 4.33%, respectively, compared with the aforementioned two classifiers. Finally, under the condition of 10% labeled samples, the performance of the collaborative training optimization algorithm was verified under three different basis classifiers. The results of this study demonstrate the effectiveness of optimization algorithms based on collaborative training in dealing with health big data classification problems. By improving the performance of the classifier, health big data can be predicted and analyzed more accurately, thereby improving the accuracy and efficiency of medical decision-making. Meanwhile, the application of this optimization algorithm also provides new research directions for other semisupervised learning problems.

Landslide Susceptibility Mapping Based on Ensemble Learning in the Jiuzhaigou Region, Sichuan, China

Accurate landslide susceptibility mapping is vital for disaster forecasting and risk management. To address the problem of limited accuracy of individual classifiers and lack of model interpretability in machine learning-based models, a coupled multi-model framework for landslide susceptibility mapping is proposed. Using Jiuzhaigou County, Sichuan Province, as a case study, we developed an evaluation index system incorporating 14 factors. We employed three base models—logistic regression, support vector machine, and Gaussian Naive Bayes—assessed through four ensemble methods: Stacking, Voting, Bagging, and Boosting. The decision mechanisms of these models were explained via a SHAP (SHapley Additive exPlanations) analysis. Results demonstrate that integrating machine learning with ensemble learning and SHAP yields more reliable landslide susceptibility mapping and enhances model interpretability. This approach effectively addresses the challenges of unreliable landslide susceptibility mapping in complex environments.

BIOM-19. METHYLATION PROFILING OF PLASMA CELL-FREE DNA IN PEDIATRIC BRAIN TUMOR PATIENTS

Abstract BACKGROUND Circulating tumor DNA (ctDNA) assays are being evaluated to inform clinical decisions in cancer care in adult patients, but the study in pediatric brain tumor patients is rare partly due to very low levels of ctDNA in blood. Currently, treatment decisions require invasive diagnostic surgical biopsies that carry risks and morbidity. The aim of this study is to utilize methylomes from plasma ctDNA for non-invasive diagnosis in pediatric brain tumor patients. METHODS Using cell-free methylated DNA immunoprecipitation and high-throughput sequencing (cfMeDIP-seq), which with low input DNA requirement, we measured methylation profiles of plasma samples from 77 pediatric brain tumor patients and 16 patients with non-neoplastic diseases like epilepsy. Binomial GLMnet classifiers of tumor or tumor subtype were built, in 2000 iterations of 60% training sets. Performance was evaluated in 40% test sets. RESULTS. These 77 pediatric brain tumor patients were a mixture of the most common pediatric brain tumors, include circumscribed astrocytic gliomas (N=23), glioneuronal tumors (N=16), embryonal tumor (N=9), ependymal tumor (N=7), pediatric type diffuse high grade glioma (N= 5), glioma NOS (N=4), mesenchymal tumor (N=4) and other pediatric brain tumors (N=9). The methylation profile differentiated brain tumors from non-neoplastic diseases with 82.9% accuracy (precision = 92.6%, sensitivity = 86.2%, specificity = 66.7 %). For subtype analysis, the methylation profile detected circumscribed astrocytic glioma from non-neoplastic diseases with 85.7% accuracy (precision = 87.5%, sensitivity = 87.5 %, specificity = 83.3 %), and glioneuronal tumors from non- neoplastic diseases with 83.3% accuracy (all precision, sensitivity, and specificity = 83.3 %). CONCLUSIONS The results suggest that methylation profiling of plasma cell-free DNA has the potential to discriminate between pediatric brain tumor patients and patients with non- neoplastic diseases. Larger population cohorts to train even more accurate classifiers will be needed.

An Automated Approach for Discriminating Hole Cleaning Efficiency While Predicting Penetration Rate in Egyptian Western Desert Wells

AbstractHigher rate of penetration (ROP) indicates successful drilling operation but is not the only drilling success measure. However, Conventional ROP prediction methods focus on increasing ROP and neglect the hole cleaning state, which can be altered by ROP changes. Higher ROP in vertical and deviated wells may increase cutting concentration, leading to hole cleaning problems such as overpulling and stuck pipe. With this problem in mind, this paper utilized geological, rheological, and drilling data of 31 vertical wells across four oil fields located in the Egyptian Western Desert, developed intelligent ROP prediction model through back propagation neural network (BPNN), and compared the proposed BPNN results with an empirical model. Finally, the pattern recognition algorithms including discriminant analysis, support vector machines, and neural network pattern recognition (NNPR) were implemented to discriminate hole cleaning efficiency following the ROP prediction process. Recognition models were developed based on predicted ROP, bit wear rate, specific energy, and drilling fluid carrying capacity index to evaluate hole cleaning. The accuracy of the multi-strategy classifier was evaluated using area under curve, confusion matrix, and receiver operating characteristic. The BPNN model outperformed the empirical model in terms of linear correlation coefficient (R = 98.6%) and average absolute error (AAE = 5.5%). Additionally, the best classification performance was achieved using the NNPR algorithm with 91% accuracy and a cross-validation error equal to zero. For validity, the proposed approach predicted ROP and classified hole cleaning efficiency for new vertical well in adjacent oil field, resulting in a 6% improvement in ROP.

Rethinking Multiple Instance Learning for Whole Slide Image Classification: A Bag-Level Classifier is a Good Instance-Level Teacher.

Multiple Instance Learning (MIL) has demonstrated promise in Whole Slide Image (WSI) classification. However, a major challenge persists due to the high computational cost associated with processing these gigapixel images. Existing methods generally adopt a two-stage approach, comprising a non-learnable feature embedding stage and a classifier training stage. Though it can greatly reduce memory consumption by using a fixed feature embedder pre-trained on other domains, such a scheme also results in a disparity between the two stages, leading to suboptimal classification accuracy. To address this issue, we propose that a bag-level classifier can be a good instance-level teacher. Based on this idea, we design Iteratively Coupled Multiple Instance Learning (ICMIL) to couple the embedder and the bag classifier at a low cost. ICMIL initially fixes the patch embedder to train the bag classifier, followed by fixing the bag classifier to fine-tune the patch embedder. The refined embedder can then generate better representations in return, leading to a more accurate classifier for the next iteration. To realize more flexible and more effective embedder fine-tuning, we also introduce a teacher-student framework to efficiently distill the category knowledge in the bag classifier to help the instance-level embedder fine-tuning. Intensive experiments were conducted on four distinct datasets to validate the effectiveness of ICMIL. The experimental results consistently demonstrated that our method significantly improves the performance of existing MIL backbones, achieving state-of-the-art results. The code and the organized datasets can be accessed by: https://github.com/Dootmaan/ICMIL/tree/confidence-based.

Filtering-Assisted Airborne Point Cloud Semantic Segmentation for Transmission Lines.

Point cloud semantic segmentation is crucial for identifying and analyzing transmission lines. Due to the number of point clouds being huge, complex scenes, and unbalanced sample proportion, the mainstream machine learning methods of point cloud segmentation cannot provide high efficiency and accuracy when extending to transmission line scenes. This paper proposes a filter-assisted airborne point cloud semantic segmentation for transmission lines. First, a large number of ground point clouds is identified by introducing the well-developed cloth simulation filter to alleviate the impact of the imbalance of the target object proportion on the classifier's performance. The multi-dimensional features are then defined, and the classification model is trained to achieve the multi-element semantic segmentation of the transmission line scene. The experimental results and analysis indicate that the proposed filter-assisted algorithm can significantly improve the semantic segmentation performance of the transmission line point cloud, enhancing both the point cloud segmentation efficiency and accuracy by more than 25.46% and 3.15%, respectively. The filter-assisted point cloud semantic segmentation method reduces the volume of sample data, the number of sample classes, and the sample imbalance index in power line scenarios to a certain extent, thereby improving the classification accuracy of classifiers and reducing time consumption. This research holds significant theoretical reference value and engineering application potential for scene reconstruction and intelligent understanding of airborne laser point cloud transmission lines.

A three-stage model pipeline predicting regional avalanche danger in Switzerland (RAvaFcast v1.0.0): a decision-support tool for operational avalanche forecasting

Abstract. Despite the increasing use of physical snow cover simulations in regional avalanche forecasting, avalanche forecasting is still an expert-based decision-making process. However, recently, it has become possible to obtain fully automated avalanche danger level predictions with satisfying accuracy by combining physically based snow cover models with machine learning approaches. These predictions are made at the location of automated weather stations close to avalanche starting zones. To bridge the gap between these local predictions and fully data- and model-driven regional avalanche danger maps, we developed and evaluated a three-stage model pipeline (RAvaFcast v1.0.0), involving the steps classification, interpolation, and aggregation. More specifically, we evaluated the impact of various terrain features on the performance of a Gaussian-process-based model for interpolation of local predictions to unobserved locations on a dense grid. Aggregating these predictions using an elevation-based strategy, we estimated the regional danger level and the corresponding elevation range for predefined warning regions, resulting in a forecast similar to the human-made public avalanche forecast in Switzerland. The best-performing model matched the human-made forecasts with a mean day accuracy of approximately 66 % for the entire forecast domain and 70 % specifically for the Alps. However, the performance depended strongly on the classifier's accuracy (i.e., a mean day accuracy of 68 %) and the density of local predictions available for the interpolation task. Despite these limitations, we believe that the proposed three-stage model pipeline has the potential to improve the interpretability of machine-made danger level predictions and has, thus, the potential to assist avalanche forecasters during forecast preparation, for instance, by being integrated in the forecast process in the form of an independent virtual forecaster.

AugmenToxic: Leveraging Reinforcement Learning to Optimize LLM Instruction Fine-Tuning for Data Augmentation to Enhance Toxicity Detection

Addressing the challenge of toxic language in online discussions is crucial for the development of effective toxicity detection models. This pioneering work focuses on addressing imbalanced datasets in toxicity detection by introducing a novel approach to augment toxic language data. We create a balanced dataset by instructing fine-tuning of Large Language Models (LLMs) using Reinforcement Learning with Human Feedback (RLHF). Recognizing the challenges in collecting sufficient toxic samples from social media platforms for building a balanced dataset, our methodology involves sentence-level text data augmentation through paraphrasing existing samples using optimized generative LLMs. Leveraging generative LLM, we utilize the Proximal Policy Optimizer (PPO) as the RL algorithm to fine-tune the model further and align it with human feedback. In other words, we start by fine-tuning a LLM using an instruction dataset, specifically tailored for the task of paraphrasing while maintaining semantic consistency. Next, we apply PPO and a reward function, to further fine-tune (optimize) the instruction-tuned LLM. This RL process guides the model in generating toxic responses. We utilize the Google Perspective API as a toxicity evaluator to assess generated responses and assign rewards/penalties accordingly. This approach guides LLMs through PPO and the reward function, transforming minority class samples into augmented versions. The primary goal of our methodology is to create a balanced and diverse dataset to enhance the accuracy and performance of classifiers in identifying instances from the minority class. Utilizing two publicly available toxic datasets, we compared various techniques with our proposed method for generating toxic samples, demonstrating that our approach outperforms all others in producing a higher number of toxic samples. Starting with an initial 16,225 toxic prompts, our method successfully generated 122,951 toxic samples with a toxicity score exceeding 30%. Subsequently, we developed various classifiers using the generated balanced datasets and applied a cost-sensitive learning approach to the original imbalanced dataset. The findings highlight the superior performance of classifiers trained on data generated using our proposed method. These results highlight the importance of employing RL and a data-agnostic model as a reward mechanism for augmenting toxic data, thereby enhancing the robustness of toxicity detection models.

Automatic categorization of medical documents in Afaan Oromo using ensemble machine learning techniques

Automatic medical document classification using machine learning techniques can enhance the productivity of healthcare services by reducing processing time and cost. This work proposes an ensemble learning approach to develop a model that classifies electronic medical documents in Afaan Oromo. The main tasks in this work are preparing the corpus, pre-processing, training the models, and the classification process. We used the term frequency-inverse document frequency (TF-IDF) and bag of words (BOW) feature extraction methods. An ensemble technique in this work is that it creates multiple individual classifier predictions from naïve Bayes, random forest, SVM, and logistic regression and then combines them to advance a reliable and more accurate classifier. Evaluation measures were employed using accuracy, F1-score, recall, and precision for performance comparison. The efficiency of the proposed method is compared with the two existing boosting approaches, namely gradient boosting and adaboost. The experimental result shows the efficiency of BOW feature extraction over TF-IDF in this work on our dataset. These results also illustrated the effectiveness of the proposed model by scoring 94.81% accuracy and 94.84% F1-score. This work significantly contributes to the technological enhancement of service delivery, managing documents through classification methods, and advancing the data processing systems in healthcare sectors.

Deep learning techniques for quality of transmission estimation in optical networks

A large body of research has recently examined the estimation of the quality of transmission (QoT) in optical networks with deep learning. This paper discusses a lightpath’s quality of transmission to design fiber-optic communication and networks using deep learning algorithms. We need different major estimation parameters for advanced optical fiber communication and networks, i.e., modulation formats, baud rate, and code rate. Currently, the quality of transmission for unspecified optical paths depends on different estimation techniques i.e., (1) analytical models estimating physical layer impairments (PLIs) and (2) margined formulas. This paper focuses on deep-learning techniques that can be applied to optimization and complex systems. The deep learning algorithms contain different classifiers that can simulate results and estimate the bit-error rate, and signal-to-noise ratio of unspecified optical paths with threshold values, traffic volume, and modulation format. We must train and test the datasets for various classifiers, and classification features using Korean network topology. The classifier accuracy and Area Under the ROC Curve (AUC) simulation results are carried out using MATLAB.

The Potential for Hyperspectral Imaging and Machine Learning to Classify Internal Quality Defects in Macadamia Nuts

Tree nuts are rich in nutrients, and global production and consumption have doubled during the last decade. However, nuts have a range of quality defects that must be detected and removed during post-harvest processing. Tree nuts can develop hidden internal discoloration, and current sorting methods are prone to subjectivity and human error. Therefore, non-destructive, real-time methods to evaluate internal nut quality are needed. This study explored the potential for VNIR (400–1000 nm) hyperspectral imaging to classify brown center disorder in macadamias. This study compared the accuracy of classifiers developed using images of kernels imaged in face-up and face-down orientations. Classification accuracy was excellent using face-up (>97.9%) and face-down (>94%) images using ensemble and linear discriminate models before and after wavelength selection. Combining images to form a pooled dataset also provided high accuracy (>90%) using artificial neural network and support vector machine models. Overall, HSI has great potential for commercial application in nut processing to detect internal brown centers using images of the outside kernel surface in the VNIR range. This technology will allow rapid and non-destructive evaluation of intact nut products that can then be marketed as a high-quality, defect-free product, compared with traditional methods that rely heavily on representative sub-sampling.

Exploring the role of project status information in effective code smell detection

Repairing code smells detected in the code or design of the system is one of the activities contributing to increasing the software quality. In this study, we investigate the impact of non-numerical information of software, such as project status information combined with machine learning techniques, on improving code smell detection. For this purpose, we constructed a dataset consisting of 22 systems with various project statuses, 12,040 classes, and 18 features that included 1935 large classes. A set of experiments was conducted with ten different machine learning techniques by dividing the dataset into training, validation, and testing sets to detect the large class code smell. Feature selection and data balancing techniques have been applied. The classifier’s performance was evaluated using six indicators: precision, recall, F-measure, MCC, ROC area, and Kappa tests. The preliminary experimental results reveal that feature selection and data balancing have poor influence on the accuracy of machine learning classifiers. Moreover, they vary their behavior when utilized in sets with different values for the selected project status information of their classes. The average value of classifiers performance when fed with status information is better than without. The Random Forest achieved the best behavior according to all performance indicators (100%) with status information, while AdaBoostM1 and SMO achieved the worst in most of them (> 86%). According to the findings of this study, providing machine learning techniques with project status information about the classes to be analyzed can improve the results of large class detection.

Accuracy Assessment of Land Use Land Cover Classification Using Machine Learning Classifiers in Google Earth Engine; A Case Study of Jammu District

Abstract. LULC (Land Use and Land Cover) involves classifying and describing different land types and their usage. Using satellite imagery for LULC mapping is increasing in remote sensing. This study focuses on Jammu district in India, situated between mountain ranges from north and south make it eco-sensitive zone. Expanding of human activity and loss of natural resources make it vulnerable if mismanaged. Study of LULC is important because of it and this study deals with efficiency and accuracy of various machine learning classifiers for LULC. This study uses machine learning classifiers - Random Forest (RF), Support Vector Machine (SVM), Gradient Boosted Trees (GTB), and Classification and Regression Trees (CART) - for the task, leveraging Google Earth Engine (GEE). Sentinel-2 satellite data from January 1 to March 31, 2023, with specific spectral bands (B4, B3, B2 & B8, B4, B3), were used for image preprocessing. Six classes: built-up, water, agricultural land, fallow land, forest, and barren land, each with 100 sample points, were used for classification. After training, the classifiers' accuracy was evaluated using Overall Accuracy and Kappa Coefficient. The results showed RF as the top performer with an overall accuracy of 99.36% and a Kappa coefficient of 99.11%, followed by SVM, GTB, and CART. This highlights the effectiveness of machine learning classifiers, especially RF and SVM, in accurately mapping LULC patterns in Jammu district, suggesting RF's potential as a reliable tool for remote sensing-based LULC mapping.

Confidential outsourced support vector machine learning based on well-separated structure

Support Vector Machine (SVM) has revolutionized various domains and achieved remarkable successes. This progress relies on subtle algorithms and more on large training samples. However, the massive data collection introduces security concerns. To facilitate secure integration of data efficiently for building an accurate SVM classifier, we present a non-interactive protocol for privacy-preserving SVM, named NPSVMT. Specifically, we define a new well-separated structure for computing gradients that can decouple the fusion matter between user data and model parameters, allowing data providers to outsource the collaborative learning task to the cloud. As a result, NPSVMT is capable of removing the multiple communications and eliminating the straggler’s effect (waiting for the last), thereby going beyond those developed with interactive methods, e.g., federated learning. To further decrease the data traffic, we introduce a high-efficient coding method to compress and parse training data. In addition, unlike outsourced schemes based on homomorphic encryption or secret sharing, NPSVMT exploits functional encryption to maintain the data confidentiality, achieving dropout-tolerant secure aggregation. The implementations verify that NPSVMT is faster by orders of magnitude than the existing privacy-preserving SVM schemes on benchmark datasets.