Support Vector Machine For Classification Research Articles

Classification of flood disaster level news articles using Machine Learning

Floods have a significant socio-economic impact on Indonesian society. Much of this information is sourced from online news articles and social media. This research investigates whether the Support Vector Machine (SVM) method can be used for flood disaster level classification (low, medium, and high). Our methodology involves preparing data extracted from textual news articles on the National Disaster Management Agency (BNPB) website on the topic of flooding. We then labeled the data according to Regulation No. 02/2012 on general guidelines for disaster assessment and used the Support Vector Machine (SVM) method. Training and testing were conducted using different datasets, followed by accuracy and error evaluation. In addition, we considered the performance comparison of SVM with other classification methods, including Decision Tree, Naive Bayes, Adaboost, Random Forest, and Xgboost. The experimental results show that SVM still does not get good accuracy results for flood disaster level classification. The SVM accuracy level result of (52%) is still low compared to Random Forest (78%), and Xgboost (68%). Further research is expected to increase the accuracy of SVM for flood level classification.

An integrated deep-learning model for smart waste classification.

Efficient waste management is essential for human well-being and environmental health, as neglecting proper disposal practices can lead to financial losses and the depletion of natural resources. Given the rapid urbanization and population growth, developing an automated, innovative waste classification model becomes imperative. To address this need, our paper introduces a novel and robust solution - a smart waste classification model that leverages a hybrid deep learning model (Optimized DenseNet-121 + SVM) to categorize waste items using the TrashNet datasets. Our proposed approach uses the advanced deep learning model DenseNet-121, optimized for superior performance, to extract meaningful features from an expanded TrashNet dataset. These features are subsequently fed into a support vector machine (SVM) for precise classification. Employing data augmentation techniques further enhances classification accuracy while mitigating the risk of overfitting, especially when working with limited TrashNet data. The results of our experimental evaluation of this hybrid deep learning model are highly promising, with an impressive accuracy rate of 99.84%. This accuracy surpasses similar existing models, affirming the efficacy and potential of our approach to revolutionizing waste classification for a sustainable and cleaner future.

ChestCovidNet: An Effective DL-based Approach for COVID-19, Lung Opacity, and Pneumonia Detection Using Chest Radiographs Images.

Currently used lung disease screening tools are expensive in terms of money and time. Therefore, chest radiograph images (CRIs) are employed for prompt and accurate COVID-19 identification. Recently, many researchers have applied Deep learning (DL) based models to detect COVID-19 automatically. However, their model could have been more computationally expensive and less robust, i.e., its performance degrades when evaluated on other datasets. This study proposes a trustworthy, robust, and lightweight network (ChestCovidNet) that can detect COVID-19 by examining various CRIs datasets. The ChestCovidNet model has only 11 learned layers, eight convolutional (Conv) layers, and three fully connected (FC) layers. The framework employs both the Conv and group Conv layers, Leaky Relu activation function, shufflenet unit, Conv kernels of 3×3 and 1×1 to extract features at different scales, and two normalization procedures that are cross-channel normalization and batch normalization. We used 9013 CRIs for training whereas 3863 CRIs for testing the proposed ChestCovidNet approach. Furthermore, we compared the classification results of the proposed framework with hybrid methods in which we employed DL frameworks for feature extraction and support vector machines (SVM) for classification. The study's findings demonstrated that the embedded low-power ChestCovidNet model worked well and achieved a classification accuracy of 98.12% and recall, F1-score, and precision of 95.75%.

Evaluating Effects of Resting-State Electroencephalography Data Pre-Processing on a Machine Learning Task for Parkinson's Disease.

Resting-state electroencephalography pre-processing methods in machine learning studies into Parkinson's disease classification vary widely. Here three separate data sets were pre-processed to four different stages to investigate the effects on evaluation metrics, using power features from six regions-of-interest, Random Forest Classifiers for feature selection, and Support Vector Machines for classification. This showed muscle artefact inflated evaluation metrics, and alpha and theta band features produced the best results when fully pre-processing data.

Performance evaluation of ferro-fluids flooding in enhanced oil recovery operations based on machine learning

The process of enhanced oil recovery (EOR) and core flooding involves various challenges such as preserving cores, configuring experiment setup, scaling from the laboratory to the field, interpreting experimental data, handling interactions between multiple EOR agents, and performing cost-effective evaluations. However, recent advancements in artificial intelligence and deep learning have facilitated the prediction of EOR efficiency, simplifying the process while saving time and money - two vital aspects of petroleum engineering. This study employed deep learning to determine the impact of ferro-fluids on EOR efficiency, which reduces excessive laboratory analysis costs and saves time. The findings may aid in optimizing injectable materials for future EOR projects. Four machine learning (ML) algorithms were utilized to predict the efficacy of ferro-fluids in carbonate and sandstone core flooding: support vector machine (SVM) for classification, multilayer perceptron (MLP), radial basis function (RBF), artificial neural network (ANN), and adaptive neuro-fuzzy inference system (ANFIS) for regression. The models were trained initially, followed by creating a set of validating measurements through classification to determine the regression trend of ferro-fluids. The SVM accurately classified all data points and the resulting regressions were satisfactory, with MLP and ANFIS producing R2 values of 0.9957 and 0.9922, respectively.

Improved SVM Algorithm Financial Management Model for Data Mining

Financial crisis early warning is an important link in financial management. With its relatively simple structure and excellent classification performance, support vector machine is often used in intelligent warning of financial crises. A financial warning model based on improved support vector machine is proposed to address the shortcomings of current financial warning methods that cannot handle large-scale data well, and the accuracy and efficiency are not ideal. In response to the excessive indicators in the financial early warning indicator system, the principal component analysis method is studied for dimensionality reduction, reducing the computational complexity of the model, and improving the training and operational efficiency of the model. In response to the limited practicality of support vector machine, which can only search for the optimal solution under constraint conditions, the use of smooth support vector machine to replace support vector machine for predictive classification is studied. Finally, a financial crisis early warning model is constructed based on smooth support vector machine using principal component analysis for dimensionality reduction of input data. The performance of the model is tested using an internet listed company as an example. The findings demonstrate that Model 1 outperforms the other three models in terms of accuracy rate (95%), MAE, MSE and [Formula: see text] values (0.162, 0.174 and 0.169, respectively). Therefore, the financial crisis early warning model constructed by the research can properly forecast the company’s financial situation, thereby helping the enterprise to develop better.

Towards a true conservation zero tillage system: “A proposed solution based on computer vision to herbicide resistance.”

The Zero Tillage (ZT) or Non-Tillage system constitutes an agricultural production approach designed for improved soil conservation, reduced fossil fuel usage, mitigation of waterway pollution, water and wind erosion, and addressing soil compaction, among other objectives. In this way, ZT promises more sustainable agriculture. However, current ZT production systems depend on herbicide applications for weed control. The use of herbicides with the same modes of action over many years has led to numerous resistant weed species, which end up threatening the success of the herbicide weed control and, consequently, the overall success of the ZT system. Being able to automatically detect and classify weed species directly under the sprayer during the chemical application is an alternative to control populations of weeds since it would allow selecting the herbicide and dose for those particular species. This study, conducted over 15 years in commercial fields utilizing Zero Tillage (ZT) management, examined various ground covers intrinsic to the ZT system originating from previous crops or residues post-harvest. ZT features a broad spectrum of contexts due to the complexity of different ground cover types, which challenge computer vision techniques. The assessment focused on the automatic detection and classification of among the most problematic monocotyledonous and dicotyledonous weed species in the province of Cordoba in Argentina, with the objective of enabling a more targeted and selective approach to their control. These are Amaranthus palmeri, Echinochloa crus-galli, Eleusine indica, Parietaria debilis and Conyza sumatrensis. Additionally, some species belong to the same botanical families with several morphological resemblances that defy vision algorithms. This work proceeded through a machine learning approach applied to computer vision features such as SIFT (scale-invariant feature transform), K-means (clustering), and Bag of Features over the Support Vector Machine for classification. The results showed an accuracy between 89 % and 98 % over the species, allowing a significant reduction of herbicide while applying the adequate active ingredient. Moreover, virtual binary maps from weed patches are devised to be implemented through ISOBUS protocol. Thus the current research contributes significantly to the issue of controlling populations of weeds in the ZT agriculture system.

Label-free detection of the cytotoxicity effect of cisplatin in human leukemic cells using Raman spectroscopy in conjunction with multivariate analysis.

The single-cell Raman spectra of human leukemic Jurkat cells can be obtained by confocal microscopy Raman spectroscopy, including cell groups treated with different doses of cisplatin (0, 3.5, 7, 10.5 and 14 μmol L-1) for 24 hours and those treated with 10.5 μmol L-1 cisplatin for different times (0, 6, 12, 24 and 36 hours). The structure and amount of protein, nucleic acid and other major molecules from different cell groups show special changes in the percentage of biochemical constituents. Compared with the control group, the two protein Raman bands (1449 and 1659 cm-1) and two DNA bands (1303 and 1338 cm-1) in the treatment groups decrease in intensity with the increase of the drug dose and treatment time of cisplatin. Partial least squares combined with support vector machines was used to develop diagnostic algorithms for distinguishing between control and treatment groups. The support vector machines for classification between the control group (0 μmol L-1) and cell groups treated with 10.5 and 14 μmol L-1 cisplatin for 24 hours have achieved good diagnostic results with a high sensitivity of 100%, specificity of 100% and accuracy of 100%, respectively, indicating that 10.5 μmol L-1 can be used as an appropriate therapeutic dose. Using the same method, the diagnostic sensitivity, specificity and accuracy between the control group (0 hours) and cell groups treated with 10.5 μmol L-1 cisplatin for 24 and 36 hours are all 100%, showing that 24 hours can be used as an appropriate therapeutic time. These results showed that Raman spectroscopy in conjunction with multivariate statistical analysis could be a useful tool for evaluating the cytotoxicity induced by cisplatin in human leukemic cells.

Comparative analysis of k-nearest neighbor and support vector machine in classification of COVID 19 disease in Makassar City

Comparative analysis of k-nearest neighbor and support vector machine in classification of COVID 19 disease in Makassar City

Comparative Analysis of Naïve Bayes Classifier, Support Vector Machine and Decision Tree in Rainfall Classification Using Confusion Matrix

Comparative Analysis of Naïve Bayes Classifier, Support Vector Machine and Decision Tree in Rainfall Classification Using Confusion Matrix

Automated Tomato Leaf Disease Detection Using Image Processing: An SVM‐Based Approach with GLCM and SIFT Features

Tomato cultivation is increasingly widespread, yet it faces significant challenges, particularly from plant diseases caused by fungi, bacteria, and insects. Addressing these diseases is crucial for ensuring the quality and yield of tomato crops. To support specialists in accurately identifying and managing these diseases, we propose an advanced automatic system for detecting and identifying tomato leaf diseases using sophisticated image processing techniques. Therefore, we proposed an approach that employs robust feature extraction methods, including the gray level co‐occurrence matrix (GLCM) and scale‐invariant feature transform (SIFT), coupled with a support vector machine (SVM) for adequate classification. We curated an extensive dataset of 2700 tomato leaf images, with a minimum of 300 images for each of the nine distinct disease classes. This comprehensive dataset facilitated the training and testing of various machine learning and deep learning models. The experimental results highlight our proposed approach’s exceptional accuracy and reliability, significantly improving the detection and classification of tomato leaf diseases. A thorough comparative analysis with contemporary state‐of‐the‐art techniques further validates the superiority of our system. Our findings suggest that this framework can significantly benefit tomato cultivation by enabling timely and precise disease management. Future research can explore integrating advanced deep learning algorithms to enhance the system’s accuracy in this multiclass classification challenge.

Exploring the Teaching Reform Path of Ideological and Political Education in Colleges and Universities in the Context of Deep Learning

Abstract Deep learning significantly enhances the comprehension of ideological and political education (IPE) concepts and facilitates the advancement of IPE teaching in colleges and universities through bilinear blended teaching reforms. This innovative teaching approach employs association rules from big data and an enhanced Apriori algorithm to analyze students’ online learning data. An analytical model of student learning behaviors in IPE, utilizing a Support Vector Machine (SVM) for learner classification, forms the basis for developing an online learner profile. This profile integrates assessments of learning outcomes, group identification, and style classification to tailor learning activities to individual needs. The efficacy of this reform was tested over one year, comparing the impacts of IPE across four different regions. Results indicate that students in the experimental group significantly outperformed their counterparts in the control group. For instance, in Shanghai, scores from the experimental group averaged 89.63 compared to 83.25 in the control group, yielding a significant difference of 6.38 points. Statistical analysis showed a t-value of 6.21 and a p-value of 0.000, affirming the positive impact of the bilinear blended teaching reform on enhancing student ideological engagement and promoting educational reform in colleges.

Comparison of support vector machine and random forest algorithms for classification of songs for relaxation purposes in individuals with stress disorders

Comparison of support vector machine and random forest algorithms for classification of songs for relaxation purposes in individuals with stress disorders

Acoustic and Text Features Analysis for Adult ADHD Screening: A Data-Driven Approach Utilizing DIVA Interview.

Attention Deficit Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder commonly seen in childhood that leads to behavioural changes in social development and communication patterns, often continues into undiagnosed adulthood due to a global shortage of psychiatrists, resulting in delayed diagnoses with lasting consequences on individual's well-being and the societal impact. Recently, machine learning methodologies have been incorporated into healthcare systems to facilitate the diagnosis and enhance the potential prediction of treatment outcomes for mental health conditions. In ADHD detection, the previous research focused on utilizing functional magnetic resonance imaging (fMRI) or Electroencephalography (EEG) signals, which require costly equipment and trained personnel for data collection. In recent years, speech and text modalities have garnered increasing attention due to their cost-effectiveness and non-wearable sensing in data collection. In this research, conducted in collaboration with the Cumbria, Northumberland, Tyne and Wear NHS Foundation Trust, we gathered audio data from both ADHD patients and normal controls based on the clinically popular Diagnostic Interview for ADHD in adults (DIVA). Subsequently, we transformed the speech data into text modalities through the utilization of the Google Cloud Speech API. We extracted both acoustic and text features from the data, encompassing traditional acoustic features (e.g., MFCC), specialized feature sets (e.g., eGeMAPS), as well as deep-learned linguistic and semantic features derived from pre-trained deep learning models. These features are employed in conjunction with a support vector machine for ADHD classification, yielding promising outcomes in the utilization of audio and text data for effective adult ADHD screening. Clinical impact: This research introduces a transformative approach in ADHD diagnosis, employing speech and text analysis to facilitate early and more accessible detection, particularly beneficial in areas with limited psychiatric resources. Clinical and Translational Impact Statement: The successful application of machine learning techniques in analyzing audio and text data for ADHD screening represents a significant advancement in mental health diagnostics, paving the way for its integration into clinical settings and potentially improving patient outcomes on a broader scale.

Improving Support vector machine for Imbalanced big data classification

A significant proportion of one type of pattern and a relatively small quantity of another type of pattern can be found in many unbalanced real data sets. In addition, finding significant observations and excluding influential observations is effectively accomplished through diagnostic analysis. Support vector machines (SVM), a common classification technique, perform poorly on imbalanced datasets and when influential observations exist. In this research, the pigeon optimization algorithm as a metaheuristic algorithm is employed to address the influence observation issues in SVM. Experiments are done on three real sets of data. Our approach provides higher classification accuracy compared to other widely used algorithms. This approach could be used for further biological, chemical, and medical datasets.

An efficient Analysis based on the Internet of Things, SVM and KNN for Operative Diabetic Retinopathy Classification

These days, diabetes is an incurable disease, with millions of people suffering from it worldwide. Several variables namely lack of education, crowded living conditions, obesity and improper diet are among the causes of this recent upsurge in diabetes cases. They are identified by the name of infections induced by bacteria or viruses, harmful compounds in food, autoimmune reactions, obesity, unhealthy lifestyles, and pollution in the environment. Excessive and sight-threatening diabetic retinopathy (DR) is the most common retinal micro-vascular dysfunction that is characterized by the occurrence of a disorder of retinal blood vessels resulting in impaired vision. The IoT-based work is conducted in this work on the machine learning (ML) techniques, K-Nearest Neighbor (KNN) and Support Vector Machine (SVM). The classification of diabetic retinopathy is a topic that is under research. The range of activities of the processes of downsampling, labelling, image flattening, and format conversion is all within the dataset preparation process. An advanced prognosis model is designed which follows a combination of two machine learning techniques such as SVM and KNN. This approach classifies the images of diabetic retinopathy into five segments (subclasses), thus facilitating in-depth analysis. Our solution proposal in this case is a superior one because of its higher classification accuracy and faster processing speed as the findings showed. The robustness and accuracy that the SVM is known for are ensured by the convergence of the KNN to the SVM. The paper also proves a close linkage of clinical symptoms and blood sugar readings to an algorithmic DM prediction system that is based on IoT and ML approaches. This is another advantage of this method that it outperforms the existing classification methods. Amongst all the classifiers that we used in this project, the KNN ML classifier turned out to be the most accurate one with an accuracy rate of 93%. It was found that the algorithm performed with a 79% accuracy rate after tough testing and training and it was consistently providing number one quality DM predictions.

Multi-MicroRNA Analysis Can Improve the Diagnostic Performance of Mammography in Determining Breast Cancer Risk.

The objective of this study was to determine whether multi-microRNA analysis using a combination of four microRNA biomarkers (miR-1246, 202, 21, and 219B) could improve the diagnostic performance of mammography in determining breast cancer risk by age group (under 50 vs. over 50) and distinguish breast cancer from benign breast diseases and other cancers (thyroid, colon, stomach, lung, liver, and cervix cancers). To verify breast cancer classification performance of the four miRNA biomarkers and whether the model providing breast cancer risk score could distinguish between benign breast disease and other cancers, the model was verified using nonlinear support vector machine (SVM) and generalized linear model (GLM) and age and four miRNA qRT-PCR analysis values (dCt) were input to these models. Breast cancer risk scores for each Breast Imaging-Reporting and Data System (BI-RADS) category in multi-microRNA analysis were analyzed to examine the correlation between breast cancer risk scores and mammography categories. We generated two models using two classification algorithms, SVM and GLM, with a combination of four miRNA biomarkers showing high performance and sensitivities of 84.5% and 82.1%, a specificity of 85%, and areas under the curve (AUCs) of 0.967 and 0.965, respectively, which showed consistent performance across all stages of breast cancer and patient ages. The results of this study showed that this multi-microRNA analysis using the four miRNA biomarkers was effective in classifying breast cancer in patients under the age of 50, which is challenging to accurately diagnose. In addition, breast cancer and benign breast diseases can be classified, showing the possibility of helping with diagnosis by mammography. Verification of the performance of the four miRNA biomarkers confirmed that multi-microRNA analysis could be used as a new breast cancer screening aid to improve the accuracy of mammography. However, many factors must be considered for clinical use. Further validation with an appropriate screening population in large clinical trials is required. This trial is registered with (KNUCH 2022-04-036).

Application of a Low-Cost mHealth Solution for the Remote Monitoring of Patients With Epilepsy: Algorithm Development and Validation

Background Implementing automated seizure detection in long-term electroencephalography (EEG) analysis enables the remote monitoring of patients with epilepsy, thereby improving their quality of life. Objective The objective of this study was to explore an mHealth (mobile health) solution by investigating the feasibility of smartphones for processing large EEG recordings for the remote monitoring of patients with epilepsy. Methods We developed a mobile app to automatically analyze and classify epileptic seizures using EEG. We used the cross-database model developed in our previous study, incorporating successive decomposition index and matrix determinant as features, adaptive median feature baseline correction for overcoming interdatabase feature variation, and postprocessing-based support vector machine for classification using 5 different EEG databases. The Sezect (Seizure Detect) Android app was built using the Chaquopy software development kit, which uses the Python language in Android Studio. Various durations of EEG signals were tested on different smartphones to check the feasibility of the Sezect app. Results We observed a sensitivity of 93.5%, a specificity of 97.5%, and a false detection rate of 1.5 per hour for EEG recordings using the Sezect app. The various mobile phones did not differ substantially in processing time, which indicates a range of phone models can be used for implementation. The computational time required to process real-time EEG data via smartphones and the classification results suggests that our mHealth app could be a valuable asset for monitoring patients with epilepsy. Conclusions Smartphones have multipurpose use in health care, offering tools that can improve the quality of patients’ lives.

Blend of Deep Features and Binary Tree Growth Algorithm for Skin Lesion Classification

One of the most frequently identified cancers globally is skin cancer (SC). The computeraided categorization of numerous skin lesions via dermoscopic images is still a complicated problem. Early recognition is crucial since it considerably increases the survival chances. In this study, we introduce an approach for skin lesion categorization where, at first, a powerful hybrid deep-feature set is constructed, and then a binary tree growth (BTG)-based optimization procedure is implemented using a support vector machine (SVM) classifier with an intention to compute the categorizing error and build symmetry between categories, for selecting the most significant features which are finally fed to a multi-class SVM for classification. The hybrid deep-feature set is constructed by utilizing two pre-trained models, i.e., Densenet-201, and Inception-v3, that are fine-tuned on skin lesion data. These two deep-feature models have distinct architectures that characterize dissimilar feature abstraction strengths. This effective deep feature framework has been tested on two publicly available challenging datasets, i.e., ISIC2018 and ISIC2019. The proposed framework outperforms many existing approaches and achieves notable {accuracy, sensitivity, precision, specificity} values of {98.50%, 96.60%, 97.84%, 99.59%} and {96.60%, 94.21%, 96.38%, 99.39%} for the ISIC2018 and ISIC2019 datasets, respectively. The proposed implementation of the BTG-based optimization algorithm performs significantly better on the proposed feature blend for skin lesion classification.

Machine Learning in the Hyperspectral Classification of Glycaspis brimblecombei (Hemiptera Psyllidae) Attack Severity in Eucalyptus

Assessing different levels of red gum lerp psyllid (Glycaspis brimblecombei) can influence the hyperspectral reflectance of leaves in different ways due to changes in chlorophyll. In order to classify these levels, the use of machine learning (ML) algorithms can help process the data faster and more accurately. The objectives were: (I) to evaluate the spectral behavior of the G. brimblecombei attack levels; (II) find the most accurate ML algorithm for classifying pest attack levels; (III) find the input configuration that improves performance of the algorithms. Data were collected from a clonal eucalyptus plantation (clone AEC 0144—Eucalyptus urophilla) aged 10.3 months old. Eighty sample evaluations were carried out considering the following severity levels: control (no shells), low infestation (N1), intermediate infestation (N2), and high infestation (N3), for which leaf spectral reflectances were obtained using a spectroradiometer. The spectral range acquired by the equipment was 350 to 2500 nm. After obtaining the wavelengths, they were grouped into representative interval means in 28 bands. Data were submitted to the following ML algorithms: artificial neural networks (ANN), REPTree (DT) and J48 decision trees, random forest (RF), support vector machine (SVM), and conventional logistic regression (LR) analysis. Two input configurations were tested: using only the wavelengths (ALL) and using the spectral bands (SB) to classify the attack levels. The output variable was the severity of G. brimblecombei attack. There were differences in the hyperspectral behavior of the leaves for the different attack levels. The highest attack level shows the greatest distinction and the highest reflectance values. LR and SVM show better accuracy in classifying the severity levels of G. brimblecombei attack. For the correct classification percentage, the RL and SVM algorithms performed better, both with accuracy above 90%. Both algorithms achieved F-score values close to 0.90 and above 0.8 for Kappa. The entire spectral range guaranteed the best accuracy for both algorithms.