Support Vector Machine Technique Research Articles

CNN‐based defect detection in manufacturing

AbstractThis research introduces an advanced algorithm based on convolutional neural networks for the detection and categorization of surface defects in manufacturing processes. At its core, the algorithm employs a deep learning model that integrates residual networks and attention mechanisms to effectively extract features. Additionally, we have developed a novel feature selection method, named NR, which synergistically combines neighborhood component analysis and ReliefF techniques. This approach enables the selection of more representative deep features for subsequent analysis. For the classification task, we utilize the support vector machine technique, which demonstrates versatility in handling both binary and multi‐class classification scenarios. The reliability and superiority of our algorithm are further validated through a comparative analysis using a dataset specifically tailored for this context. The results indicate that our approach outperforms existing algorithms in accurately identifying manufacturing defects.

Spatial analysis and machine learning prediction of forest fire susceptibility: a comprehensive approach for effective management and mitigation

Forest fires (FF) in tropical seasonal forests impact ecosystem. Addressing FF in tropical ecosystems has become a priority to mitigate impacts on biodiversity loss and climate change. The escalating frequency and intensity of FF globally have become a mounting concern. Understanding their tendencies, patterns, and vulnerabilities is imperative for conserving ecosystems and facilitating the development of effective prevention and management strategies. This study investigates the trends, patterns, and spatiotemporal distribution of FF for the period of 2001–2022, and delineates the forest fire susceptibility zones in Odisha State, India. The study utilized: (a) MODIS imagery to examine active fire point data; (b) Kernel density tools; (c) FF risk prediction using two machine learning algorithms, namely Support Vector Machine (SVM) and Random Forest (RF); (d) Receiver Operating Characteristic and Area Under the Curve, along with various evaluation metrics; and (e) a total of 19 factors, including three topographical, seven climatic, four biophysical, and five anthropogenic, to create a map indicating areas vulnerable to FF. The validation results revealed that the RF model achieved a precision exceeding 94 % on the validation datasets, while the SVM model reached 89 %. The estimated forest fire susceptibility zones using RF and SVM techniques indicated that 20.14 % and 16.72 % of the area, respectively, fall under the “Very High Forest Fire” susceptibility class. Trend analysis reveals a general upward trend in forest fire occurrences (R2 = 0.59), with a notable increase after 2015, peaking in 2021. Notably, Angul district was identified as the most affected area, documenting the highest number of forest fire incidents over the past 22 years. Additionally, forest fire mitigation plans have been developed by drawing insights from forest fire management strategies implemented in various countries worldwide. Overall, this analysis provides valuable insights for policymakers and forest management authorities to develop effective strategies for forest fire prevention and mitigation.

Functional support vector machine.

Linear and generalized linear scalar-on-function modeling have been commonly used to understand the relationship between a scalar response variable (e.g. continuous, binary outcomes) and functional predictors. Such techniques are sensitive to model misspecification when the relationship between the response variable and the functional predictors is complex. On the other hand, support vector machines (SVMs) are among the most robust prediction models but do not take account of the high correlations between repeated measurements and cannot be used for irregular data. In this work, we propose a novel method to integrate functional principal component analysis with SVM techniques for classification and regression to account for the continuous nature of functional data and the nonlinear relationship between the scalar response variable and the functional predictors. We demonstrate the performance of our method through extensive simulation experiments and two real data applications: the classification of alcoholics using electroencephalography signals and the prediction of glucobrassicin concentration using near-infrared reflectance spectroscopy. Our methods especially have more advantages when the measurement errors in functional predictors are relatively large.

Wavelet Transform and SVM Based Heart Disease Monitoring for Flexible Wearable Devices

INTRODUCTION: Heart disease has been a major health challenge globally, therefore the development of reliable and real-time heart disease monitoring methods is crucial for the prevention and management of heart health. The aim of this study is to explore a flexible wearable device approach based on wavelet transform and support vector machine (SVM) to improve the accuracy and portability of heart disease monitoring. OBJECTIVES: The main objective of this study is to develop a wearable device that combines wavelet transform and SVM techniques to achieve accurate monitoring of physiological signals of heart diseases. METHODS: An integrated method for heart disease monitoring was constructed using flexible sensor technology combined with a wavelet transform and support vector machine. The Marr wavelet transform was applied to the ECG signals, and the feature vectors were constructed by feature parameter extraction. Then, the radial basis kernel SVM was utilized to identify the three ECG signals. The performance of the algorithm was optimized by adjusting the SVM parameters to improve the accurate monitoring of heart diseases. RESULTS: The experimental results show that the proposed wavelet transform and SVM-based approach for flexible wearable devices achieves satisfactory results in heart disease monitoring. In particular, the algorithm successfully extracted feature vectors and accurately classified different ECG signals by skillfully combining the wavelet transform and SVM techniques for the processing of premature beat signals. CONCLUSION: The potential application value of the wavelet transform and SVM-based flexible wearable device approach in heart disease monitoring is emphasized. By efficiently processing ECG signals, the method provides an innovative and comfortable solution for real-time monitoring of cardiac diseases.

Predicting psycho-somatic disorders in online activity using multi-layer perceptron

<p>Internet services such as social media, blogs, and websites make it possible for people to acquire knowledge instantly. Due to these websites, it is now considerably easier to communicate information. As a result, individuals increasingly devote a higher amount of time to social networking programmes. This study provides estimates about the potential future ramifications of how individuals will utilise social networks. This work presents an accurate and applicable model for forecasting undesirable consequences. The model is of sufficient quality to be useful. This has been the case throughout. Using the model that has been proposed, significant properties are identified from datasets. After recovering the properties, they are categorised using the complicated computational method of multi-layer perceptron-based (MLP) artificial neural networks (ANN). 70% of this data was utilised during the training phase of the machine learning algorithm, while the remaining 30% was utilised during the validation phase of model construction. The proposed model's results were compared to those of more standard machine learning techniques. The approach utilises social networks to predict the issue. The simulation results indicate that the suggested model generates more precise predictions than the support vector machine, logistic regression, and random forest decision tree classifier techniques combined.</p>

Comparing the performance of machine learning methods in estimating the shear wave transit time in one of the reservoirs in southwest of Iran.

Shear wave transit time is a crucial parameter in petroleum engineering and geomechanical modeling with significant implications for reservoir performance and rock behavior prediction. Without accurate shear wave velocity information, geomechanical models are unable to fully characterize reservoir rock behavior, impacting operations such as hydraulic fracturing, production planning, and well stimulation. While traditional direct measurement methods are accurate but resource-intensive, indirect methods utilizing seismic and petrophysical data, as well as artificial intelligence algorithms, offer viable alternatives for shear wave velocity estimation. Machine learning algorithms have been proposed to predict shear wave velocity. However, until now, a comprehensive comparison has not been made on the common methods of machine learning that had an acceptable performance in previous researches. This research focuses on the prediction of shear wave transit time using prevalent machine learning techniques, along with a comparative analysis of these methods. To predict this parameter, various input features have been employed: compressional wave transit time, density, porosity, depth, Caliper log, and Gamma-ray log. Among the employed methods, the random forest approach demonstrated the most favorable performance, yielding R-squared and RMSE values of 0.9495 and 9.4567, respectively. Furthermore, the artificial neural network, LSBoost, Bayesian, multivariate regression, and support vector machine techniques achieved R-squared values of 0.878, 0.8583, 0.8471, 0.847 and 0.7975, RMSE values of 22.4068, 27.8158, 28.0138, 28.0240 and 37.5822, respectively. Estimation analysis confirmed the statistical reliability of the Random Forest model. The formulated strategies offer a promising framework applicable to shear wave velocity estimation in carbonate reservoirs.

Identification of Bearing Clearance in Sugar Centrifuge Using Dimension Theory and Support Vector Machine on Vibration Measurement

Abstract Bearing fatigue life is significantly influenced by bearing clearance. Vibration monitoring of bearing clearance deviations can efficiently reveal bearing wear and give sufficient lead time for maintenance. This study investigates the dynamics of roller bearings utilizing the dimension theory with the support vector machine (SVM) technique for diagnosing the bearing clearance faults of revolving machines. The generation of local defects in rotating machines is closely related to the clearance behavior of the rotor-bearing system. A dynamic model of bearing with dimension theory by matrix method with SVM is developed for characteristics of bearing clearance considering the influence of local defects on the inner and outer bearings races. The characteristics of bearing internal radial clearance considering the impact of the defect on the bearing are analyzed. An experimental study has been performed to capture the vibration signature of radial clearance for different speeds and radial loads of the rotor. The rotor-bearing system equations are numerically integrated, and the results are validated with experimental findings. The collective effects among the four parameters (radial load, speed, defect size, radial clearance) are investigated in detail for the rotor-bearing system. The noisy signal is subsequently eliminated using the modulation signal bispectrum (MSB), and the peaks of the MSB results are represented by the bearing clearance indicator. The efficiency and reliability of the stated approach are evaluated using a specialized bearing test and a run-to-failure sugar centrifuge test. The results suggest that the proposed approach can detect a change in bearing clearance up to 40 µm.

Comparison of Machine Learning for Multispectral Signature Classification of Invasive Spathodea campanulata in Remote Sensing Imagery

Invasive plant species pose a significant threat to native ecosystems, and mapping their distribution is essential for effective management and conservation efforts. This study compared the performance of different machine learning techniques in the classification of multispectral signature of the invasive plant species Spathodea campanulata using remote sensing imagery. The ground-based locality data of vegetation was collected using a random sampling technique, and satellite imagery was acquired from Sentinel-2. The results showed that the blue, green, red, red- edge, and near-infrared bands were effective in distinguishing between invasive S. campanulata, non-vegetation, and other vegetation using machine learning techniques. The support vector machine (SVM) technique achieved the highest overall accuracy of 80%, followed by random forest (RF) with 73% and K-nearest neighbor (KNN) with 66%. The Gaussian mixture model (GMM) technique had the lowest overall accuracy of 53%. SVM and RF showed substantial agreement between predicted and observed classes, whereas KNN showed moderate agreement and GMM showed poor agreement. The maps generated by SVM depict a distribution of the invasive plant species that is characterized by isolated patches in the northern region of the study area. In contrast, the southern region – including the protected area of Mount Timolan – shows a dense presence of S. campanulata, indicating an ongoing invasion by the species. This highlights the need for effective management and conservation efforts to mitigate the negative impact of invasive plant species on native ecosystems.

Fractal based feature extraction technique for classifying EEG signal for color visualization tasks

Exploring and finding Significant features for colour visualization tasks using the EEG signals is crucial in developing a robust Brain-machine Interface (BMI). The visually evoked potential carries multiple pieces of information, and finding its best feature is a tedious task. The main objective of this research is to concentrate on various linear and non-linear features which classifies the visually evoked potential when visualizing various colours for a certain period with reduced computational time and with higher accuracy. The feature extraction techniques utilized for extracting the features of EEG signals while visualizing various colours are Power Spectral Intensity (PSI), Spectral Entropy (SE), Detrended Fluctuation analysis (DFA), Higuchi Fractal Dimension (HFD), Petrossian Fractal Dimension (PFD), Multifractal Detrended Fluctuation Analysis (MFDFA). The extracted features were classified using the Multiclass classifier using one vs rest technique Support Vector Machine algorithm. The result shows that the MFDFA method with multiclass classifier combination has achieved 97.4 percent of classification accuracy when compared with other features.

Analysis of Fault Detection in Analog Circuits Using WSF-SKC Optimized SVM Technique

Many industrial applications and control systems depend heavily on analogue electrical circuitry. The conventional method of diagnosing such circuit faults can be time-consuming and erroneous, which might have a severe impact on the industrial output. The fault detection and analysing of analogue circuits with intelligent effective model is proposed in this work. The suggested technique primarily consists of two main stages one is extraction of features and the other is classification of faults. The analysis is performed on the response of frequency in analogue circuits. For extracting features particle swarm optimization (PSO) is utilized. The PSO is used to evaluate the fitness function of Wilks A-Statistic Filters sallen-key circuit (WSF-SKC). With fault characteristics retrieved using the particle swarm approach that are carefully selected, the fault classes may be separated more quickly. To categorise different failures in a benchmark circuit, a Support Vector Machine (SVM) classifier is built. Utilising firefly optimisation, the classifier is improved. Different fault codes were tested in experiments for defect detection and identification. The findings of the experiment indicate that this proposed technique can significantly increase the accuracy of fault diagnosis. The accuracy obtained for WS-LPF is 99.95%, WS-HPF is 99.97 and WS-BPF is 99.90% respectively.

Accuracy Assessment of Land Use/Land Cover Indices for Al-Rusafa in Baghdad Governorate by Remote Sensing Technology and GIS

The management and planning of natural and artificial resources depend on accurately monitoring land cover changes. Land cover change mapping and monitoring used to require expensive field surveys. Remote sensing is cheaper and more practical for mapping land use and cover changes. The Tigris River divides the Iraqi capital, Baghdad, into two parts: Karkh and Rusafa. Al-Rusafa was selected as a study area for current research, which has had rapid population and urban growth in recent decades. The current research applies the support vector machine technique to supervised LU/LC maps’ classification into barren regions, water bodies, vegetation cover and built-up regions. Spectral indicators were calculated: Enhanced Vegetation Index, Modified Normalized Difference Water Index, Normalized Built-Up Area Index, Dry Bareness Index in addition to calculating the accuracy assessment and Kappa coefficient. Using the Landsat 9 satellite image, ArcGIS 10.8 and Envi5.3 software were used to analyze and evaluate the results and field points observed by GPS devices. The results showed that the SVM classification algorithm accurately revealed the categories of LU/LC, where the classification accuracy reached 95%, and that the arid lands covered most of the study area 848.864 km2 and water bodies 76.747 km2, the vegetation and the built-up regions 466.459 km2 and 439.077 km2, respectively. The spectral indices showed slightly different areas of barren lands (DBSI 752.589 km2, 93% accuracy), vegetation (EVI 423.651 km2, 96% accuracy), and water bodies (MNDWI 73.187 km2, 98% accuracy) and built-up areas (NBAI 501,731 km2, 90%accuracy). The Support Vector Machine method outperforms other classification methods, and the spectral indicators employed in this work are useful and dependable for extracting each LU/LC category. In conclusion, Landsat 9 satellite data can reliably and swiftly detect ground cover.

Assessment of Zero-Day Vulnerability using Machine Learning Approach

Organisations and people are seriously threatened by zero-day vulnerabilities because they may be utilised by attackers to infiltrate systems and steal private data. Currently, Machine Learning (ML) techniques are crucial for finding zero-day vulnerabilities since they can analyse huge datasets and find patterns that can point to a vulnerability. This research’s goal is to provide a reliable technique for detecting intruders and zero-day vulnerabilities in software systems. The suggested method employs a Deep Learning (DL) model and an auto-encoder model to find unusual data patterns. Additionally, a model for outlier detection that contrasts the autoencoder model with the single class-based Support Vector Machine (SVM) technique will be developed. The dataset of known vulnerabilities and intrusion attempts will be used to train and assess the models.

Power demand estimation techniques applied to microgrid

ABSTRACT The increasing global demand for power necessitates a focus on renewable energy sources to reduce environmental impact and diversify energy generation. Estimating power demand in real-time electrical systems is crucial in this context. However, the literature contains various estimation techniques without clear guidance on the most suitable option based on performance and field-specific use. To address this gap, a study compares these techniques and identifies the most effective method for power demand estimation through a training process that assesses the accuracy of predictions against actual values. This comparison is conducted using the JUPYTER platform and historical data from the ENTSO-E power systems database. Results indicate that Linear Regression and Support Vector Machine techniques are commonly used for model evaluation, but the Adaptive Neural Network consistently performs as the most robust estimation method. The choice of technique depends on researcher preference, but it's essential to note that inadequate data preparation can lead to suboptimal results. In summary, as the world's power demand rises, the study identifies the Adaptive Neural Network as a promising tool for accurate power demand estimation, emphasizing the significance of proper data preparation in this process.

Nontechnical Loss Detection for Metered Customers in Alexandria Electricity Distribution Company Using Support Vector Machine

Non-technical losses (NTL) during transmission and distribution (T&D) of electrical energy is a major problem faced by utility companies which is very difficult to fight and detect. For that, more of power utilities spend thousands dollar for research centres to find efficient methods for detecting and controlling abnormalities. Electricity theft and billing irregularities forms the main portion of NTL. With the introduction of smart meter, the frequency of reporting energy consumption data to the utility company has been increased. Incoming and outgoing energy could be monitored and analyzed. Electricity theft is a complex problem with many parameters to be evaluated before implementing any measures to detect and control that. These parameters include some issues like social, economic, regional, managerial, infrastructural, and corruption. In recent years, several data mining and research studies on fraud detection and prediction techniques have been carried out in the electricity distribution sector. Support vector machine (SVM) technique has dominated the research for classifying data and detecting fraudulent electricity customers. SVM technique has good ability in data mining and data classification. The paper objective is to analyze the metered energy consumption data recorded and predict the pattern or the form of daily user’s energy consumption, then using SVM to classify the data whether normal or theft. The suggested technique is then tested using real data from Alexandria Electricity Distribution Company (AEDC). The proposed technique was able to distinguish between healthy and theft cases.

Machine Learning based Sorting of Somatic Embryos for In-Line processing in Automated SE Fluidics System

Somatic embryogenesis (SE) can be a viable method for the clonal propagation of many economically significant forest trees, particularly coniferous trees like pines and spruces. However, large-scale production of SE plants requires automation to reduce manual labor and attain cost-efficiency. The most labor-intensive step of the SE process for SE plant production is selecting and harvesting mature embryos. Embryo maturation is not a synchronized process; selecting the most developed embryos capable of continuous development is necessary. However, there needs to be more research conducted on mapping morphological features to germination-competent mature somatic embryos. This paper lays down the preliminary work of employing machine learning techniques for classifying large volumes of images of mature somatic embryos processed using an automated SE processing system based on fluidics processing referred to as SE Fluidics system. The results show that machine learning could be an alternative classification methodology instead of the traditional manual morphology-based classification process based on image analysis. The paper discusses two popular image classification techniques, namely Convolution Neural Network (CNN) and Support Vector Machine (SVM), applying them to both binary (black and white) and grayscale images. It is observed that grayscale images provide better accuracy with the SVM technique and outperform morphology-based classification in terms of processing speed (17.6% faster) across the test envelope. On the other hand, CNN-based classification shows better processing speeds only at a lower number of convolution layers. Hence, the data scientist can optimally select the number of convolution layers to get the desired accuracy-processing speed combination.

Sentiment Analysis Regarding Candidate Presidential 2024 Using Support Vector Machine Backpropagation Based

This research has the potential to make an important contribution to the development of computationally-based sentiment analysis, particularly in the political context. Anies Baswedan, Ganjar Pranowo, and Prabowo Subianto, three candidates for the presidency of Indonesia, are examined using a Backpropagation-based Support Vector Machine (SVM) methodology in this study. This approach is used to categorize emotions into three groups: neutral, adverse, and favorable. Between July 1 and July 30, 2023, data on tweets mentioning the three presidential contenders was gathered. After processing the data, SVM was used while lowering the backpropagation process. The study's findings demonstrate that the performance of the model in determining public sentiment is greatly enhanced by the application of backpropagation-based SVM techniques. For each presidential contender, the evaluation was conducted using the f1 score, recall, and precision metrics. The evaluation's findings indicate that while the model struggles to distinguish between favorable and negative feelings toward particular presidential contenders, it performs better when categorizing neutral feelings. The SVM model is more accurately able to identify popular sentiment toward the three presidential candidates when the backpropagation approach is used. The results of the sentiment analysis are also represented by word clouds for each presidential contender, giving an intuitive sense of the words that are frequently used in public discourse. This study sheds light on the possibilities of using Twitter data to analyze political sentiment using the backpropagation-based SVM algorithm.

Ensemble Heuristic–Metaheuristic Feature Fusion Learning for Heart Disease Diagnosis Using Tabular Data

Heart disease is a global health concern of paramount importance, causing a significant number of fatalities and disabilities. Precise and timely diagnosis of heart disease is pivotal in preventing adverse outcomes and improving patient well-being, thereby creating a growing demand for intelligent approaches to predict heart disease effectively. This paper introduces an ensemble heuristic–metaheuristic feature fusion learning (EHMFFL) algorithm for heart disease diagnosis using tabular data. Within the EHMFFL algorithm, a diverse ensemble learning model is crafted, featuring different feature subsets for each heterogeneous base learner, including support vector machine, K-nearest neighbors, logistic regression, random forest, naive bayes, decision tree, and XGBoost techniques. The primary objective is to identify the most pertinent features for each base learner, leveraging a combined heuristic–metaheuristic approach that integrates the heuristic knowledge of the Pearson correlation coefficient with the metaheuristic-driven grey wolf optimizer. The second objective is to aggregate the decision outcomes of the various base learners through ensemble learning. The performance of the EHMFFL algorithm is rigorously assessed using the Cleveland and Statlog datasets, yielding remarkable results with an accuracy of 91.8% and 88.9%, respectively, surpassing state-of-the-art techniques in heart disease diagnosis. These findings underscore the potential of the EHMFFL algorithm in enhancing diagnostic accuracy for heart disease and providing valuable support to clinicians in making more informed decisions regarding patient care.

CAD System Design for Two-class Brain Tumor Classification using Transfer Learning

Background: The occurrence of brain tumors is rapidly increasing, mostly in the younger generation. Tumors can directly destroy all healthy brain cells and spread rapidly to other parts. However, tumor detection and removal still pose a challenge in the field of biomedicine. Early detection and treatment of brain tumors are vital as otherwise can prove to be fatal. Objective: This paper presents the Computer Aided Diagnostic (CAD) system design for two classification of brain tumors employing the transfer learning technique. The model is validated using machine learning techniques and other datasets. Methods: Different pre-processing and segmentation techniques were applied to the online dataset. A two-class classification CAD system was designed using pre-trained models namely VGG16, VGG19, Resnet 50, and Inception V3. Later GLDS, GLCM, and hybrid features were extracted which were classified using Support Vector Machine (SVM), k-Nearest Neighbor (kNN), and Probabilistic Neural Network (PNN) techniques Results: The overall classification accuracy using Inception V3 is observed as 83%. 85% accuracy was obtained using hybrid GLCM and GLDS features using the SVM algorithm. The model has been validated on the BraTs dataset which results in 84.5% and 82% accuracy using GLCM + GLDS + SVM and Inception V3 technique respectively. Conclusion: 2.9% accuracy improvement was attained while considering GLCM + GLDS + SVM over kNN and PNN. 0.5% and 1.2% accuracy improvement were attained for CAD system design based on GLCM + GLDS + SVM and Inception v3 model respectively.

Prediction of Student Performance Based on Behavior using E-Learning During the Covid-19 Pandemic using Support Vector Machine

The COVID-19 crisis has profoundly impacted many sectors globally, including education, necessitating the shift from traditional in-person learning to independent or online learning through various digital platforms. The integrity of e-learning can be ensured by leveraging e-learning behavioral data. The objective of this research is to develop a novel data model to navigate the educational challenges of the COVID-19 era. Previous studies employed the Support Vector Machine (SVM) technique to predict student performance in an e-learning setting, yet they failed to contrast different SVM kernels and their outcomes. In contrast, this study uses SVM and compares three types of kernels: Radial, Polynomial, and Linear. The dataset used for this research was procured from X-API-Edu-Data. The SVM technique was utilized in a unique way to process the data, which comprised 17 variables and 40 observations. Notably, all 17 variables were character variables, with only four being numeric. Two variables, Raisedhands and Discussion, were selected for analysis due to their key role in effective learning and their association with student performance in an e-learning environment. The evaluation of the model was performed using the Topic variable, which represents the subjects in the dataset. The research findings revealed a marked improvement in accuracy compared to earlier studies. Among the three SVM kernels tested - Radial, Polynomial, and Linear, the Polynomial kernel demonstrated superior accuracy with a score of 0.9979. Therefore, the Polynomial model was deemed most appropriate for analyzing the Topic variable. In conclusion, the study indicates that the application of the e-learning method, specifically during the COVID-19 pandemic, proved highly effective in forecasting student performance.

Combining a forward supervised filter learning with a sparse NMF for breast cancer histopathological image classification

Histopathological images play a important role in clinical diagnosis, particularly in identifying and assessing the severity of abnormal conditions like benign lesions and malignant tumors. Traditional machine learning techniques for processing histopathology images involve the extraction of manual features from these images, which is typically done with the assistance of industry experts. Recent advancements in Deep Learning (DL), especially with Convolutional Neural Networks (CNN), have enabled the automatic extraction of multi-level abstract features directly from raw data. This capability significantly enhances the performance of complex computer vision tasks. Classic CNN models like AlexNet and VggNet employ back-propagation algorithms to learn filters in the training phase. However, these algorithms demand large labeled datasets, resulting in extensive computational processing. Additionally, they often face the vanishing gradient problem, which can negatively impact the quality of the learning process. Besides, in many domains, acquiring enough labeled images for conducting properly the training phase is a real challenge. To address these challenges, a feed-forward propagation approach was proposed using Non-Negative Matrix Factorization(NMF). The NMF technique factorizes the input data into two latent factors (non-negative matrices). It has been shown that by enforcing constraints such as sparsity on the latent factors, dominant features that are mostly correlated with tumors types can be extracted. In this work, a novel model combining sparse NMF and Support Vector Machine (SVM) was developed for classifying histopathological images. We have derived a mathematical model of a novel feed-forward filter learning approach that combines sparse NMF (SNMF) and Support Vector Machine technique (SVM). The model was used to design and implement a feed-forward CNN classifier to classify histopathology images. This model has been evaluated on the histopathology images from Sultan Qaboos University Hospital (SQUH dataset) and the public BreaKHis dataset. The experiments we have conducted demonstrate the efficiency of the proposed model, especially on small-sized SQUH datasets achieving an AUC of 0.90, 0.89, 0.85, and 0.86 on 4x,10x, 20x, and 40x magnifications, respectively, and achieving an AUC of 0.95 BreaKHis dataset.