3- إعلام آلي
Permanent URI for this collectionhttp://172.16.2.43:4000/handle/123456789/580
Browse
Recent Submissions
Item AI-Driven Cardiac Diagnostics: Leveraging Vision Transformers for PCG Signal Classification(University of Echahid Cheikh Larbi Tébessi -Tébessa, 2025-06-06) BOUATOUATA, Chakibwork presents a comprehensive approach to classifying cardiac sound (PCG) signals into normal and abnormal categories using hybrid deep learning and ma- chine learning techniques. Our goal is to develop a solution based on artificial intelligence techniques. The study leverages the PhysioNet/CinC Challenge 2016 dataset, which consists of recordings from multiple environments and patient cases, introducing significant variability in signal quality, duration, and class balance. A robust preprocessing pipeline was implemented to normalize and segment the sig- nals by patient. Time-domain and frequency-domain features were extracted to improve representation. n our work, we proposed a hybrid architecture that inte- grates multiple models, long short-term memory (LSTM) networks to model tem- poral context, and an attention mechanism to weight the most informative heart- beats. The most relevant segments were selected using attention weights, forming the final input to a random forest classifier. Experimental results demonstrate that the proposed system achieves promising classification performance, proving its effectiveness in handling PCG variability and imbalanced real-world data. This work highlights the potential of hybrid deep learning methods in automated car- diac listening systemsItem Classification Audio pour les Maladies Respiratoires(Université de Larbi Tébessi –Tébessa, 2025-06-09) DJEDIDI, AbdelghaniLes maladies respiratoires constituent un défi sanitaire mondial majeur, nécessitant des outils diagnostiques rapides, fiables et accessibles. L’analyse audio des sons de toux offre une alternative non invasive prometteuse. Ce projet vise à développer un système basé sur l’intelligence artificielle pour la classification automatisée des maladies respira- toires à partir de l’audio de la toux, en utilisant le jeu de données COUGHVID annoté par des experts. Face aux défis tels que la variabilité acoustique et le bruit, nous avons exploré diverses techniques de prétraitement et d’extraction de caractéristiques, en nous concentrant sur la combinaison des coefficients MFCC et des spectrogrammes Mel. Plusieurs modèles de classification ont été entraînés et évalués sur ces caractéristiques. Notre approche méthodologique a abouti à la sélection d’un modèle XGBoost opti- misé comme étant le plus performant. Les expériences menées ont rigoureusement validé son efficacité : le modèle XGBoost final a atteint une précision remarquable de 98.01% sur l’ensemble de test (et 97.51% en validation) pour la classification des conditions res- piratoires basées sur les diagnostics experts. Ces résultats surpassent significativement ceux d’autres modèles testés, tels que Random Forest (96.20% test) et les approches d’Apprentissage Profond standard (94.12% test). Ce travail contribue une méthodologie optimisée combinant des caractéristiques acous- tiques pertinentes (MFCC + Mel-spectrogramme) avec un classificateur XGBoost per- formant, démontrant une haute précision sur des données validées cliniquement. Ces résultats soulignent le potentiel de l’analyse audio de la toux pour développer des outils de dépistage évolutifs et économiques, renforçant ainsi la télémédecine et les stratégies de santé publique.Item Video Tampering Detection Based on Deep Learning Techniques(University of Echahid Cheikh Larbi Tébessi -Tébessa, 2025-06-10) RAIS, TameurDashcams have become standard in modern vehicles, capturing continuous video footage that can serve as vital digital evidence in legal disputes or insurance claims. However, the reliability of these recordings is threatened by various forms of video tampering, including video forgery, frame deletion, frame insertion, and even deepfake-based alterations. These challenges demand robust and intelligent forgery detection systems capable of identifying manipulations at both frame and video levels. This master thesis presents a deep learning-based approach for automated video tampering detection in dashcam recordings. It begins with an in-depth theoretical analysis of digital video forgery, outlining its common forms and the limitations of traditional forensic methods. The study then focuses on the use of Convolutional Neu- ral Networks (CNNs), known for their effectiveness in visual anomaly detection, and proposes a novel detection pipeline built upon the ResNet50 architecture. The main contribution is the implementation of a custom model named BinaryCNN, a CNN-based architecture derived from ResNet50 and fine-tuned to classify dashcam videos into “original” or “forged” (specifically for frame deletion forgery). The de- tection pipeline includes video preprocessing, frame extraction, individual frame classification, and score aggregation to determine video-level authenticity. The system was trained and evaluated using the publicly available Video Forgery Dataset (Kaggle), which contains thousands of dashcam videos, divided into classes representing different tampering techniques: frame deletion, insertion, duplication, flipping, rotation, and zooming. For this work, we focused on a binary classification task: detecting frame deletion forgery vs. original. The dataset includes 1181 videos per class, split evenly between training and testing sets, and covers diverse driving conditions. The proposed BinaryCNN model achieved an overall accuracy of 79%, with a precision of 81%, recall of 78%, and F1-score of 79%, demonstrating its effectiveness in identifying forged videos under real-world conditions. These results confirm the capability of CNN-based models to tackle video forgery detection with high reliability. Additionally, a user-friendly web interface was developed using Flask to demonstrate the real-time application of this deep learning video tampering detection system. Users can upload a dashcam video and instantly receive a classification result. This research contributes to the growing field of automated forgery detection and highlights the potential of deep learning techniques, especially ResNet50-based CNNs, in building scalable, intelligent, and robust systems for video tampering de- tection.Item Mefria : Une application super multiservice(Université de Larbi Tébessi –Tébessa, 2025-06-07) TRIKI Tarek, KALLA AbdelmounamMefria est une application mobile multiservices innovante conçue pour simplifier le quoti- dien des utilisateurs en centralisant plusieurs services essentiels au sein d’une seule plateforme. L’application propose initialement la prise de rendez-vous pour les coiffeurs et les coiffeuse, et prévoit d’étendre ses fonctionnalités vers la livraison de repas, les courses alimentaires, le transport de personnes, ainsi que la livraison de colis. Mefria vise à répondre aux besoins croissants des citoyens en matière de praticité, de gain de temps et d’efficacité, tout en soutenant la digitalisation des petits commerçants, artisans et prestataires de services. L’application se distingue par son interface conviviale, son système de réservation intuitif, ainsi que son approche locale et adaptée au contexte algérien. Ce mémoire explore les aspects techniques, ergonomiques et économiques de l’application, tout en analysant son impact potentiel sur les habitudes de consommation et la transformation numérique des services de proximité.Item Détection des anomalies et maintenance predictive avec l’intelligence artificielle dans l’internet des objets(Université de Larbi Tébessi –Tébessa, 2025-06-09) NACEUREDDINE, KaoutherL’Internet des objets (IoT) représente l’une des évolutions technologiques les plus marquantes dans l’industrie moderne. Il a permis aux systèmes de surveiller leur propre fonctionnement grâce à des réseaux de capteurs intelligents. Ces capteurs collectent des données opérationnelles en temps réel, ce qui permet d’analyser le comportement des équipements dans des conditions réelles. Cette capacité est devenue essentielle pour développer des stratégies de maintenance prédictive, afin de réduire les pannes inattendues et améliorer la fiabilité des systèmes. Dans ce cadre, ce travail présente une étude appliquée sur la surveillance des moteurs d’avion, considérés comme des systèmes industriels sensibles et complexes. Une architecture basée sur l’IoT a été utilisée pour collecter les données du moteur, comprenant différentes mesures comme la pression, la température et la vitesse. Ces données sont utilisées comme entrées d’un modèle intelligent conçu pour analyser le comportement du moteur et détecter les anomalies pouvant indiquer un dysfonctionnement imminent. Le modèle proposé suit une structure en plusieurs étapes et utilise des techniques d’intelligence artificielle. Il comprend : une première phase de détection des anomalies avec un Autoencoder utilisant un mécanisme d’attention, suivie d’une phase de classification de l’état du moteur (sain ou non) à l’aide de l’algorithme Balanced Random Forest, puis une phase de prédiction de la durée de vie restante (RUL) avec un réseau LSTM doté d’attention. Le modèle a été évalué à l’aide de la base de données C-MAPSS, et les résultats ont montré une nette amélioration dans la détection précoce et l’aide à la prise de décision en maintenance prédictive. Cette étude met en valeur l’intérêt de combiner les technologies de l’IoT et de l’intelligence artificielle pour construire des systèmes industriels plus intelligents, capables de réagir aux situations critiques avant qu’elles ne deviennent de véritables pannes.Item Classification de catégorie d’âge par Deep Learning(Université de Larbi Tébessi –Tébessa, 2025-06-09) ADJAL, ChamsLa classification de l'âge et du genre à partir d'images faciales joue un rôle important dans divers domaines tels que la sécurité, l'interaction homme-machine et la personnalisation des services. Cette étude présente une approche basée sur l'apprentissage profond pour la classification de l'âge (en 8 catégories) et la prédiction du genre. En utilisant l'apprentissage par transfert, plusieurs modèles pré-entraînés (VGG16, ResNet101, DenseNet121) sont évalués sur l’ensemble de données UTKFace (23 708 images), selon deux protocoles. Afin d'améliorer les performances, un modèle hybride combinant VGG16 et DenseNet121 est proposé, tirant parti de la complémentarité des deux architectures. Ce modèle de fusion atteint une précision de 75,14 % pour la classification d’âge et 92,90 % pour la reconnaissance du genre, surpassant les modèles individuels.Item Blockchain Technology for Product Traceability: Application to Pharmaceutical Products.(University of Echahid Cheikh Larbi Tébessi -Tébessa, 2025-06-09) KHIMECHE, AimeneThis thesis discusses the design and implementation of a blockchain-based system for pharmaceutical product tracking and tracing. The objective here is to give transparency, integrity, and security in tracking the product across the supply chain. Current systems are centralized and hence prone to tampering and inefficiency. To address these issues, we recommend the implementation of a decentralized application using modern web technologies like React for frontend development, Express.js for the backend, PostgreSQL for database, and a blockchain network was implemented using special web libraries to enable peer-to-peer communication and consensus without the need for a central authority. The process allows users to create, authenticate, and transfer product records on a shared ledger such that each stage of the life cycle of a product is recorded irretrievably. Peers authenticate transactions and mine blocks published on the network collectively. The process enhances the integrity of pharma supply chains by eliminating counterfeiting, fraud, and unauthorized product tampering. The prototype created herein demonstrates the effectiveness of the combination of blockchain and web technologies towards secure and scalable traceability. This work contributes to the advancement of decentralized supply chain management solutions, most importantly in industries where trust and traceability are a necessity.Item Offline Writer Identification Using Textural Descriptors(University of Echahid Cheikh Larbi Tébessi -Tébessa, 2025-06-09) DJABRI, FaroukThis thesis presents a novel approach for offline writer identification by applying textural descriptors to extract the unique writing style of individuals from static samples of handwriting. Offline writer identification is a challenging biometric problem due to the variability in handwriting caused by emotional, physical, and environmental factors, as well as missing dynamic writing details such as stroke order or pressure, to address these challenges, this manuscript investigates Run-Length-based feature extraction techniques, with a focus on Run-Length statistics. Our study begins by investigating traditional Run-Length (RL) descriptors, which are used to describe the distribution and repetition of binary patterns or runs in handwriting images. We enhance their discriminative ability with a new descriptor Z-Run-Length (Z-RL) that traverses images along a zigzag course to detect more subtle structural differences in handwriting. Furthermore, we evaluate a hybrid approach that combines baseline Run-Length features with the proposed Z-RL descriptor. The resulting feature vectors are classified using machine learning algorithms such as Support Vector Machines (SVM) and K-Nearest Neighbors (K-NN), and are validated using robust evaluation protocols, including k-fold cross-validation and Leave- One-Out Cross-Validation (LOOCV). Experimental results on the BFL dataset demonstrate that the proposed system achieves high accuracy and exhibits strong consistency across diverse handwriting samples.Item Embedded System for Corrosion Analysis and Prediction(University of Echahid Cheikh Larbi Tébessi -Tébessa, 2025-06-09) BRAHMI, RaidCorrosion is a major challenge in the oil and gas industry, compromising the safety, reliability, and sustainability of transportation infrastructure. This project proposes the design of an embedded system dedicated to analyzing and predicting corrosion in metal pipelines. It is based on the exploitation of data from inspections carried out over several years, combined with a BiLSTM (Bidirectional Long Short-Term Memory) deep learning model to predict the maximum corrosion depth at a given horizon. The dataset undergoes various preparation steps: missing value processing, correlation analysis and feature matching between inspection years. The developed model demonstrated high predictive accuracy with a low mean absolute error. This approach enables effective predictive maintenance and contributes to failure prevention, thus providing an innovative and suitable solution for real-time corrosion monitoring in embedded industrial systems.Item A hybrid approach for offline handwritten and printed script identification(University of Echahid Cheikh Larbi Tébessi -Tébessa, 2025-06-10) DJELAL, RayaneWith the growing need to understand and analyze multilingual documents, recognizing different of scripts has become a major challenge in optical character recognition (OCR) and automated document analysis. This work presents a hybrid approach that combines the power of deep learning for extracting visual features with the flexibility of traditional machine learning algorithms for classification. The proposed system is based on a deep neural network called YafNet, designed to extract specific features for each script from images of printed and handwritten words. These features are then used by several classic classifiers, especially logistic regression, to identify the script type. The methodology includes several steps: image preprocessing, deep feature extraction with YafNet, and classification using eight different models (such as SVM, RF, KNN, XGBoost, etc.). Experiments were conducted on the MDIW-13 dataset, which includes 13 different writing systems, covering both printed and handwritten texts. The YafNet–LogReg model performed well in all scenarios, successfully distinguishing between various script types. Error analysis showed that most confusions occurred between visually or linguistically similar scripts, highlighting the difficulty of the task and the strength of the proposed method. This study shows the benefit of combining deep representations from CNNs with simple and efficient classifiers. The proposed approach can be applied in real-world use cases such as automatic language or script identification in document archiving systems, or as a first step in multilingual OCR systems.Item Animal Skin Disease Identification Using Machine Learning Methods(University of Echahid Cheikh Larbi Tébessi -Tébessa, 2025-06-09) SERRADJ, ZahiaSkin diseases in animals are a growing concern in veterinary medicine, particularly in rural and underserved areas that lack access to specialized care services. With the rapid advancements in computer vision and artificial intelligence techniques, novel methods have emerged to facilitate the early and accurate diagnosis of these pathological conditions. This research proposes a predictive approach based on deep learning techniques, utilizing Convolutional Neural Networks (CNNs) and multispectral imaging to detect and classify skin diseases in dogs. The work is divided into three main axes: the first axis covers the fundamentals of animal skin diseases, their clinical importance, and traditional diagnostic methods. The second axis discusses the role of Transfer Learning and the use of state-of-the- art neural network architectures in medical image classification. The third axis details the experimental protocol, which includes the use of multispectral images, the class_weight technique to address data imbalance, and the evaluation of several CNN models such as Xception, EfficientNet, and DenseNet121. The results demonstrated that the Xception model outperformed the other models, achieving an accuracy of 98.9%, an Area Under the Curve (AUC) of 99.7%, and a Matthews Correlation Coefficient (MCC) of 98.4%. These findings highlight the efficacy of combining multispectral imaging with deep learning techniques in veterinary diagnostics and underscore the significant potential of AI-powere d tools in enhancing animal health, especially in resource-limited regions.Item Machine learning-Based Approach to Annotate Social Media Comments for Product Recommendation(University of Echahid Cheikh Larbi Tébessi -Tébessa, 2025-06-09) ZOUAOUI, WafaSentiment analysis is a core task in Natural Language Processing (NLP), widely applied in sectors such as digital marketing, politics, and customer feedback systems. However, its application to Arabic dialects—particularly the Algerian dialect—remains challeng- ing due to linguistic diversity, lack of standardized orthography, and the prevalence of non-standard scripts like Arabizi. Existing works are often limited to Modern Standard Arabic or fail to effectively capture dialect-specific nuances. To address these limitations, this work proposes a late fusion architecture that com- bines contextual embeddings from two complementary pre-trained models: DziriBERT, specialized for Algerian dialect, and MARBERT, trained on a broad set of Arabic di- alects. The fused representations are then passed to a BiLSTM classifier, selected for its strong capacity to capture sequential context. Experiments conducted on two real-world datasets—Djezzy and BrandDZ—demonstrated the effectiveness of this approach. The BiLSTM-based late fusion model achieved ac- curacies of 0.83% , respectively, These results confirm that integrating dialectal and broader Arabic representations, along with appropriate sequence modeling, leads to more robust and reliable sentiment classification for under-resourced dialects such as Algerian Arabic.Item Flood Detection in Satellite Imeges Using Deep Learning(University of Echahid Cheikh Larbi Tébessi -Tébessa, 2025-06-10) RABAH, ManarFloods are one of the most common and dangerous natural disasters. They can damage homes, roads, and important services, and they often happen without warning. To reduce the damage, it is important to detect floods quickly and accurately. This study uses deep learning methods to detect floods in satellite images. A total of 380 images were used, divided equally into flood and non-flood classes. Several deep learning models were tested, including VGG, ResNet, DenseNet, MobileNet, and Inception. These models were improved using transfer learning, which helps save time and improves results. The best models, such as ResNet50V2 and DenseNet, achieved high accuracy (up to 97%) in classifying the images. The study also tested a new idea by adding a quantum layer to the models using the PennyLane library. These hybrid models showed promising results. This research shows that deep learning and transfer learning are powerful tools for flood detection. In the future, this work can be improved by using more data and trying new ideas to make it even more accurate and useful.Item Holonic optimization system for sustainable agriculture planning in Algeria(University of Echahid Cheikh Larbi Tébessi -Tébessa, 2025-06-09) BRAHMI, AichaAgricultural planning in Algeria remains largely centralized, facing limitations in adaptability, responsiveness, and sustainability. With rising environmental challenges and the need for local-level customization, traditional systems struggle to efficiently manage complex agricultural data and decision-making processes. This thesis proposes a novel holonic optimization system for sustainable agricultural planning, designed to bridge the gap between local needs and national strategies. Inspired by holonic and multi-agent system principles, the solution introduces a hierarchical model composed of four interdependent levels: local, district, regional, and national. Each level functions as an autonomous agent (holon) capable of making partial decisions that contribute to a global agricultural plan. To operationalize this model, real data was collected from the Wilaya of Tébessa, structured into regional catalogs and parcel-level datasets. These were then used to simulate and evaluate different optimization techniques, namely the Genetic Algorithm (GA), Weighted Sum Method (WSM), and the BKPACS algorithm, which is the proposed approach in this work. Comparative results demonstrated that the holonic system using BKPACS outperforms traditional methods in terms of execution time, sustainability, and solution quality. The integration of AUML diagrams and a structured database further enhanced system clarity and applicability. Ultimately, this research offers a scalable, intelligent, and practical solution for transforming agricultural planning in Algeria, paving the way for digital, data-driven decision support in the agricultural sector.Item Développement d'un modèle prédictif pour la gestion des épidémies(Université de Larbi Tébessi –Tébessa, 2025-06-10) GOUTEL, AbirLes épidémies représentent un défi majeur pour les systèmes de santé publique à travers le monde, et la pandémie de COVID-19 a révélé les limites des méthodes traditionnelles de gestion et de prévision. Ce travail propose un modèle prédictif hybride combinant le modèle épidémiologique classique SEIR avec les réseaux de neurones LSTM pour améliorer la précision des prévisions quotidiennes de l’évolution des cas. Les modèles épidémiologiques traditionnels et les contributions de l’intelligence artificielle ont été étudiés, puis un prétraitement des données a été effectué à l’aide de la base de données de Johns Hopkins. Le modèle SEIR + LSTM a été développé et évalué à l’aide de métriques telles que MAE, RMSE, R2 et MAPE, démontrant des performances supérieures aux approches traditionnelles. Les résultats soulignent le potentiel de l’intelligence artificielle pour améliorer les prédictions épidémiologiques, ouvrant des perspectives prometteuses pour une gestion proactive des pandémies.Item Detection of plant diseases using artificial intelligence techniques(University of Echahid Cheikh Larbi Tébessi -Tébessa, 2025-06-10) SALMI, SouriaArtificial intelligence (AI), and in particular deep learning, is transforming the agricultural sector by enabling automated, fast, and reliable detection of plant diseases. This study explores an image classification approach for leaf images using Convolutional Neural Networks (CNNs), with the aim of automatically diagnosing several plant pathologies. The study is initially based on the standardized PlantVillage dataset, which is used to train and compare different architectures, including both pre-trained models (such as MobileNetV2) and a custom CNN model. The best model achieved an accuracy of 98% while maintaining low computational complexity. To make this solution accessible to users, a web interface was also developed, allowing automated diagnosis from user-uploaded images. This work thus contributes to the design of AI-assisted diagnostic systems for more effective plant disease management. Additionally, an initial series of tests was conducted on images from real-world conditions, marking an exploratory step toward validating the model in uncontrolled environments.Item Détection de l'engagement académique des étudiants par suivi oculaire(Université de Larbi Tébessi –Tébessa, 2025-06-09) FARES, WissalL’engagement académique est un facteur clé dans la réussite scolaire des étudiants. Sa mesure, souvent subjective, peut être optimisée grâce aux technologies émergentes comme le suivi oculaire et l’intelligence artificielle. Ce travail propose un modèle intelligent combinant un réseau de neurones convolutif pré-entraîné (VGG16) et un réseau LSTM afin de détecter automatiquement le niveau d’engagement académique à partir de séquences d’images d’yeux. Après avoir étudié les approches existantes, un prétraitement des données a été réalisé sur le dataset Eye Dataset de Kayvan Shah. Le modèle proposé a été évalué en utilisant des métriques telles que la précision, le rappel et le F1-score, montrant des performances supérieures aux méthodes traditionnelles. Cette approche ouvre des perspectives prometteuses pour l'amélioration des dispositifs éducatifs intelligents.Item Automated Glaucoma Diagnosis from Fundus ImagesUsing Deep Learning(University of Echahid Cheikh Larbi Tébessi -Tébessa, 2025-06-10) AHMED SISTA, Mohamed AliGlaucoma is a leading cause of irreversible blindness worldwide, often progressing without early symptoms and resulting in significant vision loss if left undiagnosed. Early detection is therefore critical, but current manual screening methods are time-consuming and require specialized expertise, limiting accessibility in many regions. This thesis presents a fully automated framework for glaucoma diagnosis using deep learning on retinal fundus images. The proposed method integrates three main phases: image enhancement with Contrast-Limited Adaptive Histogram Equalization (CLAHE), hierarchical segmentation of the optic disc and cup using a fine-tuned Segment Anything Model 2 (SAM2), and glaucoma classification based on neuroretinal rim analysis with a Vision Transformer (ViT). Experimental results on the ORIGA and REFUGE datasets demonstrate high segmentation accuracy, with Dice scores of 95.84% and 90.63% for the optic disc, and over 90% for the optic cup. For classification, the ViT model achieves an F1-score of 95.02% and a precision of 97.39% on the REFUGE dataset, outperforming several traditional convolutional neural networks. The framework is efficient, interpretable, and suitable for real-time clinical deployment. While specificity remains a challenge due to class imbalance and lack of data augmentation, the results highlight the potential of combining advanced segmentation and transformer-based classification for robust, scalable glaucoma screening. This work lays a strong foundation for future improvements and broader clinical application in automated eye disease diagnossi.Item Automatisation de tri de courrier(Université Larbi Tébessi-Tébessa, 2024-06-10) BRADJI, RayenDans le domaine du transport, de nombreuses entreprises proposent leur propre application de suivi de colis. Cependant, une application mobile "All-in-One" capable de suivre les colis de différentes entreprises serait plus avantageuse. Notre projet s'inscrit dans ce domaine en développant une application mobile permettant de suivre les envois de plusieurs entreprises de transport. L'objectif est de créer une plateforme centralisée offrant une expérience de suivi simplifiée et efficace pour les utilisateurs, tout en configurant notre propre API pour intégrer les données de suivi de différentes entreprises de transport.Item Texture features using machine learning techniques for offline facial expression recognition(University Larbi Tébessi – Tébessa, 2023) AYACHI, Ahmed Chiheb EddineThe graduation thesis aims to study and develop a facial expression recognition system using machine learning. Machine learning techniques are utilized to develop models and algorithms for data analysis, pattern recognition, and automatic prediction, without the need for direct programming. In the context of facial recognition and expression analysis, machine learning can be used to develop models capable of classifying different facial expressions. These models are trained on a large dataset containing pre-labeled facial expressions. After training on this data, the model becomes capable of recognizing facial patterns and classifying new expressions presented to it. Several experimental results have been shared in the thesis, using various classification algorithms such as SVM and KNN on the FER2013 dataset. The results have demonstrated satisfactory performance using these algorithms. To improve these results, a set of filters has been applied to the dataset, and the available data volume has been increased. Multiple studies have been conducted on these processes, and better results have been achieved after implementing the enhancements. In general, it can be concluded that the use of SVM and KNN algorithms in facial expression recognition has shown satisfactory results. Additional processes such as filter application and data augmentation have improved performance. This approach holds promise in developing an effective facial expression recognition system using machine learning. These results and proposed enhancements have been comprehensively discussed in the thesis. The research in the thesis aims to expand knowledge and gain a better understanding of machine learning applications in facial expression recognition. It seeks to achieve further advancements and improvements in performance and accuracy. This approach is promising and may contribute to the development of more efficient facial expression recognition systems applicable in various fields.