فرح عباس عبيد ساري

This paper introduces a method for fake accounts detection on social media through machine learning using several classification algorithms and sophisticated feature extraction. Preprocessing steps included dataset integration, feature normalization, entropy-based feature extraction, behavioral ratio computation, and class imbalance handling using SMOTE and SMOTEENN. Ensemble learning models, including LightGBM, CatBoost, XGBoost, and Logistic Regression, were trained on the training dataset and subsequently evaluated on the unseen testing dataset to assess their classification performance. Models were evaluated using accuracy, precision, recall, F1 score, and confusion matrices. SHAP explanation displayed the feature contributions. Experimental results demonstrated the effectiveness of the proposed approach, achieving the highest predictive accuracy of 99.71% and a balanced F1-score of 99.70%. These findings demonstrate excellent discriminative performance in distinguishing real accounts from the fake accounts and generated the following implications. Mixtures of feature selection, resampling strategies, and explainable AI can clearly enhance fake account detection. The proposed framework provides a scalable and effective solution for identifying fraudulent social media accounts and enhancing trust and security across online platforms.

This paper introduces a method for fake accounts detection on social media through machine learning using several classification algorithms and sophisticated feature extraction. Preprocessing steps included dataset integration, feature normalization, entropy-based feature extraction, behavioral ratio computation, and class imbalance handling using SMOTE and SMOTEENN. Ensemble learning models, including LightGBM, CatBoost, XGBoost, and Logistic Regression, were trained on the training dataset and subsequently evaluated on the unseen testing dataset to assess their classification performance. Models were evaluated using accuracy, precision, recall, F1 score, and confusion matrices. SHAP explanation displayed the feature contributions. Experimental results demonstrated the effectiveness of the proposed approach, achieving the highest predictive accuracy of 99.71% and a balanced F1-score of 99.70%. These findings demonstrate excellent discriminative performance in distinguishing real accounts from the fake accounts and generated the following implications. Mixtures of feature selection, resampling strategies, and explainable AI can clearly enhance fake account detection. The proposed framework provides a scalable and effective solution for identifying fraudulent social media accounts and enhancing trust and security across online platforms.

The use of digital technologies in agriculture has become very important to ensure the protection of trees from disease and limit their development, which leads to increased production, so the paper proposes a modified analytical model to analyze the data and graphical parts of the leaves of fruit trees using priority fuzzy C-means (PFCM). Based on the proposed distance scale to obtain a clustering with a less error rate and fairly close to accuracy for the purpose of monitoring the development of diseases of fruit trees, by classifying the diseases and medications needed for each disease, a database was created containing large samples of data and images, where the results of Analysis of previous studies that analyzes of large amounts of data give accurate results. The proposed method was used in smart gardens with large areas and we got the desired results.

The development of a mathematical model of the risks of cardiovascular events in humans based on risk factors for cardiovascular diseases is an unresolved problem of modern knowledge. We have identified and grouped by type the following risk factors for cardiovascular diseases: biological indicators (anthropometric, biochemical, morphological, physiological), indicators of diseases and social indicators. These risk factors are designated as corresponding linguistic variables. A logical-linguistic model has been built. Further, the risk degree of each of the indicators is assessed, where the risk degree is first calculated using the membership formula, then using a logical-linguistic model, the hazard class is assessed according to the degree of risk.

A patient who has fully recovered from COVID-19 can help patients currently fighting infection by donating plasma. Because it is an infection killer, the plasma now contains antibodies against COVID-19. These antibodies provided the immune system with one way to fight the virus when it was sick, so plasma can be used to fight diseases. Therefore, this paper monitoring recovering patients based on the clustering of data and classifying them using fuzzy hierarchical clustering to reach the plasma as soon as possible.

The paper presents the formulation and solution of the fuzzy problem of the distribution of nodes in the structure of an information system with a star-tree structure under uncertainty. Fuzzy parameters are specified as fuzzy numbers of the L-R type.