Search

Breast cancer causes the highest number of deaths weekly. It is the most common type of cancer and the leading cause of death in women in the world. Deep learning and machine learning are a way to predict perfect therapeutic protocols for patients. This paper presents different models to perform the prediction of breast cancer chemotherapy protocol and the number of cycles of cure using deep learning and machine learning algorithms. The database of therapeutic protocols related to 600 patients with breast cancer pathology was constructed at the University Hospital Center of Marrakesh, Morocco, it was treated following three distinct procedures before being used. In the first procedure, even though it is imbalanced, the database was left as it was. In the second procedure, the database was divided into three equal classes of 200 samples (Three balanced classes). As for the third, a database augmentation was realized, so that the minority classes have the same number of samples as the majority classes to have a balanced database. The main objective is to determine with our restricted database that we have collected since 2018, the best algorithm for the prediction of the therapeutic protocol and number of cycles, in accuracy terms, balanced accuracy, time, and precision. The results show that machine learning algorithms, especially Random Forest and XGBoost gave the best scores when the database was amplified. [ABSTRACT FROM AUTHOR]

Denial of Service (DoS/DDoS) intrusions are damaging cyberattacks, and their identification is of great interest to the Intrusion Detection System (IDS). Existing IDS are mainly based on Machine Learning (ML) methods including Deep Neural Networks (DNN), but which are rarely hybridized with other techniques. The intrusion data used are generally imbalanced and contain multiple features. Thus, the proposed approach aims to use a DNN-based method to detect DoS/DDoS attacks using CICIDS2017, CSE-CICIDS2018 and CICDDoS 2019 datasets, according to the following key points. a) Three imbalanced CICIDS2017-2018-2019 datasets, including Benign and DoS/DDoS attack classes, are used. b) A new technique based on K-means is developed to obtain semi-balanced datasets. c) As a feature selectionmethod, LDA (Linear Discriminant Analysis) performance measure is chosen. d) Four metaheuristic algorithms, counting Artificial Immune System (AIS), Firefly Algorithm (FA), Invasive Weeds Optimization (IWO) and Cuckoo Search (CS) are used, for the first time together, to increase the performance of the suggested DNN-based DoS attacks detection. The experimental results, based on semi-balanced training and test datasets, indicated that AIS, FA, IWO and CS-based DNNs can achieve promising results, even when cross-validated. AIS-DNN yields a tested accuracy of 99.97%, 99.98% and 99.99%, for the three considered datasets, respectively, outperforming performance established in several related works. [ABSTRACT FROM AUTHOR]

Due to the increasing number of cyber-attacks, the necessity to develop efficient intrusion detection systems (IDS) is more imperative than ever. In IDS research, the most effectively used methodology is based on supervised Neural Networks (NN) and unsupervised clustering, but there are few works dedicated to their hybridization with metaheuristic algorithms. As intrusion detection data usually contains several features, it is essential to select the best ones appropriately. Linear Discriminant Analysis (LDA) and t-statistic are considered as efficient conventional techniques to select the best features, but they have been little exploited in IDS design. Thus, the research proposed in this paper can be summarized as follows. a) The proposed approach aims to use hybridized unsupervised and hybridized supervised detection processes of all the attack categories in the CICIDS2017 Dataset. Nevertheless, owing to the large size of the CICIDS2017 Dataset, only 25% of the data was used. b) As a feature selection method, the LDAperformancemeasure is chosenand combinedwith the t-statistic. c)For intrusiondetection, unsupervised Fuzzy C-means (FCM) clustering and supervised Back-propagation NN are adopted. d) In addition and in order to enhance the suggested classifiers, FCM and NN are hybridized with the seven most known metaheuristic algorithms, including Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Differential Evolution (DE), Cultural Algorithm (CA), Harmony Search (HS), Ant-Lion Optimizer (ALO) and Black Hole (BH) Algorithm. Performance metrics extracted from confusion matrices, such as accuracy, precision, sensitivity and F1-score are exploited. The experimental result for the proposed intrusion detection, based on training and test CICIDS2017 datasets, indicated that PSO, GA and ALO-based NNs can achieve promising results. PSO-NN produces a tested accuracy, global sensitivity and F1-score of 99.97%, 99.95% and 99.96%, respectively, outperforming performance concluded in several related works. Furthermore, the best-proposed approaches are valued in the most recent intrusion detection datasets: CSE-CICIDS2018 and LUFlow2020. The evaluation fallouts consolidate the previous results and confirm their correctness. [ABSTRACT FROM AUTHOR]

Building machine learning algorithms for real-time clinical decision support systems has become a current research hotspot. The success of these algorithms depends on their ability to handle data stream characteristics. For example, we cite as characteristics the large amounts of data, the high speed and rate of incoming data, and the change in data nature and distribution over time. The Very Fast Decision Tree (VFDT) is a method for incrementally building decision trees. Since its proposition in the literature, it has become one of the most popular tools for data stream classification. This paper aims to optimize a new version of VFDT called McDiarmid Tree (MT) for the early prediction of heart diseases in real-time clinical decision support systems. The proposed method for improving MT performance consists of two main mechanisms: detecting the presence of missing and meaningless values in data attributes and handling the impact of this presence. The proposed MT has been compared with MT and VFDT. Simulation results show that the proposed MT attains significantly higher prediction accuracy with less time and model cost (RAM-Hours) than the other two algorithms. [ABSTRACT FROM AUTHOR]