Informatique

Intrusion Detection Systems are widely used today for the security of computer systems. Various machine learning techniques have been used to develop IDS. With the advent of deep learning technologies and its successes in many areas, we have developed an approach based on these methods.
The proposed approach has been tested on the public database CICIDS2017. We reduced the dimensionality of the database using a stacked auto-encoder. We achieved a very low loss rate (10-5 to 10-6). Then we performed a classification (multi-classes) with a convolutional neural network. We achieved a very encouraging accuracy (99.20%). The chosen model (SAE/CNN) gave very satisfactory results in comparison with those of previous work where other approaches were used.

Information communication technologies (ICTs) are becoming very essential and unavoidable in our daily life. Today in Algeria, ICTs exemplified by the internet and interactive multimedia are obviously an important focus for future education. They need effective integration into formal teaching. The problem question of this study is: To what extent is the contribution of ICT to didactics in Algeria? Hence, this study analyzes trends in using ICT by teachers and students in classrooms. Quantitative research was used to collect data from respondents using online Google forms. The principal issues and challenges deemed important in using ICTs in education were: Quality of internet connection, limited time, and lack of effective training.

this paper presents a method to instruct students on how to write their first scientific paper. The method adopts a holistic approach that discusses several different aspects of paper writing. Covered issues include the choice of the paper subject, hypothesis formulation and the design of experiments to test it, the collection and analysis of data, and the planning and writing of the final text. The method has been successfully used to teach graduate and undergraduate students to write their first articles.

this paper presents a method to instruct students on how to write their first scientific paper. The method adopts a holistic approach that discusses several different aspects of paper writing. Covered issues include the choice of the paper subject, hypothesis formulation and the design of experiments to test it, the collection and analysis of data, and the planning and writing of the final text. The method has been successfully used to teach graduate and undergraduate students to write their first articles.

Unmanned aerial vehicles (UAVs) are the emerging vital components of millimeter wave (mmWave) wireless systems. Accurate beam alignment is essential for efficient beam based mmWave communications of UAVs with base stations (BSs). Conventional beam sweeping approaches often have large overhead due to the high mobility and autonomous operation of UAVs. Learning-based approaches greatly reduce the overhead by leveraging UAV data, like position to identify optimal beam directions. In this paper, we propose a reinforcement learning (RL)-based framework for UAV-BS beam alignment using deep Q-Network (DQN) in a mmWave setting. We consider uplink communications where the UAV hovers around 5G new radio (NR) BS coverage area, with varying channel conditions. The proposed learning framework uses the location information to maximize data rate through the optimal beam-pairs efficiently, upon every communication request from UAV inside the multilocation environment. We compare our proposed framework against Multi-Armed Bandit (MAB) learning-based approach and the traditional exhaustive approach, respectively and also analyse the training performance of DQN-based beam alignment over different coverage area requirements and channel conditions. Our results show that the proposed DQN-based beam alignment converge faster and generic for different environmental conditions. The framework can also learn optimal beam alignment comparable to the exhaustive approach in an online manner under real-time conditions.

Unmanned aerial vehicles (UAVs) are the emerging vital components of millimeter wave (mmWave) wireless systems. Accurate beam alignment is essential for efficient beam based mmWave communications of UAVs with base stations (BSs). Conventional beam sweeping approaches often have large overhead due to the high mobility and autonomous operation of UAVs. Learning-based approaches greatly reduce the overhead by leveraging UAV data, like position to identify optimal beam directions. In this paper, we propose a reinforcement learning (RL)-based framework for UAV-BS beam alignment using deep Q-Network (DQN) in a mmWave setting. We consider uplink communications where the UAV hovers around 5G new radio (NR) BS coverage area, with varying channel conditions. The proposed learning framework uses the location information to maximize data rate through the optimal beam-pairs efficiently, upon every communication request from UAV inside the multilocation environment. We compare our proposed framework against Multi-Armed Bandit (MAB) learning-based approach and the traditional exhaustive approach, respectively and also analyse the training performance of DQN-based beam alignment over different coverage area requirements and channel conditions. Our results show that the proposed DQN-based beam alignment converge faster and generic for different environmental conditions. The framework can also learn optimal beam alignment comparable to the exhaustive approach in an online manner under real-time conditions.

In a world where linguistic and logical barriers are dissolving and communication must be possible on demand and in a language one is familiar with, Machine Translation (MT) has become essential. This natural language processing sub-eld is now the most exciting but also the most dicult topic to solve, as it automates the translation process and decreases the reliance on human translators. Researchers are attempting to translate English into their native language and vice versa, but achieving perfect MT has proven to be a dicult task for researchers all over the world. This paper provides a high-level overview of this subject reviewing the accomplishments and ndings of various MT studies in the context of Arabic language, which is known by its richness and complexity. At the end of the survey, open challenges and issues are discussed. Therefore, this study might be useful for future research and for selecting the best method for a certain purpose.