In the modern age we live in, the internet has become an essential part of our daily life. A significant portion of our personal data is stored online and organizations run their business online. In addition, with the development of the internet, many devices such as autonomous systems, investment portfolio tools and entertainment tools in our homes and workplaces have become or are becoming intelligent. In parallel with this development, cyberattacks aimed at damaging smart systems are increasing day by day. As cyberattack methods become more sophisticated, the damage done by attackers is increasing exponentially. Traditional computer algorithms may be insufficient against these attacks in the virtual world. Therefore, artificial intelligence-based methods are needed. Reinforcement Learning (RL), a machine learning method, is used in the field of cyber security. Although RL for cyber security is a new topic in the literature, studies are carried out to predict, prevent and stop attacks. In this study; we reviewed the literature on RL's penetration testing, intrusion detection systems (IDS) and cyberattacks in cyber security.

Advances in wireless systems have encouraged the growth and improvement of tiny, low-cost, efficient, and multi-tasking smart sensors. Wireless sensor networks are employed for sensing, collecting data and information, analysing it, and sending it to the main center to be considered quickly. During the performance, a wide range of issues appear, and the existence of some restrictions on the movement of sensors, reliable data collection mechanisms and transferring them properly and securely to the main center, and issues in the sensor network topology. In fact, all traditional methods need more ability to deal with and solve these situations. In this regard, the authors decided to highlight the importance and role of artificial intelligence technologies in designing wireless sensor networks by proposing a hybrid model that helps to construct a more practical sensor network and solve all the situations it faces. These techniques significantly improve network performance, extend battery life, develop smart nodes, and improve network functionality.

This paper introduces BioGPT, a generative transformer-based framework designed to advance personalized genomic medicine and rare disease diagnosis. Unlike conventional models that process either genomic sequences or clinical narratives in isolation, BioGPT employs a cross-modal architecture that effectively fuses both data streams, enabling precise classification and interpretable natural language generation. The model is pre-trained on large-scale genomic and electronic health record datasets and fine-tuned for rare disease tasks. Comprehensive experiments demonstrate BioGPT’s superiority over state-of-the-art biomedical models, including RarePT, BioBERT, and DNABERT, with improvements of up to 10% in F1-score and over 20 BLEU points in justification fluency. Ablation studies highlight the essential contribution of cross-attention mechanisms in enhancing multi-modal synergy. Furthermore, attention-based interpretability techniques show strong alignment with expert clinical markers, ensuring trust and transparency in diagnostic outputs. With sub-second inference times and compatibility with edge deployment strategies, BioGPT proves both effective and deployable in real-world clinical settings. This work establishes BioGPT as a robust, scalable, and explainable AI solution, setting a new benchmark for intelligent diagnostic systems in precision medicine.

Knowledge of the pathogen-host interactions between the species is essentialin order to develop a solution strategy against infectious diseases. In vitro methods take extended periods of time to detect interactions and provide very few of the possible interaction pairs. Hence, modelling interactions between proteins has necessitated the development of computational methods. The main scope of this paper is integrating the known protein interactions between thehost and pathogen organisms to improve the prediction success rate of unknown pathogen-host interactions. Thus, the truepositive rate of the predictions was expected to increase.In order to perform this study extensively, encoding methods and learning algorithms of several proteins were tested. Along with human as the host organism, two different pathogen organisms were used in the experiments. For each combination of protein-encoding and prediction method, both the original prediction algorithms were tested using only pathogen-host interactions and the same methodwas testedagain after integrating the known protein interactions within each organism. The effect of merging the networks of pathogen-host interactions of different species on the prediction performance of state-of-the-art methods was also observed. Successwas measured in terms of Matthews correlation coefficient, precision, recall, F1 score, and accuracy metrics. Empirical results showed that integrating the host and pathogen interactions yields better performance consistently in almost all experiments.