Significant growth of internet users signifies that more autonomous systems need to be designed. Software-Defined Network (SDN) replaces traditional networks due to their solubility and capability to integrate with artificial intelligence. It also makes networking future poof. A technological shift often comes with shortcomings, and SDN is no exception. Anomaly detection is a crucial task. Existing anomaly detection methods use traditional statistical and machine learning-based models that may perform poorly due to two key issues: (i) the unavailability of all training network-packets at the beginning since these packets are received over time and the training dataset gradually grows over time, and (ii) limited memory and computation capacity of the computing system for model building, preventing the storage of all network-packets that arrive over time for a prolonged period. To address the issues, this paper introduces a novel functionality in SDN called Adaptive anomaly Detection Method (ADM), which utilises an existing technique for facilitating incremental learning so that ADM can handle new anomaly types by keeping track of historical information. ADM is evaluated using four criteria by comparing its performances against four state-of-the-art methods on two publicly available datasets. Our experimental results demonstrate that ADM performs significantly better than existing techniques. The source code of this work has been made publicly available at https://github.com/eferdous/adm

BACKGROUND: Informed consent is fundamental to ethical medical practice, ensuring that patients understand the procedures they undergo, the associated risks, and available alternatives. The advent of artificial intelligence (AI) in healthcare, particularly in diagnostics, introduces complexities that traditional informed consent forms do not adequately address. AI technologies, such as image analysis and decision-support systems, offer significant benefits but also raise ethical, legal, and practical concerns regarding patient information and autonomy. MAIN BODY: The integration of AI in healthcare diagnostics necessitates a re-evaluation of current informed consent practices to ensure that patients are fully aware of AI's role, capabilities, and limitations in their care. Existing standards, such as those in the UK's National Health Service and the US, highlight the need for transparency and patient understanding but often fall short when applied to AI. The "black box" phenomenon, where the inner workings of AI systems are not transparent, poses a significant challenge. This lack of transparency can lead to over-reliance or distrust in AI tools by clinicians and patients alike. Additionally, the current informed consent process often fails to provide detailed explanations about AI algorithms, the data they use, and inherent biases. There is also a notable gap in the training and education of healthcare professionals on AI technologies, which impacts their ability to communicate effectively with patients. Ethical and legal considerations, including data privacy and algorithmic fairness, are frequently inadequately addressed in consent forms. Furthermore, integrating AI into clinical workflows presents practical challenges that require careful planning and robust support systems. CONCLUSION: This review proposes strategies for redesigning informed consent forms. These include using plain language, visual aids, and personalised information to improve patient understanding and trust. Implementing continuous monitoring and feedback mechanisms can ensure the ongoing effectiveness of these forms. Future research should focus on developing comprehensive regulatory frameworks and enhancing communication techniques to convey complex AI concepts to patients. By improving informed consent practices, we can uphold ethical standards, foster patient trust, and support the responsible integration of AI in healthcare, ultimately benefiting both patients and healthcare providers.

Transfer learning aims to learn classifiers for a target domain by transferring knowledge from a source domain. However, due to two main issues: feature discrepancy and distribution divergence, transfer learning can be a very difficult problem in practice. In this paper, we present a framework called TLF that builds a classifier for the target domain having only a few labeled training records by transferring knowledge from the source domain having many labeled records. While existing methods often focus on one issue and leave the other one for further work, TLF is capable of handling both issues simultaneously. In TLF, we alleviate feature discrepancy by identifying shared label distributions that act as the pivots to bridge the domains. We handle distribution divergence by simultaneously optimizing the structural risk functional, joint distributions between domains, and the manifold consistency underlying marginal distributions. Moreover, for the manifold consistency we exploit its intrinsic properties by identifying$k$nearest neighbors of a record, where$k$is determined automatically. We evaluate TLF on seven publicly available datasets and compare performances of TLF and fourteen state-of-the-art techniques. Our experimental results, including statistical sign test and Nemenyi test analyses, indicate a clear superiority of TLF over the state-of-the-art techniques.

Data pre-processing and cleansing play a vital role in data mining for ensuring good quality of data. Data cleansing tasks include imputation of missing values, and identification and correction of incorrect/noisy data. In this paper, we present a novel approach called Co-appearance based Analysis for Incorrect Records and Attribute-values Detection (CAIRAD). For a data set having incorrect/noisy values CAIRAD separates the noisy records from the clean records. It thereby produces two data sets; a clean data set and a data set having all noisy records. It also reports noisy attribute values of each noisy record. We evaluate CAIRAD on four publicly available natural data sets by comparing its performance with the performance of two high quality existing techniques namely RDCL and EDIR. We use various patterns (of noisy values) each having different noise levels. Several evaluation criteria such as error recall (????), error precision (????), F-measure, record removal ratio (??????), and area under a receiver operating characteristics curve (AUC) are used. Our experimental results indicate that CAIRAD performs significantly better (based on t-test analysis) than RDCL and EDIR.