Network data is expanding and that too at an alarming rate. Besides, the sophisticated attack tools used by hackers lead to capricious cyber threat landscape. Traditional models proposed in the field of network intrusion detection using machine learning algorithms emphasize more on improving attack detection rate and reducing false alarms but time efficiency is often overlooked. Therefore, in order to address this limitation, a modern solution has been presented using Machine Learning-as-a-Service platform. The proposed work analyses the performance of eight two-class and three multiclass algorithms using UNSW NB-15, a modern intrusion detection dataset. 82,332 testing samples were considered to evaluate the performance of algorithms. The proposed two class decision forest model exhibited 99.2% accuracy and took 6 seconds to learn 1,75,341 network instances. Multiclass classification task was also undertaken wherein attack types like generic, exploits, shellcode and worms were classified with a recall percentage of 99%, 94.49%, 91.79% and 90.9% respectively by the multiclass decision forest model that also leapfrogged others in terms of training and execution time.

Aphasia is a loss or an impairment of language that is due to brain damage, which affects the production of speech and the ability to read or write. his lead to the necessity of large human resources in terms of a speech therapist and physicians, caretakers, etc. Having an automatic computer/mobile-based technology for the detection, evaluation and providing timely feedback would give a great benefit to the medical community immensely. It also speeds up the process with minimum efforts. This has motivated us to create a knowledge base to build machine learning model to understand and learn from the available data for an accurate prediction. The Automatic Speech Recognition (ASR) technology has the potential to enable individuals with aphasia using computer based conversion of spoken word to a text form. This would make it possible for language therapy software to provide feedback about the correctness of the user's spoken utterances or to engage the user in spoken dialog practice or other language therapy activities. ASR hence must be developed further in order to make it feasible for people with speech impairments.

Accurate segmentation of spinal structures, including vertebrae, intervertebral discs (IVDs), and the spinal canal, is crucial for diagnosing lumbar spine disorders. Deep learning-based semantic segmentation has significantly improved accuracy in medical imaging. This study proposes an enhanced U-Net incorporating an Inception module for multi-scale feature extraction and a dual-output mechanism for improved training stability and feature refinement. The model is trained on the SPIDER lumbar spine MRI dataset and evaluated using Accuracy, Precision, Recall, F1-score, and mean Intersection over Union (mIoU). Comparative analysis with the baseline models—U-Net, ResUNet, Attention U-Net, and TransUNet—shows that the proposed model achieves superior segmentation accuracy, with improved boundary delineation and better handling of class imbalance. An evaluation of loss functions identified Dice loss as the most effective, enabling the model to achieve an mIoU of 0.8974, an accuracy of 0.9742, a precision of 0.9417, a recall of 0.9470, and an F1-score of 0.9444, outperforming all four baseline models. The Inception module enhances feature extraction at multiple scales, while the dual-output mechanism improves gradient flow and segmentation consistency. Initially focused on binary segmentation, the approach was extended to multiclass segmentation, enabling separate identification of vertebrae, IVDs, and the spinal canal. These enhancements offer a more precise and efficient solution for automated lumbar spine segmentation in MRI, thereby supporting enhanced diagnostic workflows in medical imaging.

Free smartphone applications that aim to promote physical activity or reduce sedentary behavior at workplaces were questioned for its content such as technical features, behavior change techniques (BCT) and security concerns. To evaluate systematically whether smartphone apps are mapped adequately with BCTs and security concerns. Free apps in Google Play store were searched and coded for BCTs systematically in July 2020. Two reviewers independently coded apps that aimed to promote physical activity (n = 18) and reduce sedentary behavior (n = 16) using a taxonomy of 26 BCTs. BCTs, features and security concerns among apps promoting physical activity and reducing sedentary behavior were analyzed. While the features (social support, gamification, or rewards) help increase physical activity and decrease sedentary behavior, the apps do not typically include the above features. Further, the apps were based on an average of 13 (4–21) BCTs. The BCTs coded most frequently were “self-monitoring” (100%), “general encouragement” (97%), and “goal setting” (n = 94%). No significant relation between user rating and BCTs was found in the apps. Majority of the free apps that aimed to increase physical activity or reduce sedentary behavior at workplaces were not adequately based on BCTs and lacked essential features facilitating long term behavior compliance at workplaces. Hence, there is a need to develop newer applications mapped adequately with BCTs, involving the collaborative work of behavioral scientists, app developers and policymakers. Clinical Trial Registry of India CTRI/2020/03/024138.

Although the File Transfer Protocol (FTP) was not designed with strict security measures in place, which is quite evident from the fact that all the commands and data are transmitted in clear text, subsequent additions to the protocol have implemented security features like upgrading the connection to Transport Layer Security (TLS) in FTPS. In spite of such developments, FTP remains an inherently insecure protocol, that refuses to die. The FTP, despite its quirks, has proven to be extremely resilient. As of 2015, more than 13 million FTP servers exist in the Internet Protocol Version 4 (IPv4) address space. With the advent of the Internet of Things (IoT) and the proliferation of internet enabled hand held devices, the security of computer networks has become a very important issue. There are two broad divisions of security threats, Remote exploits and local attacks. Remote to local attacks are a serious threat to computer networks. In this paper, we implement two specific remote to local attacks, the Warezmaster (WM) and the Warezclient attack (WC), which exploit the vulnerabilities present in “anonymous” FTP on both Linux and Windows. We then go on to discuss the ramifications of such an attack, it's limitations as well as the methods to prevent such attacks.