Alva's Institute of Engineering and Technology

Rivest-Shamir-Adleman (RSA) algorithm is essential for safe interaction, authentication, and data security in low-power and space-constrained devices. But because these devices have limited resources, RSA needs to be utilized in a way that maximizes battery life, reduces hardware footprint, and strikes an efficient balance between security and usability. To accomplish the required security within the constraints of the device, this frequently entails utilizing reduced key sizes, improving algorithmic implementations, and combining RSA with other cryptographic approaches. This proposed work, focused a minimum of 16 keys in order to guarantee performance and reduce total overhead by considering the given scenario. This proposed method ensures excellent speed and security. The results of the simulation demonstrate that the output will be revealed at the receiving end when the decryption key is d = 03. Specifically, this approach is used to create Internet of Things (IoT) devices when area, power, and speed are more critical than security.

Development of polymeric nanofibers using the electrospinning technique has found potential candidature for tremendous applications including protective clothing, sensors, energy sectors, tissue engineering, wound healing, air filtration, liquid filtration and cosmetics. This is because of the potentiality of the electrospinning method in producing ultrathin fibers ranging from nano to micrometers. Numerous efforts have been made by researchers using electrospinning techniques to regulate the morphology of the developed fibers. Similarly, various polymers and polymer composites have been employed for the development of fibers using electrospinning techniques for different applications. In the current research work, the electrospun nanofibers were produced from a composite solution of polydimethylsiloxane-polyvinyl alcohol (PDMS-PVA) polymers and synthesized zinc silicate (ZnSiO3) nanoparticles. Nanofibrous membranes were produced from the prepared polymer composite solution using electrospinning under high voltage. A morphological study conducted by scanning electron microscopy revealed uniform fibers with diameters ranging from 142 nm to 410 nm. Wettability studies of the electrospun nanofibrous membrane showed an increase in water contact angle from 43 degrees for pristine PDMS-PVA to 92 degrees for ZnSiO3 loaded membranes thereby indicating enhanced hydrophobicity. Air filtration performance testing on the membranes demonstrated that the filtration efficiency was improved from 63 degrees for P0 to 92 degrees for P3, while the quality factor increased from 0.034 Pa-1 to 0.368 Pa-1. Among all the samples, the membrane P3 demonstrated the best performance, whereas P0 showed the lowest performance. Thus, it can be inferred that the incorporation of ZnSiO3 greatly enhanced the air filtration capability of the composite membranes, thereby making the membranes a suitable candidate for air filtration applications and air quality management.

The conventional method of visually inspecting tea crops is time-consuming, requires human judgement, and is unsuitable for large scale tea plantations. This paper proposes a deep learning automated framework to detect tea leaf diseases using Convolutional Neural Networks (CNN) combined with Generative Adversarial Networks (GAN). The GAN will provide realistic synthetic images to augment the training set; this is an effective approach to addressing class imbalance and improve the generalization of the model. The study finds that the CNN captures the discriminatory spatial characteristics of the tea leaf images and classifies the leaf images into three categories of diseases. The framework’s classification results indicate an overall accuracy of 97.6%, precision of 96.9%, recall of 97.8%, F1 score of 97.3%, and specificity of 97.1%. The study also finds that the proposed CNN-GAN framework greatly improves the robustness and classification accuracy of tea disease detection compared to traditional CNN, ResNet-50, Inception-v3 and EfficientNet-B3. Additionally, the proposed CNN-GAN framework provides efficient inference time which can be employed in real-life applications and highlight GAN assisted deep learning methods for achieving accurate results at low cost.

The global mental health crisis, exacerbated by stigma and a shortage of professionals, necessitates innovative, accessible solutions. This paper presents EmpatheticAI, a hybrid mental health support chatbot that synergizes classical Natural Language Processing (NLP) techniques with robust machine learning to deliver context-aware, proactive emotional support. Our model leverages a multi-faceted feature extraction pipeline, combining TF–IDF vectorization, sentiment intensity analysis (VADER), and linguistic metadata to accurately classify user emotional states into Distressed, Neutral, and Positive categories. We empirically compare the performance of Logistic Regression against an Ensemble Random Forest classifier, with the latter achieving superior performance (91% accuracy, 90% F1-score). The deployed system integrates these capabilities into a holistic support application, featuring guided meditation prompts, a reflective journaling module, a peer-support community forum, and a therapist discovery service. This work underscores the significant potential of transparent, hybrid AI models in providing scalable, immediate, and ethically-grounded mental health first aid, while rigorously addressing limitations and charting a path for future integration of deep learning architectures.

The rise in digital text, PDF files, and coding has made plagiarism detection even more intricate and complex. Some of these complex plagiarism cases involve paraphrasing and coding. This is because traditional plagiarism detection methods do not work well in these situations. This paper discusses a proposed Smart Content Integrity System that incorporates traditional methods of natural language processing and language models in detecting plagiarism. The proposed method detects plagiarism in digital text files and source coding. The method has been shown to work better in detecting plagiarism than other methods. This enhances academic integrity in digital environments.