Dr. Babasaheb Ambedkar Technological University

Artificial Intelligence (AI) is emerging as a transformative force in the agricultural sector by improving productivity, optimizing resource utilization, and promoting sustainable farming practices. The integration of AI technologies such as machine learning, computer vision, robotics, Internet of Things (IoT), and big data analytics has enabled precision agriculture, smart irrigation systems, yield prediction models, pest and disease detection, and supply chain optimization. In countries like India, where agriculture remains a major source of livelihood, AI has the potential to address structural issues such as low productivity, climate variability, fragmented landholdings, and market inefficiencies. AI-powered platforms assist farmers in real-time decision-making by analyzing weather data, soil conditions, crop health, and market trends. Despite its transformative potential, the adoption of AI in agriculture faces several challenges including high initial costs, inadequate digital infrastructure, lack of technical awareness, data privacy concerns, and limited access to quality datasets. This research paper examines the role of AI in agriculture, evaluates its economic and social implications, and identifies the challenges associated with its implementation. The study concludes that effective policy support, public-private partnerships, capacity-building initiatives, and digital infrastructure development are crucial to maximizing the benefits of AI in agriculture.

Nanostructured electrode materials are increasingly recognized for enhancing the efficiency and stability of perovskite solar cells (PSCs). This study systematically investigates the synthesis, integration, and performance of diverse nanostructured electrodes, including nanoparticles, nanowires, nanotubes, and hybrid configurations, focusing on their impact on charge transport, light absorption, and overall device reliability. Incorporation of conductive nanosilver and carbon‐based nanostructures into electrode designs demonstrates a marked improvement in charge collection and a notable reduction in resistive losses. Quantitative measurements show significant gains in power conversion efficiencies: standard silver electrodes achieved 23%, nanosilver electrodes 26%, carbon nanotube electrodes 21%, and hybrid nanostructured electrodes reached up to 30%, underscoring the advantage of optimized morphologies and enhanced surface properties. Stability analysis reveals improved durability across all nanostructured variants, with hybrid electrodes showing promising long‐term performance. Scalability was assessed by employing screen‐printing and solution‐based fabrication techniques suitable for large‐area deposition, confirming that these nanostructured materials retain high efficiency when integrated into flexible and sizable substrates. Comprehensive material characterization, including scanning electron microscopy and impedance spectroscopy, substantiates the superior electrochemical and morphological features of nanostructured electrodes. This work demonstrates that nanostructured electrodes offer a practical path for upscaling PSCs, facilitating industrial adoption and flexible device fabrication without compromising efficiency or stability. Future directions include advancing synthesis methods and expanding to tandem and flexible solar technologies to further enhance performance, stability, and scalability, paving the way for practical, high‐efficiency renewable energy solutions.

In an era of social media and online learning platforms, there are several opportunities for learning different technologies and topics that students do not easily understand. However, it also presents challenges by diverting students’ attention, such as notifications, multitasking activities, advertisements, etc. Assessing students' level of focus during cognitive tasks is crucial and complex. This study evaluates students' cognitive engagement through various activities, including arithmetic calculations, reading technical articles, listening to technical podcasts, reading transcripts, browsing the internet, and engaging in relaxation exercises, utilizing EEG signals. Concentration levels are classified using deep learning algorithms, specifically Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and Artificial Neural Networks (ANN). The performance of these algorithms is also evaluated based on metrics such as accuracy, F1 Score, precision, and loss.

Objectives: Adenium obesum has a long history of usage in management of inflammatory illnesses; the current research intends to learn more about its phytochemical composition and assess its anti-arthritic and anti-inflammatory efficacy in vitro using an ethanolic extract and column fractions. Methods:The plant was gathered, identified and Soxhlet apparatus used to extract it ethanol. A preliminary phytochemical screening was conducted in order to find out important constituents. Seven solvent systems were used to carry out TLC profiling to assist in proper fractionation. Subsequently, column chromatography was done and four pooled fractions were produced onbase of comparable TLC pattern. The stability of HRBC membrane was used to assess anti-arthritic action in vitro, whereas suppression of COX-2 was used to test anti-inflammatory capability.The GC-MS based analytical method was used to conclude bioactive compound in crude extract. Results:TLC profiling confirmed the presence of single spots with distinct Rf values across solvent systems. HRBC assay revealed that Fraction 4 showed the highest membrane stabilization at 100 µg/mL (91.80%), comparable to Aspirin (93.45%). In COX-2 inhibition assay, Fraction 3 exhibited the highest inhibition (72.08% at 100 µg/mL) associated to standard (85.06%). GC-MS analysis identified citronellal (64.37%), linalool (3.04%), and terpinen-4-ol (1.06%) among others, compounds well-documented for their anti-inflammatory activities. Conclusion:Identification and extraction ethanol with Soxhlet apparatus was performed on the plant. The initial screen of phytochemical was carried out to determine significant constituents. TLC profiling was done in seven solvent systems that helped in proper fractionation. Column chromatography was then performed and four fractions were pooled according to shared similar pattern of TLC results. The in vitro determination of anti-arthritic action was performed via HRBC membrane stabilization test and inhibitory response was evaluated in a COX-2 test. Identification of the bioactive compound in crude extract was done based on the analytical method of GC-MS.

In silico drug-target interaction (DTI) prediction plays a key role in accelerating drug discovery and understanding molecular mechanisms. Traditional methods often struggle with the complexity and scale of biochemical data, thus limiting prediction accuracy. This study presents VGAN-DTI, a generative AI framework that combines generative adversarial networks (GANs), variational autoencoders (VAEs), and multilayer perceptrons (MLPs) to improve DTI predictions. The framework precisely encodes molecular features, uncovers underlying molecular mechanisms, and enhances predictive capabilities. GANs generate diverse molecular candidates, whereas VAEs optimize the feature representations. MLPs trained on BindingDB classify interactions and predict binding affinities. The model achieves 96% accuracy, 95% precision, 94% recall, and 94% F1 score, outperforming existing methods. Rigorous ablation studies validated the robustness of the framework. The proposed system enhances drug discovery by optimizing innovation, synthetic feasibility, and predictive accuracy, ensuring reliable DTI predictions and advancing data-driven pharmaceutical research.