SVKM's NMIMS is a private deemed university located in Mumbai. The University also has campuses at Shirpur, Bangalore, Hyderabad, Indore and Navi Mumbai and two upcoming campuses in Dhule and Chandigarh. It has 17 constituent schools that offer both undergraduate and postgraduate courses in management, engineering, commerce, pharmacy, architecture, economics, mathematical sciences, hospitality, science, law, aviation, liberal arts, performing arts, architecture & design. It is accredited by NAAC with 3.59 CGPA and Grade A+. NMIMS was also awarded Category I University statusby MHRD.
Rice is one of the simple food crops that has been cultivated in the majority of countries. Rice leaf diseases (RLDs) are a major problem in crop production since they may result in low productivity and economic losses. Traditional ways of detecting an illness may be time-consuming and even labor-intensive, and at times may need specialized skills. The popularity of preceding works on detecting RLDs has relied on machine learning (ML) and image processing approaches. On the other hand, deep learning (DL) methodologies are more applicable in disease detection problems because they can learn stipulated patterns on big data without using feature extraction methods. This systematic review explores various ML and DL methods used in the literature for RLD detection, which includes survey articles based on convolutional neural network (CNN), transfer learning, and advanced AI approaches. The review of existing open-source datasets is also discussed in this survey. In addition, it examines limitations of current models related to practical implementation, data diversity, domain adaptation, and hardware limitations. Lastly, this survey identifies future research directions to improve the strength and usage of DL models in real-world agriculture settings. This survey comprehensively reviews more than 70 peer-reviewed publications (2019-2025) sourced from IEEE, Elsevier, Springer, ACM, and MDPI digital libraries.
In the critically polluted Yamuna River, a lifeline for the Delhi-NCR region, the interplay between seasonal contaminant loads and natural phytoremediation remains poorly understood. This study presents the quantification of eight priority pharmaceutical and personal care products (PPCPs) in water with evaluation of the bioaccumulation potential in Eichhornia crassipes along a 22 km area in pre- and post-monsoon. Paracetamol (up to 2866.9 ± 41.60 ng/L) and mefenamic acid (up to 829.5 ± 5.48 ng/L) were the most abundant analytes found in post-monsoon with frequencies of detection (FoD) > 90%. The bioaccumulation assessment was performed in E. crassipes that was naturally abundant in pre-monsoon season but less so in winter, coinciding with a notable rise in PPCP concentrations. The study revealed significant uptake with the assistance of the bioaccumulation factor. The contaminant's retention and distribution in surface water are directly influenced by the seasonal dieback of E. crassipes in river water. Notably, seasonal analysis revealed that pre-monsoon concentrations of carbendazim were significantly higher than post-monsoon levels (p < 0.05). These findings emphasize the regulation of PPCP dynamics in urban rivers by the phytoremediation potential of E. crassipes and highlight the need for integrating natural wetland vegetation into water quality management strategies.
This paper examines whether tighter global dollar conditions weaken firm-level market liquidity in India's benchmark equity segment and whether that relationship varies with baseline structural fragility. We use a balanced firm-day panel of 50 NIFTY-50 firms covering 30 January 2018 to 31 December 2024. To capture the external tightening environment, we construct a Global Dollar Tightening (GDT ) index from standardised daily changes in the US 10-year Treasury yield, the VIX, and the USD-INR exchange rate. We also construct a Firm Fragility Index (FFI) from pre-2020 liquidity and risk characteristics in order to distinguish structural vulnerability from contemporaneous market stress. The estimates show that a one-standard-deviation tightening shock is associated with an immediate increase of about 0.26% in firm-level illiquidity. Although that short-run effect is modest, its economic significance becomes more visible once liquidity persistence is taken into account, with the implied longer-run effect rising to roughly 5.6%. Firms with higher baseline fragility display moderately stronger liquidity sensitivity, whereas evidence of additional amplification during market-wide stress becomes weaker under conservative two-way clustered inference. Overall, the findings suggest that external tightening affects trading depth even within India's most liquid benchmark segment, but the cross-sectional amplification should be interpreted cautiously. These findings should be read within the paper's empirical scope, which is limited to India's benchmark large-cap equity segment and a reduced-form proxy for external tightening rather than a structural identification of monetary-policy surprises.
Amidst the rapid advances of large language models (LLMs), most LLMs still struggle with mixed-language inputs, limited Code-switching (CSW) datasets, and evaluation biases, which hinder their deployment in multilingual societies. This survey provides the first comprehensive analysis of CSW-aware LLM research, reviewing 327 studies spanning five research areas, 15+ NLP tasks, 30+ datasets, and 80+ languages. We classify recent advances by architecture, training strategy, and evaluation methodology, outlining how LLMs have reshaped CSW modelling and what challenges persist. The paper concludes with a roadmap emphasizing the need for inclusive datasets, fair evaluation, and linguistically grounded models to achieve truly multilingual intelligence. A curated collection of all resources is maintained at https://github.com/lingo-iitgn/awesome-code-mixing/.
The proposed research addresses the challenge of detecting malicious network traffic in IoT environments, focusing on enhancing detection accuracy while ensuring interpretability. The proposed attention fusion classification model utilizes both long-term and short-term attention mechanisms to capture temporal patterns and protocol-specific features, which improves the differentiation between benign and malicious traffic. Empirical results indicate strong performance, with precision-recall scores of 0.9999 for both the DDoS_TCP and DDoS_UDP classes, and a perfect score of 1.0000 for the Normal class. The model also demonstrates solid performance for the DDoS_HTTP (0.9791), Password (0.9418), and SQL_Injection (0.9461) classes. Furthermore, it excels at identifying complex behaviors in upload-based attacks and network vulnerabilities, achieving precision-recall scores of 0.9333 for the Uploading class and 0.9963 for the Vulnerability Scanner class. The binary classification accuracy is 99.9966
