Welcome to PSGR KCW's Digital Repository
Preserving Knowledge, Empowering Research
- 36 Undergraduate Programmes
- 17 Postgraduate Programmes
- 13 Research Programmes
PSGR Krishnammal College for Women (KCW), inaugurated in June 1963 under the aegis of the GRG Trust of Coimbatore has grown over the last six decades into a temple of learning and academic excellence. Founded on a motto of ‘empowering through education’, the ‘women-only’ KCW symbolizes knowledge, love, and service. KCW is an autonomous college of higher education for women. It is affiliated to the Bharathiar University, Coimbatore.
Access, Share, and Discover Research Materials and Scholarly Works at PSGR Krishnammal College for Women
Recent Submissions
VISUALIZING THE DETECTED COMMUNITIES USING TRADITIONAL ALGORITHMS ON KEYWORD CO-OCCURRENCE NETWORKS (Article)
(Dr. Yashwant Research Labs Pvt. Ltd., 2026) Department of Computer Science (UG); Kiruthika R; Sakkarapani, Krishnaveni
Keyword co-occurrence analysis is essential for understanding emerging trends in research and discovering specific studies. The process of detecting communities by the group nodes in a network based on their interconnection as a structure is called community detection. The community detection algorithm helps analyze and detect the real connection as clusters as a structure within the network. Visualization is one of the significant ways to understand complex networks like community structures. The main aim of this work focuses on visualizing the detected communities based on the co-occurrence of keywords using traditional community detection algorithms. The methodology involves a process of gathering deep learning-based articles from Scopus Bibliographic Dataset (SBD) information based on three major time frames as network datasets, namely SBD_1 as 2006-2013, SBD_2 as 2014-2016 and SBD_3 as 2017. This data is mainly worked with Indexed keyword fields as nodes and their weighted co-occurrences as edges into networks. This work proposed a framework for converting the bibliographic data into graphs for visualizing the detected communities. This work helps scholars to understand the connections among keywords and patterns for their effective research works like extracting academic research articles through exact keyword matchin.
SYZYGIUM CARYOPHYLLATUM LEAF EXTRACT AS A BIOGENIC AGENT FOR ZINC OXIDE NANOPARTICLES: SYNTHESIS, CHARACTERIZATION AND BIOACTIVITY ASSESSMENT (Article)
(Mahendra Publications, 2026) Krishnasreya, Mohandas; Elakkiya, Maruthamuthu Rathinam; Saranya S; Arun, Muthukrishnan; Chithradevi, Balasundaramsaraswathy; Department of Botany; Krishnasreya, Mohandas; Elakkiya, Maruthamuthu Rathinam; Chithradevi, Balasundaramsaraswathy
The successful green synthesis of zinc oxide nanoparticles was accomplished by using the leaf extract of Syzygium caryophyllatum (SC) as a stabilizing agent. The biosynthesized nanoparticles were characterized by UV-Vis spectroscopy, XRD, FE-SEM with EDS and FTIR analysis, confirming the structure, crystalline nature and elemental purity with predominant zinc and oxygen content. The nanoscale dimensions of particles ranging from 23.04 nm to 42.44 nm, with mean particle size 33.91 nm. The DPPH and ABTS radical scavenging activity demonstrated an IC50 of 47.3 µg/mL and 50.43 µg/mL suggesting its strong antioxidant capacity. The antibacterial study showed a dose-dependent inhibition in the growth of E. coli, B. subtilis, MSSA and P. aeruginosa with inhibition zones 24.7 ± 0.6 mm, 23.6 ± 0.8 mm, 27.2 ± 0.6 mm and 22.8 ± 0.3 mm respectively. Cytotoxicity tested on L929 mouse fibroblast cell line indicated high cell viability confirming good biocompatibility. The concentrat ion of 80 to 100 µg/mL provided the favourable balance between strong bioactivity and acceptable biocompatibility in L929 cells. This eco-friendly biosynthesis highlights the multifunctional biological potential of S. caryophyllatum mediated zinc oxide nanoparticles for promising biomedical applications.
SYNTHESIS, CRYSTAL STRUCTURE ELUCIDATION, AND MOLECULAR INTERACTION STUDIES OF A GUANIDINIUM NAPHTHOATE ADDUCT VIA HIRSHFELD AND DOCKING ANALYSES (Article)
(Springer Science and Business Media Deutschland GmbH, 2026) Sivakumar, Nikithaa; Swathika M; Tyagi, Rajdeep; Sagar, Ram; Singh, Ravindra Pratap; Pandey, Shyam S; Natarajan, Arunadevi; Singh, Jay; Department of Chemistry; Natarajan, Arunadevi
A new guanidinium naphthoate adduct in the form of a single crystal was established by adopting a slow evaporation growth technique with a suitable solvent. The bonding interactions, the presence of functional groups, and thermal stability were confirmed using various characterization techniques. The synthesized transparent compound crystallizes in a monoclinic lattice and exhibits centrosymmetric space-group symmetry P1 21/c1. Hirshfeld surface analysis and DFT calculation were performed to investigate the molecular interactions in the crystal. The dependencies of the extinction coefficient and skin depth on photon energy were illustrated. The slope for non-spontaneous, endothermic, and slow reactions was found to be 0.88–0.99 at different temperatures. The abundance of H–H and O–H/H–O interactions reveals that both van der Waals interactions and hydrogen bonding are the primary driving forces in the crystal packing. The blood–brain barrier (BBB) value is 0.46, indicating that the compound can cross the BBB and exhibit biological activity within the central nervous system. The compound demonstrates 95.53% absorption in the human gastrointestinal tract, suggesting good oral bioavailability.
SEMANTIC SIMILARITY LEARNING FOR MEDICAL LITERATURE: A MULTIMODAL FRAMEWORK COMBINING LLM AND DEEP METRIC LEARNING (Article)
(Dr. Yashwant Research Labs Pvt. Ltd, 2026) Department of Computer Science (UG); Deepika A; Rajeswari M
There is an increasing need for artificial intelligence in the field of medical related learning frameworks which is capable of efficiently handling the multimodal data which is under resource constrained situations. Conventional machine learning and deep learning algorithms are usually depending on the large labelled dataset and limited environment with less applicability for real world applications. The adoption of the Artificial intelligence in the field of healthcare is the need for the learning and effective handling of the multimodal healthcare data. Traditional machine learning and deep learning algorithms depends on the large-scale datasets with label, limiting the usage of the current clinical environment. This will limit the applicability of the real-world clinical environments by the data lackness and variance in the domain. In this work, Large Language Models are used with deep metric learning framework for multimodal medical journal classification. In this work, Large Language Model is implemented using deep metric learning framework for the multimodal dataset. The multimodal dataset comprises of text, table and images from online repository. The data manually downloaded from the PubMed website is used in this work. The proposed LLM guide framework outperformed all other models trained by various benchmark datasets.
NON-CATALYTIC HYDROGEN ABSTRACTION KINETICS OF ALDEHYDE–METHYL RADICAL REACTIONS: A DFT BENCHMARK PERSPECTIVE (Article)
(Ram Arti Publishers, 2026) Sambathkumar, Yuvarani; Angamuthu, Abiram; Gopalan, Praveena; Department of Physics; Gopalan, Praveena
Hydrogen abstraction plays a key role in the atmospheric oxidation of aldehydes and halogenated aldehydes. Their reactivity with atmospheric radicals determines the transformation mechanisms and contribution to secondary pollutant formation, which is essential for evaluating their environmental behavior. The primary objective of this study is to benchmark the performance of three density functionals–B3LYP, M06-2X, and ωB97X-D across three basis set combinations (6-311+G(d,p), 6-311++G(d,p) and 6-311++G(2d,2p)) in predicting the non-catalytic H-abstraction reactions of substituted aldehydes (XCHO, where X = CH3, CF3, CCl3, H, F, Cl) in the presence of methyl radical (Ċ H3). The rate constants were computed using conventional transition state theory (TST) with Eckart tunnelling effect over a temperature range of 300-1700 K. The accuracy of each functional is evaluated by comparing the reaction energies, transition states, thermochemical properties and rate constants obtained in both gas and water-mediated environments with the available experimental and high-level theoretical data. The meta-hybrid GGA functional, M06-2X with better electron correlation and electron density localization effects consistently outperformed others in predicting the barrier energies, reaction enthalpies and rate constants. Also, in addition to M06-2X, the range-separated (RS) hybrid GGA functional ωB97X-D showed reasonable accuracy for rate constants at elevated temperatures (1300-1700 K), while B3LYP underestimated barrier energies and overestimated the rate constants. Among the chosen aldehydes, the aldehydic H-abstraction of ClCHO and CH3CHO exhibited the most favourable paths, while FCHO and H-abstraction from methyl site of CH3CHO took least favourable routes across the temperature range. The benchmarking of the considered H-abstraction reactions in general highlights the interplay between the choice of functionals, basis-sets and substituent-driven electronic effects in predicting reaction barriers and rate constants.