AULIB-IR Central Library

Communities in Avinashilingam Institute for Home Science and Higher Education for Women - AULIB-IR Central Library

Select a community to browse its collections.

Now showing 1 - 3 of 3

Recent Submissions

Item

Analytical Study on Social Networking among the Youth

(Avinashilingam, 2025-10) Sinjitha V; Guide - Dr.R.Jansi Rani

Social networking sites have gained prominence as powerful digital platforms that have crucial influence on the social, emotional and scholastic development of youth. Social networking platforms have become necessary tools for communication, education and career development. This present study aims to explore the knowledge, attitude and usage patterns of social networking among youth from rural and urban settings, in Thiruvananthapuram district, Kerala. The research incorporates a mixed method design that incorporates quantitative and qualitative approaches based on interview, case studies and awareness programmes. To assess the effectiveness of awareness initiatives, pre test and post test assessments were conducted to measure changes identified in respondents understanding. Among the 600 sample, from student youth and employed youth, 150 samples were fragmented from the rural student youth and 150 samples from rural employed youth. Similarly, 150 samples were selected among urban student youth and 150 from urban employed youth. Among rural student youth, 89 percent demonstrate adequate knowledge of Facebook, 76 percent were capable of profile creation and 88 percent used these platforms primarily for communication. Furthermore, 84 percent recognized social networking as a source of knowledge development and 64 percent employed it for educational purposes. The study also revealed that 87 percent of rural youth reported improvement in language proficiency, 73 percent in personal calibre and 83 percent in photography and creative skills. Meanwhile, 81 percent refrained from chatting with strangers. Among urban youth, a higher proportion 61 percent spend time on online games, whereas 79 percent of employed youth believed that social networking has enhanced their employment opportunities. The analysis highlights a vital association between the location of youth and their levels of social networking knowledge, emotional and behavioural responses and overall attitudes. The findings affirmed that potential of social networking sites to promote youth empowerment and socio economic development. The study concluded that structured education and awareness programmes can guide youth toward responsible, productive and balanced use of social networking. Furthermore, the increasing internet accessibility in rural regions is helping bridge the digital divide, facilitating equal participation in the nation’s development process. The study faced challenges due to COVID-19–related restrictions, which constrained fieldwork, participant interaction, and the inclusion of respondents beyond one district and age group. Key words: Youth, Social Networking, Rural Students, Rural Employed, Urban Students, Urban Employed,

Item

Evaluation of banks’ and customers’ perception towards the adoption of sustainable banking practices in the select indian commercial banks

(Avinashilingam, 2025-12) Sina E. S.; Guide - Dr.D.Vennila

The integration of sustainable practices into business operations and investment decisions depicts a commitment to a sustainable future. This validates a climate-focused initiative, namely the Net-Zero. India is striving to address climate change and achieve Net Zero by 2070 by promoting responsible consumption and a low-carbon strategy. Indian banks are geared up to play a leading role in facilitating this transition, with an unwavering commitment to sustainability. Although sustainability adoption and its opportunities have been widely studied across advanced and emerging economies, research focusing specifically on the Indian banking sector remains limited. In particular, the relationship between environmental concerns and banks’ commitment to sustainability has not been thoroughly explored. Moreover, many existing studies overlook a holistic perspective that considers both banks and their customers, creating a population gap. From a methodological standpoint, while some frameworks have been proposed, little empirical research has applied Structural Equation Modelling (SEM) to examine perceived environmental commitment, behavioural intention, and actual use of sustainable banking practices by customers in the Indian context. Conceptually, most literature highlights external drivers such as regulatory pressures and market demand, but there is limited understanding of how the perceptions of banks and customers influence sustainability-related decision-making. This study aims to address these gaps by theoretically and empirically investigating the factors shaping the intention to adopt sustainability in banking operations. The objectives of this study are to analyse the sustainable banking practices adopted by Indian commercial banks, the factors influencing the customer’s behavioural intention to use and actual use of sustainable banking practices, the benefits and challenges faced by each group. The research employs a quantitative approach, utilising a well-structured questionnaire from 850 bank branches and 1150 customers across India, determined through a proportionate sampling technique. Grounded in the TAM and VBN, the study uses Structural Equation Modelling to analyse the structural relationship of the independent variables: perceived ease of use, perceived usefulness, and perceived environmental commitment on the dependent variable: behavioural intention to use and actual use of sustainable banking practices. The results depict that Indian banks have increasingly adopted sustainability in their operations. Commonly adopted practices include digital banking and e-statements to reduce paper usage, financing green projects and renewable energy sectors, green building certifications for bank branches and solar-powered ATMs. Public sector banks focus more on regulatory compliance, while private sector banks are leading in voluntary green initiatives. Compared to public sector banks, Private sector banks tend to be more proactive and innovative in adopting green policies. The access to green financial products (like loans for EVs or solar panels), energy-efficient purchases and investments led to a sense of contribution to environmental sustainability. Customers report limited awareness, lack of proper education in sustainable banking practices and digital divide (especially in rural areas), concerns over data privacy and digital transaction safety and scepticism about bank motives. Customers perceive benefits such as convenience, a secure and paperless banking experience, and alignment with personal environmental values. Banks face barriers such as high initial investment in technology and green infrastructure, a lack of standardised policies, high implementation costs, and inadequate training of employees. Absence of unified sustainable banking guidelines is also a challenge faced by the banks. Both banks and customers express the need for better incentives and clearer communication of sustainable practices. The adoption and reporting of sustainable banking practices increases brand image, customer satisfaction, data security and customer privacy. Banks can take initiatives on financial inclusion and community development,product innovation with ESG consideration, sustainable business strategy, sustainable financing framework for the Bank’s issuance of sustainable instruments.” Keywords: Adoption of sustainable banking practices, Customers’ perception, Indian Commercial Banking Sector, Structural Equation Modelling

Item

Trembling Tenors of Life: A Psycho-social Approach to Trauma in Select Arab and Iranian Women Writings in English

(Avinashilingam, 2025-01) Kavitha P K; Guide - Dr. S. Kalamani

Trembling Tenors of Life: A Psycho-social Approach to Trauma in Select Arab and Iranian Women Writings in English Scheherazade, the legendary storyteller of One Thousand and One Nights, used her voice and stories to resist death, a metaphor for how Arab and Iranian women writers today use literature to fight silencing and marginalisation. Just as Scheherazade’s tales ensured her survival, these contemporary writers employ storytelling to navigate the complex realities of trauma, identity, and resistance in their lives. Arab and Iranian women writers, such as Azar Nafisi, Basma Abdel Aziz, and Jokha Alharthi, face immense challenges in gaining visibility within a literary world, long dominated by Western narratives and perspectives. Despite this, their works have emerged as powerful contributions to global literature, providing critical insights into the intersections of gender, trauma, and socio-political resistance. The present thesis analyses Azar Nafisi’s Reading Lolita in Tehran, Basma Abdel Aziz’s The Queue, and Jokha Alharthi’s Celestial Bodies, focusing on how the characters are traumatised by both personal and political contexts, shaping the identity and resilience of female protagonists in contemporary Arab and Iranian literature based on the broader historical and socio-political forces at play, such as the Iranian Revolution and the Arab Spring. The research applies trauma theories of Cathy Caruth, Shoshana Felman, Dori Laub and several other theorists to investigate how women’s experiences of trauma are portrayed and processed in these narratives. The thesis explores whether the responses to trauma in these texts are gender-specific, focusing on how women navigate societal expectations related to class, politics, and gender roles. It also analyses how literature, fine arts and crafts can help in alleviating the mental agony the women have due to ix trauma experienced in their lives. These creative outlets become essential for the characters’ survival, offering not only relief from mental anguish but also a method for reclaiming agency in a society that seeks to control their narratives. Through the lens of postcolonial theory of trauma studies in literature, it also examines how these female writers resist and rewrite the Western-dominated literary landscape, creating space for indigenous voices that challenge colonial and patriarchal narratives. Nafisi’s Reading Lolita in Tehran presents a poignant reflection on how trauma, particularly the trauma of living under an oppressive regime, impacts the intellectual and emotional lives of Iranian women. The personal trauma of these women is mirrored by the collective trauma of a society undergoing profound political repression. Similarly, Aziz’s The Queue examines the psychological toll of living under an authoritarian military regime in post-Arab Spring Egypt. It speaks of the role of silence and invisibility imposed on citizens in authoritarian contexts; challenging the assumption that trauma is always voiced. Alharthi’s Celestial Bodies offers a deep exploration of both domestic and transgenerational trauma passed down through families and former slaves in Oman. The novel intricately weaves together the lives of people especially women across generations, examining how colonial legacies, slavery, and socio-political transformation affect their identity and resilience. Ultimately, the thesis reveals how contemporary Arab and Iranian women writers not only reflect the trauma of their societies but also craft alternative spaces for healing and empowerment. The study contributes to a broader understanding of how literature and arts serve as vital tools for resisting trauma, reclaiming identity, and creating change in both personal and political contexts.

Item

Performance Efficient Methods to Handle Zero Day Attacks in Cloud Environment

(Avinashilingam, 2026-01) Swathy Akshaya M; Guide - Dr.G.Padmavathi

A zero-day attack is a type of cyber-attack that exploits a previously unknown vulnerability in a software application or a system. These attacks are called "zero-day" because the software developer has zero days to address the vulnerability since it has just been discovered. The term zero-day attack refers to a single or specific instance of an attack exploiting a newly discovered vulnerability, whereas zero-day attacks refers to multiple instances of such attacks or the general concept of attacks exploiting zero-day vulnerabilities. Zero-day attacks have become a significant cybersecurity concern as attackers continuously exploit previously unknown vulnerabilities, causing severe consequences such as data breaches, financial losses, and reputational damage. Although vulnerability identification can be performed using techniques such as code analysis, vulnerability scanning, penetration testing, and threat intelligence, traditional detection approaches remain largely dependent on signature-based and reactive mechanisms, making them ineffective against unknown and evolving threats. These limitations are further amplified in dynamic cloud environments, where increased system complexity, shared infrastructure, and large-scale network traffic expand the attack surface and hinder accurate real-time monitoring. Existing methods also suffer from high false positive rates, limited adaptability, inadequate behavioral learning, and increased computational overhead, reducing their effectiveness in large-scale deployments. To address these challenges, this research proposes a proactive, intelligent, and adaptive multi-phase framework for predicting, identifying, and detecting zero-day attacks. The proposed methodology integrates Machine Learning (ML), Deep Learning (DL), probabilistic modeling, game theory, and optimization techniques to enable behavior-driven predictive security. By moving beyond signature-based detection and learning attacker behaviors and system vulnerabilities, the framework enhances detection accuracy, minimizes false alarms, and provides scalable real-time protection suitable for dynamic cloud computing environments. The proposed research outlines a four-phase methodology for predicting zero-day attacks by combining machine learning and deep learning techniques. The first phase employs an Enhanced BPNN with CloudSim simulator to map the attack paths. In subsequent phases, data is systematically preprocessed, feature selection is performed, and multiple ML (Machine Learning) / DL (Deep Learning) models are trained and evaluated, followed by a comparative analysis of results across approaches. The objective is to proactively suggest performance efficient methods to identify and predict previously unseen (zero-day) threats, enabling organizations to detect attacks before they manifest. This approach proves to enhance the accuracy and reliability of intrusion detection systems by leveraging both traditional machine learning and advanced deep neural network methods, ultimately improving protection of digital assets through early, learned prediction of emerging threats. The research is conducted using a combination of simulation and real-world data to evaluate the proposed models performance. The simulation is performed using CloudSim, synthetic datasets Dataset 1 (Path Dataset) and Dataset 2 (Attack Dataset) is obtained from actual zero- day attack. The results of the comparative analysis will provide insights into the effectiveness of the proposed methodology and its potential for practical applications in cyber security. The core contribution of the research for Phase 1 is dedicated to identify the potential paths through which a zero-day attack can infiltrate the system. This phase traces the attack path in the cloud environment using a Probabilistic Graph Approach combined with an Enhanced Back Propagation Neural Network (BPNN), supported by Improved Decision Trees (DT) and Weighted K-Means clustering. Outcomes include improved accuracy by 3.01%, precision by 2.63%, recall by 2.27%, and F1 score by 2.34% compared to Bayesian, Scalable Bayesian, and Probabilistic Bayesian models. Phase 2 employs a hybrid method of Game Theory based on the Nash Equilibrium and an Adaptive Gaming Model based on a Modified Bi-Directional Long Short-Term Memory (LSTM) network to predict zero-day attacks. It enhances the ability to predict the potential zero day threats within the system through enhanced accuracy prediction, system performance and improved communication between security components achieving up to 11.4% higher accuracy, 11.3% higher precision, 5.5% higher recall, and 5.4% higher F1-score compared to DT, SVM, GNB, and Logistic Regression. Phase 3 propose a Residual Network integrated Deep Convolutional Zero-Day Adversarial Safety Network, designed for live zero-day attacks. It uses deep learning over the cloud communication network traffic analyzing and detecting events given in Dataset 1 (Path Dataset) and Dataset 2 (Attack Dataset) for real-time anomalies. Advanced convolutional layers and residual connections enhance generalization and classification up to 12.4% higher accuracy, 10.2% higher precision, 9.1% higher recall, and 9.2% higher F1- score over HMM, and up to 9.7% higher accuracy and 10.8% higher recall and ensure faster learning to identify zero-day attacks effectively compared to Bayesian GNN, BERT, TransVAE, and HADE-SVM models. Phase 4 presents an OLFFOA-optimized hybrid deep learning model. This comparative analysis allows for evaluating different models to determine the most efficient and accurate approach for zero-day attack prediction. The model achieves up to 98.1% accuracy, 95.2% precision, 94.4% recall, and 94.8% F1-score on Dataset 2, outperforming ResNet50, CNN-LSTM, and Bi-LSTM individually by 1.9% – 3.1%. The results show improved recognition rates, fewer false positives, and minimized time complexity, thus making the prediction models credible and scalable. Overall, this research work recommends performance efficient methods to handle zero-day attacks in cloud environment. The results are evident when tested with simulated dataset and benchmark dataset in zero-day attack. The anticipated research impact contributes to the development in enabling proactive, intelligent, and adaptive cloud security mechanisms, while the practical implication of this work supports real-world deployment in cloud infrastructures for early threat prediction, resilient intrusion detection, and enhanced security reliability in dynamic and large-scale environments.