The global transition toward eco-friendly transportation has positioned the reliability of Electric Vehicle (EV) propulsion systems as a key industrial concern. Sudden motor or battery malfunctions during high-speed transit present severe safety hazards. Traditional maintenance models, predominantly reactive or rigidly scheduled, often fail to address incipient faults effectively, resulting in avoidable downtime and high ownership costs.
This systematic review analyzes 16 high-impact research papers and technical reports to evaluate the transition from traditional condition monitoring to intelligent maintenance frameworks powered by Agentic AI and the Internet of Things (IoT). Existing reviews often lack coverage of recent developments (2024–2026) in Agentic Multi-Agent Systems (MAS) integrated with full-stack environments such as the MERN stack for autonomous fleet management.
Our analysis identifies a modular agentic architecture—balancing real-time edge diagnostics using ESP32 microcontrollers and cloud-based Prognostics and Health Management (PHM)—as the most viable solution for modern EV reliability. Results indicate that AI-driven models achieve fault classification accuracies of 85%–95%, potentially reducing unplanned outages by 35%.

The increasing adoption of Internet of Things (IoT) technologies in healthcare has enabled continuous monitoring of patients’ physiological conditions through interconnected sensors and medical devices. However, conventional cloud-based healthcare IoT systems face challenges relating to latency, communication overhead, energy consumption, data privacy, device heterogeneity and Quality of Service (QoS). This study developed an IoT optimization system in healthcare using Federated Learning (FL) to support decentralized and privacy-preserving processing of healthcare data. The system was designed around local data preprocessing and model training at IoT devices, device selection based on available resources, and global model aggregation using the Federated Averaging (FedAvg) algorithm. A Hybrid Agile–Incremental development approach was adopted. The implementation used Python, Django, SQLite and machine-learning libraries identified in the thesis. Simulated healthcare sensor records containing physiological and device-related parameters were used for testing, and the model was trained with an 80:20 training–testing split. Evaluation of the binary global model, using a confusion matrix reported for 1,000 records, produced an accuracy of 93%, precision of 70%, recall of 78%, F1-score of 74%, false-positive rate of 5%, false-negative rate of 22% and negative predictive value of 96%. Functional tests also showed successful device connection, sensor-data handling, local training, model-update transmission, aggregation, authentication checks and dashboard access. The comparative assessment of the centralized and proposed federated models was qualitative: the thesis did not report direct numerical measurements for latency, energy consumption, communication overhead or QoS. The study demonstrates the implementation of a healthcare IoT monitoring framework that keeps raw patient data at local devices while coordinating model updates through federated learning.

Before a software release goes live, someone has to check a lot of things: which versions of Java, Spring Boot, the database and the frontend framework it uses, whether those versions work together, whether any of the libraries have known security holes, and whether the tests actually passed. Doing this by hand is slow, and things get missed. In this work we built ReleaseGuard AI, a release approval system on ServiceNow that does the first round of these checks automatically. A developer raises a release request and either types in the component details or uploads the pom.xml and package.json files. A Spring Boot service reads those files and pulls out the versions. The system then compares them with the organization’s version rules, checks compatibility, looks for known vulnerabilities (CVEs) and reviews the test and build evidence. From this it gives the release a risk level and one of three recommendations: Approved, Needs Review or Blocked. The recommendation goes into a ServiceNow approval workflow, where a release manager makes the final call and every step is logged. We tried the prototype on simulated release records and sample Java and React configurations. A release that used a Log4j version affected by CVE-2021-44228 was blocked automatically, and a low-risk release with a readiness score of 95 was marked eligible for approval. The system is meant to help reviewers, not replace them. We have not yet measured how it compares with manual checking, and we leave that for future work.

This study concentrated on Agriculture shows as one of the business promotion tools used in sales promotion in Chipata District, Eastern Province Zambia. The researcher commenced this work by taking a quick look at the definitions, evaluation and background information of agriculture show use of several data collection techniques which include: Questionnaire, personal interviews, observations. The analysis of this data collected revealed certain prospects and problems shows in the researcher’s presentation and analysis of data. On that note, the researcher made the following findings. Agriculture shows were organized by Agriculture show board under ministry of Agriculture and other private sectors. Therefore, call for improvement. There is a general awareness about Agriculture show as very vital in the revitalization of the nation’s economy. Agriculture show exhibitions have frequently organized that people are calling for its or regulation and control. Based on the above findings, the research made the following major recommendations participants in Agriculture shows must have clearly defined objectives for participation in Agriculture show. Scientific and technical ideas must be about Agriculture in our Shows rather than monetary transactions. Emphasis should be made on purchase rather than immediate sales. Finally, a body be established by government to control the frequency at which shows are organized in Chipata

The National Education Policy (NEP) 2020 represents a major reorientation of Indian higher education from predominantly discipline-centred and examination-oriented models toward flexible, multidisciplinary, experiential and outcome-oriented learning. Zoology is particularly well placed to benefit from this transition because its knowledge base intersects with biodiversity, ecology, fisheries, wildlife management, biotechnology, environmental health, animal resources and community-based conservation. This paper examines the implications of NEP 2020 for undergraduate Zoology education in India through a qualitative policy and curriculum analysis. The analysis considers the National Education Policy 2020, University Grants Commission (UGC) frameworks on undergraduate curricula, learning outcomes, internships, research development, environmental education, Indian Knowledge Systems and multidisciplinary higher education, together with relevant literature on active learning and science education. The paper identifies six interconnected areas of transformation: flexible curriculum architecture, outcome-based learning, practical and field-based education, undergraduate research, digital and interdisciplinary competence, and employability-oriented skill development. It argues that curriculum revision alone is insufficient. Effective implementation requires alignment among learning outcomes, pedagogy, laboratory and field infrastructure, assessment, faculty development, research opportunities and institutional partnerships. A NEP-Zoology Integrated Education Framework is proposed to connect disciplinary knowledge with practical competence, field investigation, research capability, environmental responsibility and career readiness. The paper concludes that NEP 2020 can strengthen undergraduate Zoology as an applied and research-oriented life science, provided institutions move beyond nominal syllabus restructuring and develop coherent systems for experiential learning, authentic assessment and locally relevant biological research.

प्रस्तुत शोध पत्र भारतीय समाज में LGBTQ+ (लेस्बियन, गे, बाइसेक्सुअल, ट्रांसजेंडर, क्वीर एवं अन्य) समुदाय की ऐतिहासिक स्थिति, सामाजिक-कानूनी चुनौतियों, मानवाधिकारों और उनके मुख्यधारा में समावेशन की दिशा का विशद अध्ययन प्रस्तुत करता है। भारतीय संस्कृति और पौराणिक आख्यानों में लैंगिक तथा यौनिक विविधता के प्रमाण मिलते हैं, किंतु औपनिवेशिक काल के दमनकारी कानूनों विशेष रूप से भारतीय दंड संहिता की धारा 377 ने इस समुदाय को समाज के अंतिम पायदान पर धकेल दिया।
यह शोध पत्र सर्वोच्च न्यायालय के ऐतिहासिक निर्णयों (जैसे NALSA निर्णय 2014, नवतेज सिंह जौहर मामला 2018, तथा सुप्रीमो चक्रवर्ती मामला 2023) का गुणात्मक विश्लेषण करते हुए यह उजागर करता है कि केवल वि-अपराधीकरण (Decriminalization) ही पूर्ण समानता की गारंटी नहीं है। आज भी यह समुदाय पारिवारिक तिरस्कार, शिक्षा एवं रोजगार में भेदभाव, स्वास्थ्य सेवाओं तक सीमित पहुँच, मानसिक तनाव और विवाह एवं गोद लेने जैसे नागरिक अधिकारों के अभाव से जूझ रहा है। प्रस्तुत अध्ययन प्राथमिक व द्वितीयक स्रोतों, समाजशास्त्रीय सिद्धांतों तथा केस स्टडीज़ पर आधारित है। शोध के निष्कर्ष यह दर्शाते हैं कि एक समावेशी भारत के निर्माण के लिए विधायी सुधारों के साथ-साथ शैक्षणिक पाठ्यचर्या में बदलाव, संवेदीकरण (Sensitization) और संस्थागत सहयोग अत्यंत आवश्यक हैं।

This study examined the relationship between training and development and employee retention in selected beverage firms in South-South Nigeria. Specifically, the study assessed the relationship between off-the-job training and employee retention, as well as the relationship between on-the-job training and employee retention. A cross-sectional survey research design was adopted. The study covered a population of 211 employees, from which a sample size of 136 was determined using the Krejcie and Morgan formula. A structured questionnaire was used for data collection, and 130 valid responses were obtained and analysed. Descriptive statistics and multiple regression analysis were employed to analyse the data at a 0.05 level of significance. The findings revealed that off-the-job training had a positive and statistically significant relationship with employee retention (B = 1.153, t = 9.908, p < 0.001). Conversely, on-the-job training had a negative but statistically significant relationship with employee retention (B = -0.301, t = -3.035, p = 0.003). The overall regression model was statistically significant (F = 135.242, p < 0.001), with an R-square value of 0.680, indicating that the two dimensions of training and development jointly explained 68.0% of the variation in employee retention. The study concluded that training and development are important considerations in employee retention, although their individual relationships with retention differ. It was therefore recommended that beverage firms should strengthen relevant off-the-job training programmes and review their existing on-the-job training practices to ensure that workplace learning adequately supports employees' professional development and retention.

This study examined the effect of expansion strategies on business survival of selected manufacturing companies in Uyo, Akwa Ibom State. The study specifically investigated the effect of market expansion and diversification expansion strategies. A survey research design was adopted for the study and data were collected from 205 staff of selected manufacturing companies using a structured questionnaire, out of which 193 were validly completed and returned, representing a response rate of 94%. The instrument was validated using face and content validity through the usage of appropriate questionnaires to the four (4) manufacturing companies in Uyo L.G.A, Akwa Ibom State and adequate coverage of dimension respectively while its reliability was confirmed using the regression analysis, which yielded an index above the acceptable threshold. Data collected were analyzed using descriptive statistics, while regression analysis was employed to test the hypotheses. The findings revealed that expansion strategies have a significant positive effect on business survival. Specifically, market expansion and diversification expansion strategies was found to have a stronger influence on business survival by enhancing revenue opportunities and reducing dependency on a single market. The study concludes that manufacturing firms that adopt effective expansion strategies are better positioned to sustain operations and remain competitive in a dynamic business environment. Based on the findings, the study recommends that manufacturing firms should strategically be expanded with market expansion and diversification expansion strategies.

Artificial Intelligence (AI) has significantly improved Clinical Decision Support Systems (CDSS) by enabling intelligent analysis of healthcare data for disease diagnosis and clinical decision-making. Earlier CDSS mainly relied on rule-based systems and structured healthcare records, which limited their adaptability and diagnostic capability. Recent advancements in deep learning, Natural Language Processing (NLP), and multimodal learning have enabled the integration of medical images, patient symptoms, clinical text, and structured healthcare data for more accurate and context-aware diagnosis. Multimodal AI systems combine complementary information from different healthcare modalities to improve prediction accuracy and clinical reasoning. This survey reviews recent developments in multimodal clinical diagnosis systems, including structured healthcare data systems, CNN-based medical imaging models, NLP-based symptom analysis systems, multimodal fusion approaches, foundation models, and generative AI techniques in healthcare.

Digital platforms enable small businesses to market products and coordinate sales without permanent retail premises. This study describes business practices, perceived livelihood benefits and support needs among 50 young entrepreneurs in Chipata District, Zambia. A descriptive cross-sectional questionnaire design was used, and aggregate responses were analysed using frequencies and percentages. Clothing and fashion accounted for 60% of businesses. Facebook was the main platform for 40% of respondents, while WhatsApp and TikTok each accounted for 30%. Eighty per cent identified online business as their main income source, and all respondents reported generating income. Customer reach, flexibility and skills development were also perceived benefits. High internet and data costs were the main challenge for 50%, followed by low customer trust for 30% and competition for 20%. Sixty per cent reported fraud or dishonest transactions, while 70% had received no formal training. Affordable connectivity was the most requested support, followed by business management and digital marketing training. These findings describe the participating entrepreneurs' perceptions; they do not establish profitability, business survival or a national growth trend. Support should combine affordable connectivity with practical business skills, transaction verification and delivery planning. Larger studies with documented recruitment procedures and financial records are needed to evaluate sustainability.

Robotics is an important area of modern science and technology. Robots are used to perform tasks that may be repetitive, physically demanding, precise, or unsafe for people. Their use is expanding across manufacturing, healthcare, agriculture, education, transport, and domestic settings. This paper discusses the meaning of robotics, its major applications, and its possible benefits and challenges for society. It uses a brief literature-based review of published research and institutional material rather than new experimental data. The discussion indicates that robots can improve productivity, consistency, and support for people, but they also raise concerns about employment transitions, safety, privacy, fairness, and accountability. The paper concludes that robotics is most beneficial when human needs, responsible design, appropriate regulation, and education are considered alongside technical progress.

The increasing application of digital technologies in public services has created opportunities to modernize electoral processes through remote voting. However, the effective use of voter-related data to identify patterns in remote voting systems remains an important area for further investigation. This study developed an enhanced remote voting system for smart cities and incorporated K-Means clustering to analyze voter-related characteristics. A system-development approach was adopted, using a simulated dataset of 500 voters. Age, voting duration, and login attempts were selected as clustering variables. K-Means clustering using Euclidean distance was configured with four clusters and a maximum of 100 iterations, converging after eight iterations. The resulting clusters contained 60 (12.0%), 84 (16.8%), 186 (37.2%), and 170 (34.0%) voter records, respectively, with differences in their mean characteristics. System functionality evaluation recorded an Authentication Accuracy (AA) of 98%, Vote Recording Accuracy (VRA) of 83.7%, and Voting Success Rate (VSR) of 83.7%. The findings demonstrate the feasibility of integrating K-Means clustering into a remote voting platform to identify patterns in voter-related system data. The developed system combines remote voting functionality with an analytical component that may support system evaluation and improvement.

Machine L:earning (ML) is an important branch of Artificial Intelligence that enables computers to learn from data and make predictions or decisions without being explicitly programmed for every task. In recent years, machine learning has become increasingly important in the healthcare sector. It is being used for disease prediction, medical image analysis, patient monitoring, drug discovery, and personalized treatment. ML algorithms can analyze large amounts of medical data and identify patterns that may be difficult to detect manually. However, the use of ML in healthcare also creates challenges related to data privacy, accuracy, bias, and interpretability. This paper discusses the major applications, benefits, challenges, and future scope of machine learning in healthcare.

The Medical Internet of Things (MIoT) has revolutionized healthcare by enabling real-time health monitoring through wearable devices like smartwatches. The biometric data are collected by these devices such as heart rate and blood oxygen saturation for preventive care and long-term management. This work creates a model with a Neural Network classifier that predicts health risks in real-time and is based on biometric data obtained through smartwatches. The improvement in the reliability of MIoT systems for clinical applications shall be achieved by tackling issues such as battery life, sensor noise, and data privacy. Our work has shown promising accuracy for disease prediction and hence personalization for healthcare solutions.

Personal safety applications commonly require users to initiate a panic action, which may be difficult during an emergency. This paper describes the documented implementation of Abhaya, an Android-first application developed in Flutter and Dart that combines a manual SOS interface, locally stored trusted contacts, device-based location retrieval, an audible siren, and a microphone-driven distress detection pipeline. Audio is sampled at 16 kHz and passed to an on-device TensorFlow Lite convolutional classifier through a detector abstraction. A stub-first architecture separates testing of the alert workflow from model development. The proposed machine-learning experiment compares conventional training with low-pass filtering, additive noise, and amplitude reduction to approximate muffled audio. The project report records 25 passing functional/integration cases and lists muffled-distress recall values of 72.4% and 89.6% for baseline and augmented models; however, the corresponding experiment artifacts and evaluation protocol are absent, so these values are treated as unverified report claims rather than validated research results. The paper delineates implemented functions, planned extensions, and reproducibility requirements.

The prime spectrum of a commutative ring carries a natural topology in which closed sets are determined by ideals. The irreducible components of this space are controlled by minimal prime ideals, but finite irreducible decomposition is not automatic when the ring is not Noetherian. This article gives a self-contained account of the relationship between finite irreducible decomposition of a closed subset of a prime spectrum and finiteness of the minimal primes over its defining ideal. The central criterion is stated and proved using the correspondence between irreducible closed subsets and prime ideals. An explicit example in a polynomial ring in countably many variables exhibits infinitely many distinct minimal primes over a single ideal and hence a closed subset with infinitely many irreducible components. The discussion also distinguishes topological decomposition from primary decomposition and explains why quasi-compactness alone does not force a finite irreducible decomposition. The results are standard consequences of foundational commutative algebra; the purpose of the article is to assemble the criterion, proof, and non-Noetherian example into a coherent structural treatment, rather than to claim a new general theorem.

Most biographies follow a life through time. This one follows a life through an equation. Satish Gajawada, an Integrated Dual Degree (B.Tech + M.Tech) graduate of the Indian Institute of Technology Roorkee (2012), has spent fourteen years as an independent researcher who gives names, structure and heart to ideas in artificial intelligence: Excellent AI (EAI), Very Highly Advanced AI (VHAAI), Out of the Box AI (OBAI), a society (ISVHAAI), a family of value-themed swarm algorithms, thirteen theories of human purpose, a 108-right taxonomy of human rights, and a lifelong “thesis” supervised by his parents and GOD. Reading the particle swarm update rule as a metaphor for a research life, this article identifies four forces (inertia, personal best, global best and exploration) and one outcome: a position update that, in 2026, turned decisively toward testing and open disclosure. Drawing on eight source documents, with each statement attributed to a published work, a manuscript or the subject’s own account, we show that a reproducible genetic-algorithm and PSO experiment found a fairness weight that cut a demographic-parity gap from 0.121 to 0.016 for 1.9 points of accuracy, and a gentler weight that improved accuracy and fairness together. Itemised output rose from one entry in 2012 to twelve in 2026. We conclude that Gajawada is a builder of human-centred vocabulary and institutions, a grateful son, a joyful experimenter and a candid author, and that his life offers every learner a portable update rule: keep your direction, honour your best, listen to your guides, stay curious, and keep moving.

Alcohol-associated liver disease (ALD) is a major global health concern encompassing hepatic steatosis, alcohol-associated hepatitis, fibrosis, cirrhosis and liver failure. Chronic alcohol consumption contributes to hepatic injury through complex biochemical, metabolic, inflammatory and oxidative mechanisms. Ethanol metabolism involves alcohol dehydrogenase, aldehyde dehydrogenase, cytochrome P450 2E1 and non-oxidative pathways. These processes generate acetaldehyde and promote oxidative stress, mitochondrial dysfunction, lipid accumulation and inflammatory responses, accelerating hepatocellular damage and disease progression. Clinical manifestations range from asymptomatic steatosis to jaundice, systemic inflammation and advanced hepatic dysfunction. Diagnosis integrates alcohol consumption history, clinical assessment, biochemical investigations, imaging and non-invasive fibrosis evaluation, while liver biopsy is reserved for selected cases requiring diagnostic clarification. Management emphasizes sustained alcohol abstinence, nutritional rehabilitation, adequate protein intake and appropriate treatment of complications, including infections and hepatic decompensation. Selected patients with severe alcohol-associated hepatitis may require pharmacological treatment and multidisciplinary evaluation. Early detection, accurate risk stratification and individualized therapeutic interventions are essential to improving clinical outcomes.

This study aimed to develop, validate, and evaluate a Canva-based digitalized reading module to enhance the reading comprehension skills of Grade 6 learners in terms of literal, inferential, and critical comprehension. It employed a Research and Development (R&D) design anchored on the ASSURE Model, which guided the development, implementation, and evaluation of the instructional material. The learners’ reading comprehension was assessed using the Philippine Informal Reading Inventory (Phil-IRI) Group Screening Test (GST) through pretest and posttest scores. The developed module featured interactive reading texts, multimedia presentations, comprehension activities, formative assessments, and synthesis exercises designed to promote learner engagement and understanding. Expert validators assessed the module’s content, instructional design, functionality, and overall acceptability, and their feedback served as the basis for revisions and improvements. Findings revealed that Grade 6 learners initially obtained a mean score of 10.67 (53.35%), indicating a frustration level in reading comprehension. Following the implementation of the digitalized reading module, the mean score increased to 14.45 (72.28%), demonstrating improved reading performance. Improvements were observed across literal, inferential, and critical comprehension, although critical comprehension remained an area requiring further instructional support. Statistical analysis further revealed a significant difference between the learners’ pretest and posttest scores, indicating improvement in their reading comprehension performance after the intervention. The study concluded that the Canva-based digitalized reading module is an effective supplementary instructional material for strengthening Grade 6 learners’ reading comprehension skills. It recommends the integration of interactive digital reading resources into school literacy programs, alongside differentiated reading strategies, and encourages future researchers to explore similar interventions involving larger populations, longer implementation periods, and other grade levels.

Artificial intelligence is changing the nature of digital evidence. Images, videos, voices, documents and other forms of synthetic media can now be generated or altered with a degree of realism that makes traditional visual inspection increasingly unreliable. For Indian criminal justice, this development is significant because the evidentiary framework has already moved decisively toward electronic and digital records through the Bharatiya Sakshya Adhiniyam, 2023 (BSA), while the Bharatiya Nagarik Suraksha Sanhita, 2023 (BNSS) strengthens technology-assisted investigation and trial procedures. The central legal problem is no longer simply whether a digital record can be produced before a court, but whether its origin, integrity, authenticity and reliability can be demonstrated when artificial intelligence may have created or modified it. This article examines the existing Indian framework, the evidentiary difficulties created by deepfakes and other AI-generated material, the relevance of chain of custody and expert examination, and the constitutional implications for fair trial rights. It argues that Indian courts should adopt a technology-neutral but authenticity-sensitive approach: AI-generated material should not be automatically excluded, but neither should apparent realism be treated as proof of truth. The article proposes procedural safeguards including provenance documentation, forensic verification, disclosure of relevant metadata, expert standards, judicial training, and clear rules for challenging synthetic evidence. The future of criminal trials will therefore depend not on resisting artificial intelligence, but on developing reliable methods for distinguishing trustworthy digital evidence from persuasive digital fabrication.

The contemporary international system is marked by overlapping conflicts, strategic rivalry, economic fragmentation and weakening confidence in multilateral institutions. The Russia–Ukraine war, continuing instability in West Asia, intensifying United States–China competition and contestation in the Indo-Pacific have created a complex environment for India. As a rising power with major economic, demographic, military and diplomatic capabilities, India cannot remain detached from these developments. At the same time, India’s interests do not fit neatly into any single geopolitical camp. Its relationships with Russia, the United States, Europe, Japan, the Gulf states, Iran and the Global South require a flexible foreign policy that protects strategic autonomy while expanding international cooperation. This paper examines India’s strategic position through the concepts of strategic autonomy, multi-alignment, national interest, international law and preventive diplomacy. It argues that India’s approach should not be understood as passive neutrality. Rather, it represents issue-based engagement designed to preserve decision-making space, protect energy and economic security, strengthen defence preparedness and contribute to conflict resolution. India’s role in the G20, BRICS, SCO, Quad and United Nations provides multiple platforms for such engagement. The paper concludes that India can make its greatest contribution to global peace by combining credible national power with principled diplomacy, support for the UN Charter, humanitarian engagement, economic resilience and dialogue among competing powers.

The separation of powers seeks to prevent the concentration of governmental authority by distributing legislative, executive and judicial functions among different institutions. India and the United Kingdom share a parliamentary form of government in which the executive is drawn from the legislature, but they differ in the legal foundations and enforcement of institutional limits. India follows constitutional supremacy: the Constitution binds every organ of the State, and courts may invalidate laws that violate constitutional limits. The United Kingdom follows parliamentary sovereignty, under which courts generally cannot invalidate an Act of Parliament. This short doctrinal and comparative study examines the constitutional framework, judicial independence, judicial review and legislative accountability in both countries. It argues that India provides stronger legal protection against unconstitutional action, while the United Kingdom relies more heavily on statutes, conventions and political accountability. Both systems, however, face risks of executive dominance and institutional overreach. The article concludes that India can benefit from more transparent judicial appointments and stronger legislative scrutiny without adopting parliamentary sovereignty.

This study examined the impact of Artificial Intelligence (AI) in enhancing library services at the Federal University of Education, Zaria. The study was guided by four objectives, which were to examine the extent of AI adoption in library services, assess the perceived impact of AI on service delivery, identify the challenges hindering AI adoption and suggest strategies for improving AI integration. A quantitative survey research design was adopted. The population comprised 150 professional librarians and ICT staff of the University Library, from which a sample of 108 respondents was determined using the Krejcie and Morgan sample size determination table. Stratified random sampling was employed to ensure representation of professional librarians and ICT staff. Data were collected using a structured questionnaire developed in line with the study objectives. The instrument adopted a five-point Likert scale ranging from Strongly Agree to Strongly Disagree. A pilot study involving 20 librarians and ICT staff was proposed at the Federal University of Education, Adeyemi, to assess the clarity and reliability of the instrument. The data were intended to be analysed using descriptive statistics, including frequency counts, percentages, mean scores and standard deviations, with SPSS Version 26. The study focused on AI applications such as plagiarism detection, automated cataloguing and classification, chatbots, information retrieval and digital preservation. It also examined potential barriers, including inadequate ICT infrastructure, insufficient funding, limited staff competencies and the absence of clear institutional policies. The study proposed that improved infrastructure, adequate funding, continuous staff training, ethical guidelines and collaboration with technology providers could support effective AI integration. The study is expected to provide insights that can guide the university library in adopting AI technologies to improve service efficiency, accessibility and responsiveness to users’ information needs.

This study examined the impact of reference services and medical mobile applications on information utilization among medical students in Federal Teaching Hospital Libraries in North-West Nigeria. The study was guided by three research questions and three null hypotheses and adopted Wilson’s Information Behaviour Model (1999) as its theoretical framework. A descriptive correlational survey research design was employed. The population comprised 642 clinical medical students from five selected institutions, from which a sample of 246 respondents was drawn using a multistage sampling technique. Data were collected using a structured questionnaire validated by experts and tested for reliability, yielding a Cronbach’s Alpha coefficient of 0.910. Data were analysed using descriptive statistics, simple linear regression and multiple linear regression at the 0.05 level of significance. The findings revealed that reference services had a high perceived impact on information utilization, with a grand mean of 2.59. Assistance with library catalogues, indexes and bibliographies recorded the highest mean of 2.65. Medical mobile applications also had a high perceived impact, with a grand mean of 2.91. Their use for examination and presentation preparation recorded the highest mean of 3.00. The provisional multiple regression results indicated that reference services and medical mobile applications jointly explained 33.1% of the variation in information utilization ((R^2 = 0.331)), with the model statistically significant ((F(2,237) = 58.640, p < 0.001)). The study recommends strengthening reference services, promoting reliable medical mobile applications and improving information-literacy training to support effective information utilization among medical students.

Amitav Ghosh’s The Hungry Tide (2004) is an important work of contemporary Indian English literature that brings together ecology, human relationships, displacement, cultural memory, environmental ethics and the struggle for survival. Set primarily in the Sundarbans, the novel presents a geographical space where land and water are constantly changing and where human life is deeply connected with tides, rivers, forests and animals. Rather than presenting nature as a passive background, Ghosh represents the natural environment as an active force that influences human decisions, relationships, identities and conflicts. The novel also questions conventional approaches to environmental conservation by examining the tension between the protection of wildlife and the survival needs of marginalized communities. Through characters such as Piya Roy, Fokir, Kanai Dutt, Kusum and Nirmal, Ghosh presents different ways of understanding and experiencing the environment. Scientific knowledge, local ecological knowledge, political ideology and personal experience often come into conflict, but the novel also demonstrates that these different forms of knowledge can complement one another. The history of Marichjhapi further reveals how environmental policies can become connected with political power, displacement and social injustice. This paper examines The Hungry Tide through ecocriticism, postcolonial ecocriticism and environmental humanities.

In Nigeria&#39;s public sector, fraud continues to be a significant obstacle to accountability and efficient financial management, necessitating the development of more potent fraud detection techniques. The effect of forensic accounting methods on fraud detection in Nigeria&#39;s public sector was investigated in this study. In particular, it evaluated the impact of computer-assisted auditing tools, data mining analytics, digital forensics, and litigation support expertise. Accountants, auditors, internal control officers, forensic accountants, and other pertinent staff in Federal Ministries, Departments, and Agencies (MDAs) in Ondo State, Nigeria, were given structured questionnaires to complete online as part of the study&#39;s survey research design. 146 stakeholders made up the population, and purposive sampling and a census were employed. Using SPSS version 25, descriptive statistics and multinomial logistic regression were used to analyze the data. The results showed that data mining analytics had a positive but insignificant effect on fraud detection (B = 0.177, p = 0.300), whereas digital forensics had a negative but insignificant effect (B = -0.099, p = 0.574). Positive and substantial impacts were observed using computer-assisted auditing tools (B = 0.422, p

Direct digital sales can provide farmers with an additional route to market, but online transactions depend on credible seller information, practical delivery arrangements, and usable pricing support. This paper describes the design and implementation of FarmDirect, a web-based marketplace for agricultural products that integrates these functions within a role-based application. Farmers can register, submit certificates, list produce, manage stock and orders, and communicate with customers. Consumers can browse listings, inspect the verification status of a seller, place orders, and follow delivery updates. A central element is a certificate-review workflow that checks for reused certificate identifiers, compares submitted records with farmer-profile details, and allows an administrator to examine supporting evidence, including official QR-linked information where available. The interface distinguishes approved and pending reviews through visible badges. The application also includes reference-price analysis, map-assisted farmer discovery, UPI-oriented payment support, messaging, and a delivery-partner dashboard. The implementation uses React, Node.js, Express, SQLite, Socket.IO, and Leaflet/OpenStreetMap. The present assessment documents the implemented functions and their design implications; controlled user trials and quantitative measurements of price accuracy, verification performance, or economic impact have not yet been reported. FarmDirect therefore provides an integrated prototype and a basis for future field evaluation of trust-aware farmer-to-consumer commerce.

With the increasing demand for faster and more efficient retail experiences, conventional billing systems continue to pose challenges such as long queues and delays, particularly for customers purchasing a small number of items. This paper proposes a novel RFID-based smart shopping solution specifically designed for daily buyers and quick shoppers who require rapid check-in and checkout processes. The system introduces a compact, clip-on device, similar in concept to electronic toll collection systems such as FASTag, which can be attached to a customer’s shopping bag. The proposed device integrates an RFID reader to automatically detect products added to or removed from the bag, maintaining a real-time virtual cart through seamless synchronization with a mobile application. Each product is equipped with a unique RFID tag to ensure accurate identification and tracking. Unlike traditional smart cart systems aimed at bulk or monthly shopping, this solution is optimized for low-item, high-frequency purchases, enabling minimal interaction and significantly reduced transaction time. The system employs geofencing to detect store entry and exit points, facilitating automatic billing upon exit. The total purchase amount is deducted directly from the user’s digital wallet, eliminating the need for manual checkout. Digital receipts are generated and shared with both customers and retailers. Additionally, a dual-application framework is implemented, comprising a customer interface for expense tracking and wallet management, and a retailer interface for monitoring sales and transaction analytics. To ensure security, an anti-theft mechanism is incorporated wherein unpaid or unregistered items retain active RFID status and are detected at the exit gate, triggering an alert system. The proposed solution aims to deliver a frictionless, fast, and secure shopping experience tailored for modern consumers seeking convenience in everyday retail transactions.

Judicial review is a mechanism through which courts examine whether legislative and executive action conforms to constitutional requirements. It protects constitutional supremacy, limits arbitrary public power and supports the enforcement of rights. Yet its constitutional foundations and institutional operation differ across legal systems. This article compares judicial review in India and the United States, focusing on constitutional foundations, judicial doctrine, protection of rights and democratic accountability. In India, Articles 13, 32 and 226 of the Constitution provide important textual foundations for constitutional review. The Supreme Court has also developed the basic structure doctrine, limiting Parliament’s power to amend the Constitution in ways that damage or destroy its essential features. In the United States, judicial review developed principally through constitutional interpretation, most famously in Marbury v Madison (1803). The article argues that both systems treat constitutional supremacy as a limit on ordinary governmental power, but India’s express remedial provisions and basic structure doctrine create a distinctive framework. It further argues that judicial review is most legitimate when courts give reasoned judgments, apply constitutional standards consistently and remain attentive to institutional competence. The comparison demonstrates that judicial review is neither simply an anti-democratic intervention nor an unlimited judicial mandate; it is a constitutional responsibility that must be exercised with independence and restraint.

Business intelligence dashboards help organizations make sense of sales, customer activity, inventory, and other areas of performance. However, a dashboard can present a precise-looking number even when the data behind it contain duplicate transactions, missing values, inconsistent labels, or outdated records. When these problems are not visible, decision-makers may rely on indicators that do not fully represent the situation they are trying to understand. This conceptual paper proposes Counterfactual Data-Quality Impact Auditing (CDQIA), an approach for examining how a key performance indicator (KPI) may change when a documented data-quality treatment is applied. The proposed framework connects six activities: identifying a possible data issue, tracing it to affected KPIs, recording the treatment rule, recalculating the KPI under that rule, interpreting the size and significance of the change, and communicating the result to dashboard users. The framework also considers interactive filters, since a data issue may affect one dashboard view differently from another. CDQIA does not assume that a corrected value is automatically the truth; its usefulness depends on the quality of the evidence and the transparency of the assumptions. The paper develops conceptual propositions and identifies questions for future empirical research. As a theoretical study, it does not report experiments or claim measured improvements. Its main contribution is a decision-oriented way to make the possible effects of data-quality problems easier to examine and discuss.

Auditor independence is fundamental to the credibility of financial reporting, yet the auditor’s commercial relationship with the client creates recurring risks to objectivity. This critical narrative review examines how professional ethics prevents, identifies and addresses conflicts of interest in financial-statement auditing, with particular attention to the Zambian context. It integrates the International Ethics Standards Board for Accountants’ 2025 Code, the International Auditing and Assurance Standards Board’s quality-management framework, Zambia’s Accountants Act, 2008, the Companies Act, 2017, and the Companies (Amendment) Act, 2025. It also draws on established academic literature and carefully delimited lessons from Enron and Wirecard. The analysis distinguishes conflicts of interest from the broader concept of independence and explains why disclosure, consent and review are not universal remedies: some relationships or services are prohibited and some residual threats necessitate declining or ending an engagement. The article proposes an auditable decision pathway integrating conflict screening, threat evaluation, documented responses, independent quality review and oversight by those charged with governance. Its original contribution is to connect general ethical principles to Zambia’s express statutory conflict rule and updated key-auditor tenure requirements. Because the paper is a narrative review, it does not report primary survey results or estimate the prevalence of ethical breaches in Zambia. The central conclusion is that auditor ethics becomes effective when independence is operationalised through enforceable decisions, documented evidence and institutional accountability.

Artificial Intelligence (AI) has transformed the gaming industry from simple rule-based systems to complex, intelligent, and adaptive experiences. This research paper explores the role of AI in modern video games. Initially, AI in games was limited to predefined patterns, like the ghosts in Pac-Man (1980). Today, it powers Non-Player Characters (NPCs), procedural content generation, dynamic difficulty adjustment, and realistic graphics. The paper discusses key AI techniques such as A_ Pathfinding, Finite State Machines (FSM), Behavior Trees, Machine Learning (ML), and Reinforcement Learning (RL). Case studies of games like The Last of Us Part II, Minecraft, and AlphaGo are analyzed. The paper also highlights challenges such as high computational cost, development complexity, and balancing difficulty. Finally, it discusses the future scope of Generative AI, where AI can create entire games, stories, and characters. The study concludes that AI is not just a tool but the core of next-generation gaming.

Artificial Intelligence (AI) is increasingly influencing the teaching–learning process by introducing innovative approaches to lesson planning, instructional delivery, student assessment, personalized learning and educational administration. The successful integration of AI into school education depends not only on the availability of technological resources but also on teachers' willingness, knowledge, confidence and preparedness to use emerging digital technologies. Digital readiness is therefore an essential prerequisite for the effective adoption of AI in educational settings. The present study aims to investigate artificial intelligence adoption and digital readiness among school teachers. It focuses on teachers' awareness of AI applications, perceived usefulness, ease of use, technological competence, digital confidence and preparedness to integrate AI into classroom practices. The study adopts a descriptive survey method with a proposed sample of 300 school teachers selected through a suitable sampling technique. Data may be collected using an Artificial Intelligence Adoption Scale and a Digital Readiness Scale. Percentage analysis, mean, standard deviation, t-test are proposed for analyzing the data. Findings are provided to demonstrate the presentation of statistical results. The study emphasizes the importance of strengthening teachers' digital competencies, AI literacy, professional development and ethical awareness to promote responsible and meaningful AI integration in school education.

Background: GLP-1-based therapies such as semaglutide and tirzepatide produce substantial weight loss, but reduced appetite and gastrointestinal adverse effects may compromise dietary adequacy, lean mass, and long-term weight maintenance. This narrative review examines how nutrition care can prevent sarcopenia, malnutrition, and weight regain during GLP-1-based obesity treatment.
Methods: Literature published from January 2018 to May 2026 was searched in PubMed, Scopus, and Web of Science. Trials, observational studies, systematic reviews, clinical guidance, and mechanistic studies were synthesized thematically without meta-analysis. Findings were interpreted contextually.
Results: Pivotal trials reported mean weight reductions of 14.9% with semaglutide 2.4 mg and up to 20.9% with tirzepatide, while body-composition substudies showed reductions in absolute lean mass; approximately one quarter of weight lost with tirzepatide was lean mass. Key risks include sarcopenia, inadequate protein and micronutrient intake, dehydration, and gastrointestinal intolerance. Reduced intake may create an energy gap and place iron, vitamin B12, folate, vitamin D, and thiamine at risk. Withdrawal studies show substantial weight regain after discontinuation. The proposed framework combines baseline nutritional screening, GLIM-based diagnosis of malnutrition, individualized energy and protein prescriptions, a nutrient-dense dietary pattern, hydration and micronutrient monitoring, resistance exercise, and behavioral support, adapted for older adults, type 2 diabetes, bariatric surgery, pregnancy, adolescents, and Middle Eastern and North African settings, including Ramadan.
Conclusion: Nutrition care is not adjunctive but essential to GLP-1-based obesity treatment. Prospective trials of nutrition interventions in treated patients are still needed. Dietitian-led assessment, monitoring, and maintenance support can help protect muscle, nutritional status, and long-term outcomes.

Disease may arise from multiple interacting factors rather than a single isolated cause. Homoeopathy has historically emphasized susceptibility, constitution, heredity, and miasmatic predisposition to explain individual variations in disease expression. To examine the limitations of a single-cause model and explore the conceptual relevance of biopathography in understanding disease causation, susceptibility, and individual disease expression. Biopathography reconstructs the patient's life and disease history, considering significant experiences, environmental exposures, lifestyle, relationships, and factors that may trigger or maintain illness. It provides a longitudinal study for organising antecedent circumstances and understanding individual responses. However, temporal association does not establish causation, and homoeopathic susceptibility and miasmatic predisposition should not be equated with genetically established mechanisms. Biopathography may support individualized case analysis by integrating life history with disease development. Its application requires critical interpretation and a clear distinction between homoeopathic theoretical concepts and contemporary scientific evidence.

Azadirachta indica A. Juss. (neem) is a tree belonging to the Meliaceae family. It has been traditionally used in South Asian countries for centuries as medicine. It has a number of bioactive compounds, such as limonoids, flavonoids, and phenolic compounds in the leaves, bark, seeds, and flowers and its oil. Laboratory and animal research has reported antimicrobial, anti-inflammatory, antidiabetic, anticancer, antioxidant, and insecticidal activity. There is limited clinical data from humans. Neem is relatively safe at the traditional dose, but neem oil can be poisonous, particularly to infants. The major compounds of neem, pharmacological activities, safety concerns, and research gaps are summarized in this review.

The proposal for ‘One Nation, One Election’ (ONOE) seeks to synchronise elections to the Lok Sabha and State Legislative Assemblies, with a longer-term proposal for synchronisation of local-body elections. Supporters argue that a common electoral cycle can reduce public expenditure, repeated deployment of administrative machinery, disruption caused by the Model Code of Conduct, and the policy uncertainty associated with frequent elections. Critics, however, contend that simultaneous elections may weaken the political distinctiveness of States, encourage nationalised voting behaviour, constrain the timing of democratic accountability, and create constitutional difficulties when a House loses its majority before the end of its term. This paper examines ONOE as an electoral reform through the constitutional principles of parliamentary government, federalism, democratic accountability, free and fair elections, and the basic structure doctrine. It traces the historical existence of simultaneous elections, analyses the Constitution (One Hundred and Twenty-Ninth Amendment) Bill, 2024 and the Union Territories Laws (Amendment) Bill, 2024, and evaluates arguments concerning cost, governance, voter choice, federalism and constitutional stability. It argues that synchronisation is not inherently incompatible with federalism, but its legitimacy depends on preserving the autonomy and accountability of State governments. A constitutionally durable model should therefore combine phased synchronisation, safeguards for premature dissolution, strengthened Election Commission capacity, transparent political-finance rules, and mechanisms ensuring that a State legislature is not effectively subordinated to the national electoral calendar.

The rapid expansion of digital services has transformed personal data into an essential economic and social resource. In India, this transformation has also exposed individuals to surveillance, profiling, identity theft, data breaches and opaque forms of automated decision-making. The constitutional recognition of privacy in Justice K.S. Puttaswamy (Retd.) v. Union of India marked a decisive shift by treating privacy as an intrinsic part of dignity, liberty and autonomy under Article 21 and the broader scheme of fundamental rights. The Digital Personal Data Protection Act, 2023 (DPDP Act) represents India's principal legislative framework for digital personal data and establishes obligations for Data Fiduciaries, rights and duties of Data Principals, special protections for children, regulation of Significant Data Fiduciaries and an institutional mechanism through the Data Protection Board of India. The Digital Personal Data Protection Rules, 2025 provide the operational framework and introduce a phased commencement structure. This article examines the constitutional and statutory foundations of data protection in India, analyses the principal provisions of the DPDP Act, considers important judicial decisions, and identifies continuing concerns relating to state exemptions, consent, enforcement, cross-border transfers, surveillance, children's data and effective remedies. It argues that Indian privacy law must balance innovation and legitimate public interests with meaningful individual control, proportionality, transparency and institutional accountability.