International Journal of Innovations in Science & Technology
Not a member yet
813 research outputs found
Sort by
A GIS-Based Comparative Analysis of Ground Water Quality in Administrative Towns of Lahore City (2014–2024)
Get PDFAmidst insufficiency of water reserves, groundwater plays a crucial role in meeting freshwater requirements for households, parks, and horticulture in Lahore city. However, concerns regarding groundwater quality and its associated risk to human health and the environment have intensified due to factors such as rapid urbanization, industrial growth, and over extraction. Despite different monitoring efforts, the regional variability and uncertainty in groundwater quality necessitate more sophisticated assessment approaches to support optimal decision-making. The purpose of this study is to perform a comparative analysis between traditional Groundwater Quality Index (GQI) evaluation methods and entropy-weighted models. Additionally, it aims to analyze groundwater quality in tubewells across administrative towns in Lahore by utilizing Geographic Information System (GIS) and advanced geospatial techniques. The study incorporates groundwater quality data such as pH, Total Dissolved Solids (TDS), Electrical Conductivity (EC), and heavy metal concentrations into a GIS-based spatial analytic framework. By taking uncertainty in water quality classification into account, both traditional and entropy information theory offer a more adaptable and practical evaluation of groundwater suitability than the other GQI frameworks. The findings of this study show that groundwater quality varies significantly around Lahore, with certain regions showing contamination levels above acceptable bounds. High-risk areas are identified by the study, where water quality metrics point to possible health issues and highlight the necessity of focused measures
Development of XAI-Driven Churn Prediction Framework for Proactive Retention in Telecom
Get PDFIn the telecommunications industry, customer churn is a major problem that has a big influence on profitability and competitiveness. Current systems mostly rely on reactive strategies that are unable to prevent churn proactively. We present an intelligent, explainable AI-driven system, TeleChurnAI, which predicts churn and pinpoints its root causes. The novelty of this research lies in its integration of explainable AI and customer segmentation, providing useful information for retention strategies. To develop TeleChurnAI, a machine learning model, CatBoost, is employed to accurately predict churn, and SHAP (Shapley Additive exPlanations) is utilized to interpret model results as well as to support predictions. The prediction accuracy and interpretability of the model are assessed after it is trained on the historical telecom customer dataset, “Telco Customer Churn”. We found that TeleChurnAI offers transparency through visual explanations of churn risk factors and significantly increases the accuracy of churn predictions. Through an interactive dashboard made for CRM professionals with different levels of technical expertise, using the Shiny framework in Python, the system also divides up the customer base according to demographic and behavioral trends, allowing for targeted retention actions. In addition to helping with early intervention, this dual capability lowers marketing expenses and boosts customer loyalty. In conclusion, TeleChurnAI provides a thorough and user-centered method for telecom churn management. In the future, we aim to integrate sentiment analysis and real-time prediction in our proposed system
A Hybrid Transformer and CNN-Based Approach for Classifying Mental Health Disorders from Social Media Data
Get PDFMental health disorders are a significant global concern, with increasing prevalence on social media platforms where individuals often share their experiences and emotions. This research presents a novel approach for classifying mental health disorders, specifically depression, anxiety, borderline personality disorder (BPD), and post-traumatic stress disorder (PTSD), using social media text. We propose a hybrid architecture that combines domain-specific transformer models, such as PsychBERT and MetaBERT, with Convolutional Neural Networks (CNNs) to enhance the model’s ability to understand mental health-related language and metaphorical expressions. The transformer models, pretrained on mental health and symbolic data, generate embeddings that capture the unique linguistic features in social media posts. These embeddings are processed through cascaded CNN layers to extract deep features, which are then concatenated and classified into mental illness categories. The model was evaluated using a balanced dataset comprising 40,000 social media posts, achieving an overall accuracy of 96% and an F1-score of 0.96. The proposed model outperforms existing state-of-the-art methods, including fine-tuned BERT and RoBERTa models, demonstrating superior performance in accurately classifying mental health disorders. The results highlight the effectiveness of leveraging domain-specific language models and CNNs for enhanced classification of mental health conditions in social media text. This study underscores the potential of advanced deep learning techniques in addressing mental health issues and facilitating early detection in real-world applications
Assessing Urban Expansion and Land Cover Change in City District Lahore using Multi-Stage Satellite Data: Lahore land use land cover
No full textLahore, a metropolis and 2nd second-largest city of Pakistan, has been experiencing rapid urban expansion over the past five decades. The socio-economic development and growth of the urban population have caused the rapid increase of urban expansion. The increase in the built-up area of Lahore has seen remarkable growth during the past decades. This study is aimed at detecting the Spatio-temporal changes in land use and land cover and evaluating the urban expansion of Lahore since 2003. The conversion of land to other uses is primarily because of growth in urban population, whereas the increase in economic activities is the central reason for the land-use changes. In this study, temporal Landsat imageries were used. The supervised image classification of the maximum likelihood algorithm was applied on Landsat ETM+ (2003) and OLI/TIRs (2023) images, whereas a post-classification comparison technique was employed to detect changes over time. The spatial and temporal analysis revealed that during the past twenty decades, the built-up area of Lahore city has expanded by 486 km². It was found from the analysis that in Lahore city, the urban expansion was primarily at the cost of loss of fertile agricultural land, vegetation, and other cultivable land use. The analysis further revealed that The Total agricultural area in 2003 was 725 KM². The agricultural land to Built-up area is about 325 km². Due to the population increasing, the newly added population needs more space to fulfill their basic needs. The Total Barren Land in 2003 was 7452 KM².The Barren land to Built-up color, which is about 255 km².Rapid Land use changes have been marked for a period of 20 years in Lahore. The increase in the area used for built-up land is 470 Km² in 2003 to 956 Km² in 2023 (overall increase is 28 %), respectively
Land Degradation Risk Assessment in District Dir, Pakistan
Get PDFSoil erosion is a global concern, influenced by terrain, vegetation, soil, and climate factors. Traditionally, field-based techniques have been utilized for the measurement of soil erosion. In the present study, Remote Sensing and Geographic Information System (RS/GIS) techniques are used for soil erosion estimation. The Revised Universal Soil Loss Equation (RUSLE) is frequently utilized, incorporating various elements such as soil erodibility, rainfall erosivity, slope steepness, Land Use and Land Cover (LULC), and conservation practices. This study focuses on the Dir district in Pakistan, integrating the RUSLE model with RS and GIS to identify soil erosion-prone areas. The goal is to implement targeted interventions and sustainable land management practices to mitigate soil erosion in these areas. The output of the RUSLE model identifies key zones that need to be addressed to prevent further land degradation. This study also indicates higher C-factor values in Upper and Lower Dir, ranging from 0.001 to 0.2. Soil loss was calculated using all factors (R, K, LS, CP), showing that soil loss is approximately 31.6 tons/ha/yr in Upper Dir and 22.88 tons/ha/yr in Lower Dir, which is higher in Upper Dir due to high elevation (>30m) and more rainfall in Upper Dir (1275mm). Furthermore, annual rainfall values ranging from 508 mm to 1275 mm were noted, resulting in maximum rainfall erosivity values of 572.87 MJ mm ha/h/year in Upper Dir and 568.16 MJ mm ha/h/year in Lower Dir. Thus, this study provides critical data for society and policymakers to implement targeted soil conservation measures and sustainable land management systems, thereby mitigating soil erosion and preventing further land degradation in the district of Dir
Smog, Stress, and Society
Get PDFSmog, an escalating byproduct of climate change and rapid urbanization, poses a significant risk to public health and resilience, especially in megacities like Lahore. This study examines the psychological, biological, and social impacts of smog on young adults, situating these findings within the broader discourse on climate-induced hazards and disaster risk reduction (DRR). Employing a mixed-method approach, the research involved focus group discussions, scale development with expert validation, and large-scale public perception analysis using digital text analytics (Voyant Tools) on responses from 800 participants. Results revealed eleven major themes ranging from anxiety, mood disturbances, and impaired concentration to respiratory ailments, skin conditions, and reduced social interactions, capturing the multidimensional burden of smog exposure. Traffic emissions, loss of green spaces, and industrial pollution emerged as the most cited contributors, reflecting gaps in urban environmental management. The study’s Bio-Psychosocial (BPS) Scale (CVI = 0.93) offers a validated tool for assessing smog’s multifaceted impacts, enabling targeted policy and anticipatory interventions. These insights highlight the need to integrate air quality monitoring, green infrastructure, and public awareness into climate change adaptation and DRR strategies to safeguard both health and societal cohesion in vulnerable urban populations
Barriers to Maternal Health Care Accessibility and Its Causal Determinants in Faisalabad, Pakistan: A Geospatial Assessment
Get PDFAccess to maternal healthcare is a critical determinant of maternal and neonatal health outcomes, yet it remains a neglected issue in many developing regions, including Pakistan. This study investigates the spatial distribution and accessibility of maternal healthcare facilities in Faisalabad District using Geographic Information System (GIS) techniques, including point distance analysis and multiple ring buffer analysis. A total of 230 female respondents from six tehsils were surveyed using a structured questionnaire, with data gathered from two major public hospitals in the district. There are 155 Basic Health Units (BHUs) operating within Faisalabad District. 45.2% of respondents reported good accessibility to maternal health facilities, while 43% reported bad and 11.7% worse access. The highest concentration of cases (78 respondents) was within 6–10 km of a health facility. The findings reveal significant spatial disparities in access to healthcare facilities, particularly in rural and peripheral areas such as Tandlianwala, Chak Jhumra, and Samundri. Many women reside more than 20 kilometers from the nearest well-equipped hospital, and road conditions—ranging from poor to non-existent—further limit physical accessibility. The multiple ring buffer analysis demonstrates that several remote settlements fall outside the 12-mile buffer, indicating critical service gaps. Additionally, economic constraints and lack of public transportation exacerbate the situation, limiting women\u27s ability to seek timely antenatal care. The overcrowding of urban public hospitals and the high cost and low quality of private healthcare further restrict options for low-income groups. The study underscores the urgent need for decentralized planning, infrastructure improvement, and equitable distribution of maternal health services, especially in underserved rural regions
Assessment of Soil Fertility in Jhelum, Punjab, Pakistan, using Geospatial Technologies
Get PDFSoil fertility is a key factor influencing agricultural productivity and sustainability. This study evaluates the spatial distribution of essential soil chemical properties-pH, electrical conductivity (EC), available phosphorus (P), available potassium (K), organic matter (OM), and saturation percentage in Tehsil Jhelum, Pakistan. A total of 160 topsoil samples (0–15 cm depth) were collected using random sampling, with GPS coordinates recorded for each site. Laboratory analysis was conducted to assess the nutrient status of the soils, and Ordinary Kriging interpolation was used within a GIS framework to generate spatial distribution maps. The findings revealed notable variability across the region. Soil pH ranged from 4.3 to 7.8 (mean: 7.44), indicating mostly neutral to slightly alkaline conditions. EC values ranged from 0.49 to 1.40 dS/m, suggesting low to moderate salinity. Available phosphorus varied between 1.2 and 7.8 mg/kg, while available potassium ranged from 60 to 180 mg/kg, showing moderate fertility levels. Organic matter content was uniformly low (0.20–0.66%), with a mean of 0.42%, highlighting poor organic inputs. Saturation levels varied from 22% to 72%, displaying a layered spatial pattern. The spatial heterogeneity observed in soil nutrients underscores the need for site-specific nutrient management and precision agriculture practices. The generated maps serve as valuable tools for farmers, agronomists, and policymakers to make informed decisions aimed at improving crop productivity and maintaining soil health in the region
A Hybrid Machine-Learning Framework for Intrusion Detection: Comparative Evaluation and Statistical Validation
Get PDFThe increasing sophistication and frequency of cyberattacks have intensified the need for Intrusion Detection Systems (IDS) that are both accurate and adaptive. Traditional IDS, whether signature-based or anomaly-based, provides foundational protection but faces well-documented limitations: signature-based systems struggle against zero-day exploits, while anomaly-based systems often produce high false positive rates. To address these challenges, researchers and practitioners are increasingly turning to Machine Learning (ML) as a means of enhancing IDS capabilities. This paper explores the integration of ML techniques—supervised, unsupervised, and deep learning—into IDS frameworks and evaluates their effectiveness using widely recognized datasets, including NSL-KDD and CICIDS2017. Supervised learning methods such as Random Forest and Support Vector Machines (SVM) demonstrate strong classification abilities, while unsupervised clustering approaches offer promise in identifying novel attacks. Deep learning models, particularly Recurrent Neural Networks (RNNs), show state-of-the-art performance in capturing sequential traffic patterns and detecting subtle anomalies. In addition to model comparisons, this study emphasizes the practical relevance of ML-enhanced IDS by examining its integration with established tools like Snort and Zeek. Our results highlight that ML-driven IDS consistently outperforms traditional approaches, with RNNs and Random Forest achieving the highest balance of accuracy and efficiency. The findings underscore the potential of ML-based IDS to serve as the next frontier in cybersecurity, offering improved detection accuracy, reduced false alarms, and adaptability to evolving threats. At the same time, challenges remain in terms of dataset representativeness, computational demands, and the interpretability of deep learning models. By situating the analysis within both academic research and real-world deployment contexts, this paper contributes to a clearer understanding of the opportunities and trade-offs in advancing IDS through machine learning
IoT-Enabled Smart Agriculture: Architectures, Applications, and Future Directions
Get PDFIntroduction/Importance of Study: The integration of Internet of Things (IoT) technologies into agriculture has become essential to tackle challenges of food security, climate change, and resource optimization.
Novelty statement: This study introduces the novelty of a unified, low-cost, and modular IoT framework that addresses gaps in scalability, interoperability, and affordability, particularly in developing agricultural regions.
Material and Method: A systematic literature review was conducted across IEEE Xplore, SpringerLink, ScienceDirect, MDPI, and Google Scholar, focusing on twelve peer-reviewed studies published between 2019 and 2024. Comparative thematic analysis was applied to examine IoT architectures, communication protocols, and practical implementations.
Result and Discussion: Findings highlight that IoT systems commonly adopt a three-layer architecture (perception, network, and application), with LoRa, Zigbee, and fog computing models offering reliable rural connectivity. Reported outcomes include 30–40% water savings through smart irrigation, 15–20% yield increases with IoT-based monitoring, and up to 16% energy efficiency improvements in optimized wireless sensor networks. Despite these advances, challenges remain in cost, interoperability, farmer training, and security mechanisms. Current frameworks also lack adaptability across diverse farming contexts, limiting scalability and long-term sustainability.
Concluding Remarks: IoT-enabled agriculture offers significant potential to enhance sustainability and productivity, but future research must prioritize modular platforms, lightweight AI integration, energy harvesting, and context-specific deployment strategies