International Journal of Innovations in Science & Technology
Not a member yet
813 research outputs found
Sort by
Stock Market Analysis and Prediction Using Deep Learning
Get PDFThe stock market is a complex system influenced by various factors, including economic indicators, geopolitical events, and investor sentiments. Traditional methods of stock market analysis often rely on statistical models and technical indicators, which may struggle to capture the intricate patterns and non-linear relationships present in financial data. This paper is about an innovative application which is designed to fill the gap between traditional stock market analysis and cutting-edge predictive modeling. The paper not only addresses the challenges associated with fragmented data and delayed analysis but also opens avenues for continuous monitoring and optimization of predictive models in response to dynamic market conditions. These models are seamlessly integrated into the application developed in the Analysis Phase, providing users with real-time predictions and valuable insights. Many machines learning (ML) and deep learning (DL) techniques have demonstrated to perform well in stock price prediction by prior research, and most people regard DL techniques them as one of the most accurate prediction methods, particularly when used for longer prediction ranges. In this research, after performing pre-processing steps like data normalization, we have employed an LSTM and GRU based models. Through training and testing, we determined the ideal settings for the optimizer, dropout, batch size, epochs, and other parameters. The outcome of comparing the LSTM network model with GRU we concluded that LSTM it is not suitable for short-term forecasting, and performs well for long-term forecasting whereas GRU performs well in both cases
Comparative Assessment of Classification Algorithms for Land Cover Mapping Using Multispectral and PCA Images of Landsat
Get PDFThe advancement of remote sensing technologies and the availability of free satellite data have significantly enhanced the precision of land use and land cover (LULC) mapping, facilitating the analysis of landscape transformations and ecosystem changes. However, selecting the most suitable classifier for LULC mapping remains a complex challenge. Therefore, it is essential to evaluate the accuracy of various LULC modeling algorithms to determine their effectiveness in different applications. This study conducted a comprehensive evaluation of both supervised machine learning algorithms and traditional classification methods applied to Landsat 8 imagery with a 30-meter spatial resolution, covering the Shangla and Battagram districts in Khyber Pakhtunkhwa (KPK), Pakistan. The study focused on three classification algorithms: Maximum Likelihood Classification (MLC), Support Vector Machines (SVM), and Random Forest (RF). The performance of these algorithms was assessed on both multispectral images and composite images derived from Principal Component Analysis (PCA) and Band Ratioing and/or Normalized Indices. Additionally, the accuracy of these algorithms, when applied to different datasets, was compared with the recently released World Cover LULC product by the European Space Agency (ESA). The results indicated that the SVM algorithm outperformed the others, achieving an overall accuracy of 90.43% and a kappa coefficient of 0.8792. The MLC and RF algorithms also produced promising results, with overall accuracies of 85.58% and 88.46%, respectively. Furthermore, the study found that the overall accuracy of ESA’s World Cover LULC product was 70.67% in the study area, based on similar validation samples. These findings underscore the strengths and limitations of each algorithm, providing valuable insights into their suitability for LULC classification and the applicability of existing global LULC maps
Resilience Assessment of Urban Areas in Peshawar, Pakistan, in Response to Climate Change Impacts
Get PDFIntroduction of the Study: The Urban areas in Peshawar, Pakistan, are increasingly at risk from climate change impacts, necessitating a thorough resilience assessment to mitigate these risks and enhance adaptation strategies.
Novelty Statement: This study introduces a novel framework specifically tailored to Peshawar’s context, focusing on urban resilience—a topic that has not been extensively explored before.
Material and Method: A mixed-method approach was employed, including household surveys, focus group discussions, stakeholder interviews, and the analysis of secondary data from satellite imagery and government reports.
Result and Discussion: The major findings reveal significant vulnerabilities in infrastructure, diverse community perceptions of climate risks, and varying effectiveness of current adaptation measures. The capacity for emergency response demonstrated by local institutions underscores a critical need for capacity building. The study highlights both strengths and weaknesses in urban resilience, emphasizing the importance of institutional support and community engagement.
Concluding Remarks: The study provides targeted recommendations to enhance resilience-building efforts in Peshawar, aiming to improve the city’s ability to withstand and adapt to climate change impacts
Assessment of Rooftop Potential for Solar Energy and Rainwater Harvesting in Islamabad: A Geospatial Approach towards Sustainable Urban Development
Get PDFIntroduction/Importance of Study: Water and energy crises due to abnormal temperatures, precipitation patterns and urban growth leading imbalance in sustainable process so this study explores the potential of underused area of urban settings and their potential in sustainable urban development.
Novelty Statement: This study integrates advance geospatial techniques to access rooftop characteristics and calculate the potential of rooftops for solar energy and rainwater harvesting.
Material and Method: Firstly, the total area of all rooftops is estimated by utilizing freely accessible Open Buildings V3 Polygons data with a resolution of 50 cm. Rooftop current uses and characteristics were retrieved from openly accessible remote sensing data and by implementing Geographic Information System (GIS) methodologies. To estimate the potential of renewable electricity Global Solar Atlas data is used, and the potential for rainwater harvesting is calculated using precipitation data by the Pakistan Meteorological Department.
Result and Discussion: The results show that rooftop area of Islamabad has the potential to harvest 778.92 million gallons/annual of rainwater and about 16,504.29 MWh per year, which can address the capital city\u27s energy and water demands. The sectors of I8, I9, H9, H13, I14, blue area, Rawat industrial area, and DHA have demonstrated significant potential for solar energy and rainwater harvesting. Taking into account the findings of the current research and public feedback, we can propose recommendations for future energy policies, new society planning, sustainable use of rooftop space and Islamabad can lead the way towards a more sustainable future.
Concluding Remarks: The successful implementation of proposed systems can lead to a reduction in reliance on non-renewable energy sources and might reduce urban flooding risk in region
Assessment of Groundwater Potential Zones Using Electrical Resistivity in Muzaffargarh
Get PDFThe study integrates Earth observation and geospatial data to evaluate groundwater potential and conditions in Muzaffargarh, South Punjab, Pakistan, a region grappling with freshwater scarcity due to high sediment concentrations in subsurface water. The developed approach aims to enhance sustainable water resource management in areas affected by such sediment challenges. An electrical resistivity survey was conducted at 40 locations within the study area, incorporating Vertical Electrical Sounding (VES) and spatial analysis with hydrogeological parameters to analyze and visualize the spatial distribution of freshwater. A weighted overlay analysis was employed to map freshwater and saline water zones, supported by 2D resistivity maps. The study generated several thematic layers, including data on geology, rainfall, lineaments, land use/land cover (LULC), drainage density, soil type, and slope. A groundwater potential (GWP) zone map was created, categorizing the area into four zones: very good, good, moderate, and poor. Additionally, resistivity maps were produced at depths of 2m, 10m, 50m, 80m, 200m, and 300m to analyze resistivity variations in the Ghazi Ghat and Qasba Gujrat areas of Muzaffargarh district. The study\u27s findings include curves indicating potential groundwater zones and a comprehensive understanding of subsurface characteristics through resistivity curve comparisons. These results provide valuable insights for the sustainable management of groundwater resources in the region, particularly in addressing freshwater scarcity
Exploring the Impact of Land Cover Changes on Genesis of Smog in District Lahore, Pakistan
Get PDFIntroduction: Smog is a major global issue, severely impacting Pakistan, particularly Lahore. This problem arises from a mix of natural factors and human activities, notably rapid urbanization, which has intensified fog into smog, affecting human health. In Lahore, urbanization has altered land use patterns, contributing to the urban heat island effect and elevated temperatures. Changes in land cover (LC), combined with pollution sources like industrial emissions and vehicle exhaust, play a significant role in smog formation.
Novelty Statement: This study highlights the long-term impact of LC changes on smog from 2002 to 2022. Water indirectly influences smog through meteorological conditions, while particulate matter (PM) from various sources poses health risks. The primary objective is to investigate how changes in land cover contribute to smog formation.
Material and Methods: ArcGIS was used to process data on land cover images, temperature, and air pollutants (NO₂, SO₂, and CO) within a controlled Geographic Information System (GIS) environment.
Results and Discussion: Land cover images of Lahore from different years were obtained using Google Earth Pro. ArcGIS was employed to analyze temperature data, and the inverse distance weighted (IDW) interpolation technique was used to visualize temperature variations and air pollutant concentrations over time. LC data for 2002, 2012, and 2022 were integrated into ArcGIS to demonstrate how land cover changes contribute to smog formation in Lahore.
Conclusion: The research highlights the need for effective management of urbanization and environmental challenges to address smog-related issues
A Framework of Software as a Service Using a Crowdsourcing Approach: A Case Study of Smart Classroom
Get PDFIntroduction/Importance of Study: Crowdsourcing can be effectively utilized to identify factors and develop modules by creating a platform where individuals contribute their ideas and suggestions. This research investigates the application of crowdsourcing-based cloud resources managed on a global scale, bringing together diverse skills to handle workloads on cloud platforms. Despite inherent challenges such as quality control due to the varied locations of contractors, and communication issues including language barriers, differing time zones, and security concerns, crowdsourcing provides a robust framework. It enables software developers to access a vast talent pool and deliver services more quickly and efficiently.
Novelty Statement: The crowdsourcing framework leverages the collective wisdom of diverse individuals to solve problems and generate ideas. In a smart classroom setting, this approach can be applied by setting clear objectives, engaging students through appropriate platforms, fostering collaboration, collecting data via surveys or discussions, analyzing results, and using insights to enhance learning experiences. By leveraging students\u27 contributions, educators can enhance collaboration, creativity, and engagement in the classroom, ultimately enriching the learning process for all participants.
Materials and Methods: This research is divided into three phases:
Identification Phase: Challenges are identified through a systematic literature review (SLR).
Implementation Phase: Identified factors are shortlisted to design a framework.
Validation Phase: The framework is validated using a smart classroom case study.
Results and Discussion: Our findings indicate that smart classrooms provide an opportunity to investigate how students adopt technology and innovation. Survey results show that both teachers and students believe smart classrooms enhance their knowledge and perceived ease of use, demonstrating the benefits of this educational approach.
Concluding Remarks: By exploring the case of the smart classroom, this research challenges existing pedagogical methods and introduces innovative ways to engage students through new technology acceptance perspectives. This study highlights the potential of crowdsourcing in creating more effective and interactive learning environments
Unlocking Potential: Personality-Aware TVET Course Recommendations Revolutionize Skill Development
Get PDFPersonality is a complex amalgamation of ideas, behaviors, and social constructs that shape our self-perception and influence our interactions with others. It tends to remain relatively stable over time. The development of personality-aware recommendation systems is driven by the understanding that human behavior and personality play a significant role in skill acquisition, career progression, and overall success. Technical and Vocational Education and Training (TVET) is crucial in building a skilled workforce, particularly in response to the demands of Industry 5.0. Unlike conventional recommendation systems, personality-aware systems effectively address persistent challenges such as the cold start problem and data sparsity. This paper introduces the Personality-aware TVET Course Recommender System (TCRS), which suggests the top three TVET courses by considering trainees\u27 personality traits, demographic information, and the historical success patterns of previous trainees in similar courses. A standout feature of the TCRS is its Academic System Learner, which continuously incorporates insights from individual trainees\u27 progress in TVET courses, thereby enhancing the accuracy of its machine learning model for predictive analysis. The effectiveness of the TCRS is assessed using seven classifiers, yielding notable prediction accuracies: 99% with Random Forest, 98% with Decision Tree, and 89% with k-Nearest Neighbors (kNN). In real-time testing, the TCRS demonstrated an accuracy rate of 84%
Thematic Analysis of Tourism Downfall and Economic Consequences During Covid-19: Evidence from A Rural Mountain Community in Pakistan
Get PDFThe COVID-19 pandemic had a profound impact on the tourism industry in Gilgit-Baltistan, leading to a significant reduction in tourist arrivals and severely affecting the local economy. This study demonstrated that the livelihoods of many residents, particularly those involved in tourism-related businesses, were adversely affected. The analysis revealed substantial monthly losses in tourist arrivals in 2020 compared to 2019, with a marked decline in the summer months, typically the peak season for tourism in the region. The thematic analysis highlighted the socio-economic challenges faced by the local population, including loss of income, business closures, and a decline in living standards. The study also emphasized the need for strategic interventions to support the tourism industry in Gilgit-Baltistan, including the development of policies to enhance resilience against future disruptions. The integration of advanced data analysis techniques and Geographic Information Systems (GIS) provided a comprehensive understanding of the pandemic\u27s impact, contributing valuable insights for policymakers and stakeholders in the tourism sector
A Large Language Model based Web Application for Contextual Document Conversation: A Large Language Model based Web Application for Contextual Document Conversation
No full textThe emergence of LLMs, such as ChatGPT, Gemini, and Claude has ushered in a new era of natural language processing, enabling rich textual interactions with computers. However, despite the capabilities of these new language models, they face significant challenges when queried on recent information or private data not included in the model’s dataset. Retrieval Augmented Generation (RAG) overcame the problems mentioned earlier by augmenting user queries with relevant context from a user-provided document(s), thus grounding the model’s response to inaccurate source material. In research, RAG enables users to engage interactively with their documents, instead of manually reading through their document(s). Users provide their document(s) to the system, which is then converted into vector indices, and used to inject contextual information into the user prompt during retrieval. The augmented prompt then enables the language model to contextually answer user queries. The research is composed of a web application, with an intuitive interface for interacting with the LIama 3.2 1B, an open-source LLM. Users can upload their document(s) and chat with the LLM in the context of their uploaded document(s)