International Journal of Innovations in Science & Technology
Not a member yet
813 research outputs found
Sort by
Modeling of Post-Myocardial Infarction and Its Solution Through Artificial Neural Network
Get PDFCardiovascular diseases, particularly myocardial infarction (MI) constitute a significant health concern globally. A myocardial infarction, which is commonly known as a heart attack, happens when a part of the heart muscle doesn’t get enough blood because of a blockage. Studying MI is complex and it requires looking at it from different angles. In recent years the fusion of mathematical modeling and artificial intelligence (AI) techniques has emerged as a promising avenue for understanding the complexities associated with MI. The primary goal of this study is to provide an AI-based solution for a new nonlinear mathematical model related to myocardial infarction phenomena. To obtain the solution we will use a well-known deep learning technique, known as artificial neural networks (ANNs) with the combination of the optimization technique Levenberg-Marquardt back propagation (LMB). This combined method is referred to as ANNs-LMB. The results obtained from the model using ANNs-LMB are compared with a reference dataset constructed through the adaptive MATLAB solver ode45. The numerical performance is validated through a reduction in mean square error (MSE). The MSE is around and the obtained results by ANNs-LMB almost overlapped with the reference dataset, which shows the accuracy and efficiency of the proposed methodology
On Evaluation of discrete RL agents for Traffic Scheduling and Trajectory Optimization of UAV-based IoT Network with multiple RIS unit
Get PDFUnmanned Aerial Vehicles (UAVs) have been very effective for data collection from widely spread Internet of Things Devices (IoTDs). However, in case of obstacles, the Line of Sight (LoS) link between the UAV and IoTDs will be blocked. To address this issue, the Reconfigurable Intelligent Surface (RIS) has been used, especially in urban areas, to extend communication beyond the obstacles, thus enabling efficient data transfer in situations where the LoS link does not exist. In this work, the goal is to jointly optimize the trajectory and minimize the energy consumption of UAVs on one hand and satisfy the data throughput requirement of each IoTD on the other hand. As it is a mixed integer non-convex problem, Reinforcement Learning (RL); a class of Machine Learning (ML), is used to solve it, which has proven to be computationally faster than the conventional techniques to solve such problems. In this paper, three discrete RL agents i.e. Double Deep Q Network (DDQN), Proximal Policy Optimization (PPO), and PPO with Recurrent Neural Network (PPOwRNN) are tested with multiple RISs to enhance the data transfer and trajectory optimization in an Internet of Things (IoT) network. The results show that DDQN with multiple RIS is more efficient in saving communication-related energy, while a single RIS system with the PPO agent provides more reduction in the UAV’s propulsion energy consumption when compared to other agents
Contrasting Impact of Start State on Performance of a Reinforcement Learning Recommender System
Get PDFA recommendation problem and RL problem are very similar, as both try to increase user satisfaction in a certain environment. Typical recommender systems mainly rely on history of the user to give future recommendations and doesn’t adapt well to current changing user demands. RL can be used to evolve with currently changing user demands by considering a reward function as feedback. In this paper, recommendation problem is modeled as an RL problem using a squared grid environment, with each grid cell representing a unique state generated by a biclustering algorithm Bibit. These biclusters are sorted according to their overlapping and then mapped to a squared grid. An RL agent then moves on this grid to obtain recommendations. However, the agent has to decide the most pertinent start state that can give best recommendations. To decide the start state of the agent, a contrasting impact of different start states on the performance of RL agent-based RSs is required. For this purpose, we applied seven different similarity measures to determine the start state of the RL agent. These similarity measures are diverse, attributed to the fact that some may not use rating values, some may only use rating values, or some may use global parameters like average rating value or standard deviation in rating values. Evaluation is performed on ML-100K and FilmTrust datasets under different environment settings. Results proved that careful selection of start state can greatly improve the performance of RL-based recommender systems
Integrating Multiple Datasets in Google Earth Engine for Advanced Hydrological Modeling Using the Soil Conservation Service Curve Number Method
Get PDFThis research explores the feasibility of using cloud computing and open data sources for hydrological modeling, specifically leveraging Google Earth Engine (GEE) and the Soil Conservation Service Curve Number (SCS CN) method to estimate runoff. The SCS CN approach is commonly applied in simulating rainfall-runoff processes and is effective for estimating water inflow into rivers, lakes, and streams. Google Earth Engine provides a range of functionalities, including algorithms for rapid data manipulation and visualization, and access to extensive global remote sensing and geographic information system (GIS) datasets. The study introduces an algorithm developed in GEE to analyze precipitation data and generate antecedent moisture condition (AMC) maps. This algorithm integrates MODIS land use/land cover (LULC) data with USDA soil texture data to classify hydrological soil groups. Runoff estimation utilizes three datasets: CHIRPS, GPM, and TRMM. A thorough analysis of the rainfall-runoff relationship in the Mangla watershed from 2005 to 2015 is conducted. The study quantifies runoff estimates from each dataset and performs comparative analysis to validate the accuracy and reliability of the hydrological modeling. Over the ten-year period (2005-2015), significant fluctuations in average rainfall and runoff levels are observed, with notable seasonal patterns. The highest average precipitation of 1412.194 mm occurred in 2015, resulting in an average runoff of 215.021 mm. Conversely, 2009 recorded the lowest average precipitation of 672.808 mm and an average runoff of 78.476 mm. The accuracy of the modeled runoff observations is validated using meteorological data from the Pakistan Meteorological Department (PMD), Water and Power Development Authority (WAPDA), and Climate Forecast System Reanalysis (CFSR). In 2008, 2009, and 2010, CHIRPS consistently demonstrated better accuracy compared to GPM and TRMM, with accuracies of 90%, 79%, and 86% respectively. Additionally, a sensitivity analysis of the SCS CN model parameters reveals the effects of initial abstraction and Curve Number values on runoff estimation. In conclusion, this research enhances the understanding of hydrological processes in monsoon-affected regions and offers valuable recommendations for implementing sustainable water resource management practices
Applications of RS & GIS for Tsunami and Sea Surges Risk Assessment Along the Coast of Karachi
Get PDFCoastal areas are vulnerable to various hazards, such as storm surges, inundation from sea-level rise or coastal flooding, tsunamis, and more. The situation becomes particularly disastrous if the coast is densely populated and highly developed. Pakistan has a coastline stretching 1,046 km, with Karachi being the most developed part. The Karachi coast faces frequent storms during the monsoon season and is also threatened by rising sea levels in the coming years. Additionally, Pakistan\u27s coastline is near the boundaries of two major tectonic plates—the Indo-Australian and Eurasian plates—as well as two minor plates, the Arabian and Iranian plates. In the event of a major earthquake in the Arabian Sea, a tsunami could pose a significant threat, potentially engulfing important and densely built-up commercial, residential, industrial, and sensitive military areas. This study aims to analyze potential losses due to inundation of the Karachi coast using Remote Sensing and GIS techniques
Advanced Blast Algorithm for Molecular Identification, Biodegradation and Decolorization of Synthetic Melanoidins Using Fungal Species Isolated from Soil and Spent Wash
Get PDFIntroduction/Importance of Study: Distillery spent wash contains a high organic load as Melanoidins. It is generated due to the Millard reaction, which produces sugar and amino acids, leading to extensive water and soil pollution. Anaerobic digestion removes 60-70% COD and color, so post treatment is required for degradation by using fungal species as biological process.
Objectives and Novelty statement for this study: The study aims to isolate and identify fungal species for the degradation of synthetic melanoidins from spent wash using a cost-effective, low-toxicity, and environmentally friendly fungal-based biological process.
Material and Method: Three mixed fungal culture inoculums (spent wash, wet, and dry soil) and seven isolated fungal strains were examined on solid media that degraded and decolorized melanoidins at controlled pH 5.5, 25oC, 160 rpm for 3-5 days.
Result and Discussion: The results showed that mixed culture of spent wash removed the highest COD 91.8 %, color removal was 75.7 %, F-S6 isolate identified as Penicillium showed maximum soluble COD removal was 96.7 %, and F-S5 isolate identified as Syncephalastrum showed a maximum color removal was 98.8 %.
Concluding Remarks: It was concluded that the microbial process using fungal species was successfully applied to enhance degradation and decolorization to remove melanoidins. Furthermore, Gompertz Modeling was done to check the fitting of the curve at 680 nm Optical Density (OD) analysis for seven fungal strains with the following five factors significantly estimating maximum specific growth rate µM, Asymptote A, coefficient of determination R2, lag time λ, and goodness of fit
Design and Analysis of Microstrip Patch Antenna Operating at Higher Order Mode
Get PDFThis paper proposes an enhanced bandwidth microstrip patch antenna by exciting it with higher order modes. Characteristics Mode Analysis (CMA) is used to analyze and understand the possible modes for bandwidth enhancement of microstrip patch antenna. Furthermore, Defected Ground Structure (DGS) technique is utilized for bandwidth enhancement. The proposed antenna is having a size of 67.5 × 67.5 mm2 with an operating frequency of 5.8 GHz. The impedance bandwidth is increased by 13.8% using Defected Ground Structure (DGS) by adding slots in the ground for higher order mode operation. Moreover, the proposed antenna has an overall efficiency of above 80. Therefore, enhanced impedance bandwidth, improved radiation pattern, and compatible design make the design novel and suitable for practical wireless applications
Deep Learning Based Multi Crop Disease Detection System
Get PDFThis research explores the integration of deep learning, computer vision, and edge computing to revolutionize crop disease detection. In response to the pressing need for prompt and accurate disease identification, this work leverages the capabilities of edge computing devices deployed within agricultural fields. Real-time data processing at the edge facilitates quick disease classification across various crops, enabling timely interventions. At the heart of the methodology lies a fine-tuned ResNet50 deep learning model, specifically chosen for its proficiency in handling complex visual data. Trained on a specialized dataset derived from the ImageNet database, the model exhibits promising accuracy rates in preliminary testing. Integrating edge computing into precision agriculture, this research presents a significant advancement toward sustainable agricultural practices. By empowering farmers with early detection and timely interventions, this endeavor equips agricultural communities with the knowledge and tools necessary to safeguard their crops, ensuring both food security and economic stability
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%
Assessment of Soil Erosion and Neotectonics Geomorphology of Bannu Basin using RS and GIS Techniques
Get PDFSoil erosion presents a significant environmental challenge in Bannu District, adversely impacting agricultural productivity and land sustainability. This research article offers a comprehensive approach to assessing and mitigating soil erosion risk in the region by utilizing the Revised Universal Soil Loss Equation (RUSLE) model in conjunction with hypsometric analysis. The study integrates various geospatial datasets, including mean annual rainfall, digital elevation models, soil maps, land use/land cover classifications, and satellite imagery. These datasets are essential for mapping the five key factors of the RUSLE model: Rainfall Erosivity (R), Soil Erodibility (K), Slope Length and Steepness (LS), Land Cover Management (C), and Support Practice (P). By mapping each factor individually and then integrating them, the study estimates soil erosion rates in Bannu District. Soil erosion risk is categorized into five levels, ranging from very low to excessive, to facilitate practical assessment. This classification assists in identifying areas that require immediate attention and intervention for sustainable land management and agricultural practices. The study highlights the benefits of combining Remote Sensing (RS) and Geographic Information System (GIS) technologies with the RUSLE model. This integration enables policymakers and land managers to evaluate and address soil erosion issues on a broader scale. Additionally, the study examines the role of hypsometry in understanding topography and erosion dynamics, incorporating topographic elements into the RUSLE model to explain soil erosion trends in Bannu District. Overall, this article provides a scientifically rigorous and practical soil erosion risk assessment for Bannu District. By leveraging the RUSLE model and GIS data with hypsometric analysis, the study offers valuable insights for addressing soil erosion and promoting sustainable land use in the region. The findings are intended to assist policymakers and stakeholders in safeguarding agricultural productivity and enhancing land sustainability in Bannu District