International Journal of Innovations in Science & Technology
Not a member yet
813 research outputs found
Sort by
Green Growth: AI-Driven Intelligent Farming for Effective Resource Management
Full text linkEffective fertilizer management plays a critical role in maximizing crop yield while reducing environmental harm and minimizing resource waste. This study presents an IoT-based intelligent fertilizer recommendation system designed to deliver accurate, real-time application guidance. The system integrates NPK sensors for soil nutrient detection, environmental sensors for humidity and temperature monitoring, and rain gauges to collect precipitation data. Data from the field is transmitted through an Arduino microcontroller to a cloud platform. A Random Forest classifier is used to determine the need for fertilization, while a CatBoost regressor estimates the required fertilizer quantity. The system was tested using real-time field data across 22 crop types, achieving 100% accuracy in classification and strong performance in regression tasks. Recommendations are automated and delivered via SMS to streamline field operations. The objective of this study is to develop an automated, sensor-driven fertilizer recommendation system using machine learning for precision agriculture. The novelty lies in the integration of real-time IoT sensing with hybrid AI models to optimize fertilizer use. This approach enhances productivity, reduces input waste, and supports environmentally sustainable farming
Testing Chatbot Systems using Agentic AI Approach
Full text linkAs large language models (LLMs) become increasingly integrated into real-world applications, robust and scalable evaluation methods are essential to ensure their reliability, safety, and effectiveness. This work introduces an innovative evaluation framework grounded in an agentic AI simulation approach, designed to overcome the limitations of traditional testing methodologies in newly developed chatbots. Unlike conventional methods that depend on static benchmarks or human evaluators, our approach employs autonomous AI agents capable of simulating a wide spectrum of user interactions. Within a controlled multi-agent environment, these evaluator agents interact with the target chatbot using natural language queries specifically designed to probe various functional capabilities, identify edge cases, and uncover potential failure modes. The agentic evaluation methodology systematically assesses the performance of chatbots in multiple dimensions, including task completion efficiency, contextual understanding in dynamic conversations, and adherence to safety and ethical guidelines. By incorporating recent advances in agentic metrics and automated scenario generation, our system produces detailed data-driven performance reports that capture both strengths and vulnerabilities in chatbot behavior. Preliminary results show that this approach not only reveals significantly more edge cases than conventional methods, but also reduces overall evaluation time by approximately 60-70 percent. This work contributes to a scalable, standardized testing paradigm that better aligns theoretical performance indicators with the practical challenges of deploying LLMs in real-world environments
AI-driven Early Autism Detection and Therapeutic Intervention System
Full text linkEarly identification of Autism Spectrum Disorder (ASD) is crucial for early intervention and improved outcomes. Low literacy and less exposure to computers in Pakistan’s rural areas restrict parents’ capacity to recognize ASD symptoms and receive appropriate interventions. This paper presents an AI-driven, web-based system that fills this gap by providing an accessible autism screening and therapeutic intervention platform. The proposed system integrates machine learning algorithms for symptom-based diagnosis and computer vision for image-based screening. The platform also includes awareness-raising educational content and accessible intervention guidelines for parents. The system is easy to use to ensure accessibility for low-technical-knowledge users. The results indicate that the AI-driven solution enhances the accuracy of diagnosis and provides a scalable solution for early autism screening and awareness in disadvantaged areas
Performance Analysis of HCEDV-Hop Localization Algorithm in Anisotropic Wireless Sensor Network
Full text linkAccurate and energy-efficient localization is an ongoing challenge in Anisotropic Wireless Sensor Networks (AWSNs), especially when AWSNs are deployed in irregular topologies (like valleys, coastlines, and mountainous terrain) versus regular topologies. This extended work presents additional performance evaluation of the previously introduced Hop-Correction and Energy-Efficient DV-Hop (HCEDV-Hop) algorithm. The HCEDV-Hop combines an error-correcting step with a hop-constrained broadcasting approach to improve localization accuracy and reduce energy consumption. In this study, we evaluate the HCEDV-Hop in anisotropic contexts where radio irregularities are direction-dependent and deployments in C-shaped fields are representative of real-world scenarios. The efficacy of the HCEDV-Hop is assessed using both regular and random deployments for a range of node densities, DOI values, and hop thresholds. Simulation results showed that localization errors increased in anisotropic fields but were still significantly reduced compared to conventional DV-Hop. While random deployment at DOI = 0.2 performed best, regular deployment maintained consistent accuracy. Broadcasting t hops decreased energy use without diminishing accuracy. Overall, the HCEDV-Hop performed better in ideal circumstances but remained reliable enough for real-world applications such as disaster management, environmental monitoring, and military surveillance
Analysis of Periodic Permeability on Free Convective Three-Dimensional Flow with Cattaneo-Christov heat transfer and Slip Effect
Full text linkThe present research paper contributes the slip effects on a three-dimensional viscous fluid flow for free convective boundary conditions with periodic permeability. Free convection fundamentally involves some heat transfer methods. In this work, the Cattaneo-Christov heat transfer method has been employed to develop the knowledge of heat transfer actions in complex flow porous system with periodic permeability. Moreover, the impact of the slip effect is investigated to more effectively deal with the boundary conditions. The mathematical model has designed for incompressible, viscous and laminar flow with free stream specifications. By using the regular perturbation approach, governing highly nonlinear partial differential equations are transferred into the ordinary differential equations in linear form together with certain linear partial differential equations. The separation variables approach is then used for transforming the linear PDEs to ODEs. Analytical solutions are obtained for the pressure, velocity field, components of skin friction, and temperature field. The influence of physical attributes existing in the mathematical representation of the physical occurrence is investigated and illustrated. Both the slip parameter and the Cattaneo-Christov heat flux have an impact of thickness on the thermal boundary layer of observed fluid flow
Computation of Rotational Flow of the Sun Using Satellite Data and Doppler Shift Calculations
Full text linkSolar rotational flow governs the Sun’s magnetic activity, space weather variability, and long-term dynamo processes. Traditional tracer-based techniques offer limited precision in mapping these flows, creating the need for direct spectroscopic velocity measurements. This study presents a computational framework for deriving full-disk Doppler velocity maps of the Sun using high-resolution Hα spectra from the Chinese H-alpha Solar Explorer (CHASE) mission. The H-alpha Imaging Spectrograph (HIS) data cube (2304 × 2313 × 46 pixels) was processed through a workflow of preprocessing, continuum normalization, Voigt profile fitting, and pixel-wise Doppler conversion to retrieve line-of-sight velocities. The resulting field of ~5.3 million pixels shows clear differential rotation, with blue shifts up to −7.89 km s⁻¹ on the approaching limb and red shifts up to +2.19 km s⁻¹ on the receding limb, corresponding to equatorial and polar rotation periods of ~25 and ~31 days, respectively. Localized asymmetries in active regions further reveal small-scale velocity perturbations. These results validate CHASE–HIS spectroscopy as a reliable tool for global solar flow diagnostics and highlight the utility of Voigt-based Doppler modeling in resolving fine-scale plasma dynamics. The developed approach bridges spectroscopic and Helio seismic methods, offering a reproducible foundation for future studies on solar dynamo modeling and space weather prediction
Harnessing LSTM Networks for Traffic Flow Forecasting: A Deep Learning Approach
Full text linkAccurate traffic flow forecasting in areas with different types of vehicles and varied driving behaviors is crucial for improving urban transportation systems and reducing congestion. In this paper, we introduce a Long Short-Term Memory (LSTM) approach to predict short-term traffic flow in such diverse conditions. Our model uses time-series data from real-world traffic sensors, capturing the patterns and dependencies that occur over time in mixed traffic environments. We tested the model using a dataset from seven days, with six days for training and one day for testing. The LSTM model achieved an R2 value of 0.96, a Mean Squared Error (MSE) of 2.82, and a Mean Absolute Error (MAE) of 1.13. These results demonstrate the effectiveness of LSTM networks in predicting traffic flow in complex traffic conditions, surpassing traditional machine learning models. This study provides valuable insights into using deep learning techniques for intelligent transportation systems (ITS)
A Comprehensive Toolset for Signal Processing using a Family of Hadamard Transforms
Full text linkIndependent basis of the linear vectors is of paramount significance in the advancement of digital systems that facilitate the processing and storage of information in its digital format. This study undertakes a thorough examination of discrete orthogonal transformations, with particular focus on the family of real and complex Hadamard transforms and their numerous types. The efficacy of various sequences is scrutinized, alongside their mathematical representation, inherent characteristics, and applications in signal processing. An analysis of the computational cost associated with the complex Hadamard Transform and its variants is presented. Furthermore, simulation outcomes are contrasted for the normalized sequency concerning magnitude response, image compression, and peak signal-to-noise ratio across a variety of image processing applications
A Gamified Approach to Reduce Obesity Through Physical Activity
Full text linkObesity is a major global health problem, directly affecting both illness and death rates. Sedentary lifestyles have led to high obesity rates, especially in developing countries like Pakistan. This study explores how gamified mobile apps can encourage physical activity to help fight obesity. The STRAVA app was used in a gamified intervention to track physical activity, motivation, and weight loss among participants. A quasi-experimental study design was used to assess the effectiveness of gamification in promoting behavioral change. The results show that gamification can significantly boost motivation, participation, and long-term commitment to physical activity, leading to weight loss and better health. The study highlights the potential of gamified mobile apps as affordable and scalable solutions for reducing obesity
Detecting Stance in Urdu Content on Social Media and Websites for Fake News and Propaganda Identification
Full text linkThe extensive spread of fake information has rendered various news types questionable, leading to a significant decline in trust in news. Social media is the primary channel by which fake news is disseminated widely. Worldwide, several deep learning methods have been created to identify fake news, achieving significant success with content in the English language. However, to our knowledge, there is no deep learning method available for detecting fake news or stance detection in content written in Urdu. Therefore, it is crucial to create a method that can detect fake news within Urdu language content. This study seeks to identify a method for detecting fake news in the Urdu language by proposing a framework that employs advanced Bidirectional Encoder Representations from Transformers (BERT), Embeddings from Language Models (ELMO), and various deep learning models (CNN, LSTM, Bi-LSTM) to evaluate performance accuracy on Urdu datasets (Liar-ProSOUL and Bend the Truth-Benchmark). We utilized Embeddings from Language Models (ELMO) for feature extraction and a convolutional neural network (CNN) for the classification task. The findings from the suggested framework indicate that ELMO excels with extensive dataset