International Journal of Innovations in Science & Technology
Not a member yet
813 research outputs found
Sort by
The Comparative Study of Machine Learning Algorithms for Sentiment Analysis in Multimodal Medical Data: Comparative Study of Machine Learning Algorithms for Sentiment Analysis in Multimodal Medical Data
No full textSentiment analysis, a part of data mining, uses Natural Language Processing (NLP) to understand how people feel about certain topics or individuals. It focuses on the context and polarity of information, measuring public opinions from unstructured sources like social networks and healthcare websites. By extracting useful insights from this unstructured data, healthcare professionals can improve patient care, make accurate diagnoses, and provide personalized treatments. Machine learning (ML) plays a key role in this process. ML techniques like logistic regression, decision trees, and Naive Bayes have proven effective in tasks such as sentiment analysis and named entity recognition in medical data. The goal of ML is to create algorithms that enhance data processing and decision-making by identifying patterns that might be overlooked by humans. In this study, we compare the performance of three common ML models—(a) Logistic Regression, (b) Decision Tree, and (c) Naive Bayes—for sentiment analysis on medical image captions. The Radiology Objects in Context (ROCO) multimodal image and caption dataset was used for this NLP task. Caption pre-processing is done using filtering methods to improve text quality, followed by sentiment classification using pre-trained ML models. This comparison sheds light on the effectiveness of these algorithms in performing sentiment analysis in clinical settings
Fabrication and Installation of Automatic Water Level Recorder through Global System for Mobile (GSM)
Get PDFThere are so many factors that contribute to water stress, including poor management of water distribution. These fluctuations are important to be known, as the properties of lake and river shores are significantly affected by the changes in water levels. An automatic water level recorder in this condition is essential for proper distribution of water to the fields, and for researchers to get the data via mobile. A system comprising of an Arduino Nano (open source), water-level sensor (float/magnetic sensor and ultrasonic sensor), and GSM module was proposed in this study to monitor the water level of a water body. The device performed very well in both good network areas, and bad network areas with an R2 value of 1.0 for the float sensor, and 0.9996 for the ultrasonic sensor. All the sensors were reliable and accurate, whereas in case of bad network areas the SMS received was delayed at an average of 5.7 minutes. This delay can only cause some issues when the data is needed on an immediate basis. The study concluded that the device built is reliable and can be used for the real-time monitoring of water levels
A Hybrid Model for Crop Disease Detection Based on Deep Learning and Support Vector Machine
Get PDFPakistan\u27s agriculture sector is the backbone of its economy, contributing significantly to its gross domestic product (GDP). However, a key challenge in this sector is to counteract the crop diseases timely because these diseases result in reduced production, increased cost and eventually lead to economic loss. Traditional disease control methods are costly, time-consuming, and often lack technical support, resulting in poor disease management and harmful environmental consequences. This research harnesses the unmatched capability of Artificial Intelligence (AI) and deep learning for timely disease detection in crops. This research introduces a hybrid model that combines deep learning models with a machine learning classifier for disease detection. AlexNet, Vgg-16, ResNet50, and MobileNet are the deep learning models that have been employed for the detection of various diseases in crop leaves of rice, potato, and corn. These models have been trained by using healthy and diseased leaf images of the mentioned crops and then these models are combined with a Support Vector Machine (SVM) classifier to enhance the accuracy of detection. Experimental results show the outstanding performance of this hybrid approach for timely disease detection in crops. It is further observed that the combination of MobileNet and SVM results in an impressive accuracy of 95.68% in disease detection. This technological approach would be beneficial for farmers in the effective management and control of crop diseases thus improving the crop yield and ultimately contributing to economic growth
Enhanced Trapezoidal Modulation in MMC: Comparative Analysis with Traditional Modulation Methods
Get PDFHVDC transmission and renewable energy systems extensively use Modular Multilevel Converters (MMC) because they provide outstanding scalability and modular architectural features. The performance quality of MMCs depends predominantly on which modulation technique engineers implement. This work studies Nearest Level Modulation (NLM) and conventional Trapezoidal Modulation alongside an enhanced Trapezoidal Modulation method to identify the top choice for high-voltage power implementations. The main goal of this research is to optimize modulation techniques for improving MMC harmonic performance and switching efficiency. Each modulation strategy is simulated through MATLAB/Simulink-based testing under identical operating situations. Product testing indicates NLM shows lower switching losses as well as superior power distribution efficiency but the updated Trapezoidal Modulation design combines reduced THD performance with simple implementation methods. The method\u27s innovative aspect depends on the modified trapezoidal waveform synthesis from a fundamental-frequency triangular signal enabling simplified implementation as well as lower THD and avoiding the need for high switching frequencies used in conventional approaches. The research delivers critical knowledge about MMC modulation selection which systems designers and manufacturers can use to optimize converter operation based on specific applications
Over-Voltage Protection Circuit for Tripping & Switching of 220V Appliances Using Relay Module and Python-Based Algo
No full textMitigating the risk of appliance damage caused by overvoltage conditions is a critical aspect of ensuring electrical safety and reliability. This study presents a novel "Over Voltage Protection Circuit for Tripping and Switching 220V Appliances Using a Relay Module," designed to address this challenge through an automated disconnection mechanism. The proposed system enhances protection, operational stability, and the longevity of connected appliances, making it a valuable addition to electrical safety measures. The circuit utilizes a combination of electronic components including transistors, capacitors, resistors, LEDs, diodes, and a relay module. Through the integration of these elements, the circuit detects over-voltage situations, triggers the relay to disconnect the appliance, and indicates the status through LEDs. The adjustable potentiometer allows for customization of the overvoltage threshold, enhancing flexibility and adaptability to varying electrical environments. Test results demonstrated that the circuit reliably disconnected the load at the specified voltage threshold and reconnected it when conditions stabilized. The design successfully accounted for hysteresis in relay operation, minimizing unnecessary toggling and ensuring stable performance. This practical and cost-effective solution offers reliable overvoltage protection for diverse applications
XDP-ML: A Game-Changer in Intrusion Detection Systems for Modern Cybersecurity
Get PDFIntrusion Detection system (IDS) plays a vital role in cyber security. Traditional approaches are not good enough to detect properly the large threats. Machine learning provides a promising solution and good accuracy by providing large data adaptability. This paper introduced an IDS approach using the XDP framework for real-time network traffic analysis. Objective: The primary goal of this paper is to improve IDS accuracy and effectiveness by integrating the IDS with the fast XDP-based machine learning approach. Motivation: Traditional IDS methods are defenseless to advanced attacks, so modern and adaptive solutions should be improvised. The XDP framework\u27s processing of the data at high speed makes it more resilient and ideal for real-time traffic analysis, enhancing IDS performance. Methodology: The proposed approach is evaluated using the CIC-IDS2017 and UNSW-NB15 datasets, which contain multiple network traffic features and attack labels. Results: The XDP-based machine learning approach enables real-time analysis and adapts to evolving threats. The XDP-based approach achieves a high detection rate of 98% to 99% with a low false positive rate. The performance is consistent and fast, demonstrating the productivity of the approach. Combining the IDS with XDP-based machine learning approaches makes more robust and scalable solutions for intrusion detection. The clear and accurate results show that it can handle advanced and more complex threats
Crypto Currency Compensation Model to Detect Optimal Channel of Internet of Things Through Blockchain
Get PDFThe ever-growing number of belongings of internet (IoT) devices in civilization creates a reliable, accessible, and safe infrastructure for processing the calculated data. One-point failures result from the prevalent IoT version\u27s use of an imperative cloud server approach. Because Blockchain uses a distributed community, IoT is integrated with Blockchain generation to avoid this. Consequently, this study has developed a fully autonomous and self-regulating learning system that can accurately operate channel time/spectral characteristics to communicate multi-user statistics. The future system is distinct in that it uses community metrics as its primary basis to recognize and adjust to increasing community density. Following the extraction of those capabilities, the projected protocol efficiently selects the appropriate channel for incoming nodes based on its interval features, recognizes and allocates the idle spectrum of nearby channels, and provides the optimal and appropriate channel utilization through an article called multilevel Gaussian radial and a multilayer non-linear assist vector machine (SVM) type model. The value consumption rate of the secure network and its functionalities is calculated in order to assess the performance of the proposed system. Future and conventional systems are compared. Associated to the prior model, the accuracy of the current model is 95.6%
Ion-Acoustic Density Hump Solitons in (r,q) Distributed Plasmas using Python
Get PDFElectron Velocity Distributions (EVDs) with a flat top at low energy and/or an enhanced tail at high energies are commonly observed in Earth\u27s magnetosphere and solar wind. Noteworthy is the fact that only generalized distribution with two spectral indices fit such observed flat top distributions, since at low energies neither kappa nor Maxwellian distribution can fit the observed EVDs. In the limiting cases, and ; distribution reduces the Maxwellian and kappa distributions, respectively. In the current fluid model, for the first time, electrons are treated as distributed and Sagdeev potential is derived for fully nonlinear fluid equations for ion-acoustic waves and obtained density humps to interpret the observations. We analyzed the properties of solitary structures using observed plasma parameters and values of and indices that matched the reported values. We found that flat top distribution supports the density hump solitons with larger amplitude
An IoT Distributive SM Controller for Mitigation of Circulating Currents Among Sources in a Standalone DC Microgrid
Get PDFSources of similar or different power ratings are connected in parallel within the DC microgrid. During operation, these sources generate circulating currents along with their normal currents, which disrupt proper current sharing among power electronic converters based on their capacity. Consequently, voltage regulation across the system weakens. Additionally, the resistance of the connecting lines contributes to this imbalance in current distribution. To address circulating currents, droop controllers are commonly used. This method allows converters to share power according to their capacity without requiring internal communication. However, a major drawback of conventional droop control is that as output voltage decreases, the converter\u27s output current increases linearly, leading to significant voltage fluctuations. As a result, droop control inherently involves a trade-off between voltage regulation and current sharing, making it impossible to optimize both simultaneously. To overcome this issue, this paper proposes a sliding mode (SM) controller implemented through an IoT-based distributed architecture. A system model is developed to evaluate its performance, and conditions for stability and existence are analyzed. MATLAB simulations provide detailed experimental results, demonstrating the effectiveness of the proposed technique
Smart Power Management with Small Cells: A Path to Sustainable Data Connectivity
Get PDFThe rising demand for energy-efficient networks capable of supporting high-speed data traffic poses a critical challenge for network operators. This study addresses this issue by proposing a power control strategy that dynamically adjusts small cell transmit power based on traffic patterns. Power usage is reduced to 40% of the total capacity during normal traffic, while 60% is utilized during high traffic intensity. This traffic-driven power allocation approach achieves a 13-15% improvement in energy efficiency compared to conventional small cell-controlled sleep modes. By optimizing energy consumption without compromising network performance, this research provides a practical solution for balancing efficiency and user satisfaction in modern mobile networks