Word Sense Disambiguation of Regional Language using Deep Learning
Air Quality Prediction and Analysis using Embedded Sensors and Machine Learning
Unraveling Currency Depreciation
Revolutionizing Healthcare Monitoring: An Adaptive Wearable Framework for Realtime Decision Support
Web-Based Implementation of a Logistic Regression Model for Rapid FNA Cytopathology Image Analysis in Breast Cancer Detection
Efficient Agent Based Priority Scheduling and LoadBalancing Using Fuzzy Logic in Grid Computing
A Survey of Various Task Scheduling Algorithms In Cloud Computing
Integrated Atlas Based Localisation Features in Lungs Images
A Computational Intelligence Technique for Effective Medical Diagnosis Using Decision Tree Algorithm
A Viable Solution to Prevent SQL Injection Attack Using SQL Injection
India is widely renowned for having many different languages. Indians must be quite proud of the rich diversity of languages in their country. As there are numerous languages and a large number of meaningful words. There are terms that have synonyms in every language. There will be a time when learning a new language is necessary. Natural language processing (NLP) is helpful in achieving such. It has been around for more than 50 years, and the roots of NLP may be found in the study of language. It is used in a variety of fields, including corporate intelligence, search engines, and medical research. Ambiguity is the quality of being subject to multiple interpretations, and is frequently referred to as the imprecision of a word's meaning. The method for fixing this problem is called disambiguation. This study used a large- scale dataset that included words with many meanings and senses. Additionally, the dataset is in Telugu, the regional language of Andhra Pradesh. Some of the Telugu words in this dataset are unclear when used in various contexts. Deep neural networks are utilized to do this. Two algorithms namely Bidirectional Long Short Term Memory (BiLSTM) and Bidirectional Gated Recurrent Unit (RBGRU) are used in order to obtain the noticeable sense of given ambiguous word. Accurate word sense prediction achieved an accuracy of 86.30% for word sense disambiguation for regional language results. This outcome is remarkable when compared with other approaches of word sense prediction of various regional languages.
This paper introduces a comprehensive, real-time air quality monitoring and alert system that combines sensor-based hardware with a machine learning model for data classification. The system integrates gas sensors MQ135, MQ2, MQ9, a DHT11 temperature and humidity sensor, and an Arduino Mega 2560 microcontroller. Sensor data is processed using a Python-based Random Forest classifier to predict pollution levels and environmental conditions. Alerts are triggered through buzzers, LCD displays, and GSM-based SMS notifications. Unlike traditional setups requiring Wi-Fi modules, the system transmits data to the ThingSpeak platform using serial communication and Python scripting. This approach offers a scalable and low-cost solution for detecting hazardous gases and abnormal temperatures, making it suitable for residential, industrial, and public safety applications.
Currency depreciation, also known as devaluation, plays a pivotal role in international finance and economics, influencing trade dynamics, investment flows, and macroeconomic stability. It involves complex interactions between market forces, political developments, and economic fundamentals. Depreciation occurs when a nation's currency loses value relative to others in the foreign exchange market, driven by factors such as interest rate disparities, inflation differentials, trade imbalances, geopolitical tensions, and financial speculation. Understanding these mechanisms is essential for policymakers, businesses, and investors seeking to make informed decisions and manage associated risks. The effects of currency depreciation are multifaceted. While a weaker domestic currency can enhance export competitiveness and potentially stimulate economic growth, it also raises the cost of imports, contributing to inflation and reducing consumer purchasing power. Countries dependent on imported goods or with high levels of foreign- denominated debt face increased financial pressure as repayment costs rise. Additionally, depreciation can affect capital flows and foreign direct investment, typically prompting investor withdrawals and forcing central banks to intervene through interest rate adjustments or currency market operations. Currency depreciation is further compounded by political instability and economic uncertainty, which can perpetuate downward pressure on the currency. Addressing this phenomenon requires coordinated policy responses, including fiscal discipline, monetary tightening, and structural reforms. Businesses must adopt risk management tools such as currency hedging, while investors should monitor macroeconomic trends to anticipate shifts in exchange rates. Successfully managing currency depreciation demands collaboration among governments, central banks, corporations, and financial institutions to support stability and sustainable global economic growth.
This paper presents a comprehensive methodology for wearable health monitoring systems, emphasizing the Wearable Health Monitoring and Feedback Algorithm (WHMFA). By leveraging advanced data preprocessing techniques, real- time health assessments, and adaptive learning mechanisms, the WHMFA ensures accurate health status classifications, personalized feedback, and robust data security. The system incorporates noise filtering, machine learning-based predictions, and encrypted data transmission, ensuring reliability and privacy in healthcare monitoring. Simulation results demonstrate superior performance in accuracy, latency, and resource efficiency compared to existing systems, showcasing WHMFA's potential for enhancing patient outcomes and promoting real-time health management. This work addresses critical challenges in sensor reliability, privacy, and adaptability, contributing to the advancement of wearable health technologies.
Early detection of breast cancer significantly improves patient prognosis. Fine Needle Aspiration (FNA) cytopathology is a common diagnostic procedure, but its interpretation can be subjective and time-consuming. Machine learning (ML) models have shown promise in aiding this process. This paper details the design and implementation of a user-friendly, web-based application that leverages a pre-trained Logistic Regression model for the preliminary analysis of FNA cytopathology images. The system accepts a JPG image of an FNA slide, performs automated image segmentation and feature extraction, and subsequently feeds these parameters into the classification model to predict whether the sample indicates a benign or malignant tumor. The underlying Logistic Regression model, trained on the Wisconsin Breast Cancer Dataset, demonstrated a high accuracy of 97.07% on its test set. The web application, with its frontend deployed on GitHub Pages and backend on Render, provides an accessible platform for demonstrating the potential of ML in cytopathology, simplifying the input process to a single image upload and providing an immediate, interpretable result. This work highlights the practical application of existing ML research, making sophisticated analysis tools more accessible for educational, demonstrational, or preliminary assessment purposes. The complete source code is publicly available on GitHub.
