Project Report
Browse
35 results
Search Results
Item Design and Implementation of an IoT based Smart Weather Monitoring System in Poultry Farm(Daffodil International University, 2025-02-03) Rashel, Abu; Hasan, Md. MehediThe incorporation of Internet of Things (IoT) technology in chicken farming provides creative approaches to enhance environmental conditions essential for animal well- being and efficiency. This research introduces an intelligent weather monitoring system designed specifically for poultry farms, utilizing IoT devices to gather real-time information on temperature, humidity, air quality, and light levels. The system employs wireless sensors, cloud-based data storage, and machine learning algorithms to providepredictive insights and automated control for maintaining optimal conditions. These capabilities minimize human intervention, enhance energy efficiency, and reduce operational costs while ensuring precise environmental regulation. Alerts for anomalies, such as extreme temperatures or poor air quality, enable timely correctiveactions, reducing the risk of disease and mortality. This IoT-driven approach demonstrates significant potential to improve the sustainability and productivity of poultry farms, aligning with modern agricultural advancements and the growing demand for intelligent farming solutions.Item Comparison of Breast Cancer Prediction Using Machine Learning(Daffodil International University, 2025-01-12) Hasan, Md. MehediRecent times, breast cancer has seen a concerning rise, affecting a significant proportion of women. To tackle this pressing issue, extensive research efforts have been dedicated to devising effective methodologies for early detection and prediction. Our proposed approach leverages techniques to predict potential risks also promote recent alert of breast cancer. What sets our approach apart is its practical applicability in real-world scenarios, offering a straightforward method for breast cancer prediction. We harnessed the power of four datasets hosted on the Kaggle platform and integrated various classifiers, including Decision Tree (DT), Random Forest (RF), Logistic Regression (LR), K-Nearest Classifier (KNN), among others, into our model. The results were promising, with the KNN achieving a noteworthy test accuracy of 81.14% for Dataset A, KNN of 97.2% for dataset B, KNN of 98.85% for dataset C and LR of 96.125% for dataset D. Furthermore, Bagging KNN also demonstrated accuracy matching this high standard of 99.42%. To further enhance performance, we implemented a range, including Bagging, Boosting, Stacking and Voting algorithms, optimizing each classifier with the best parameters through hyperparameter tuning. Through our experimental investigation, we not only contributed to the body of knowledge on breast cancer detection and prediction but also identified the KNNB (K-Nearest Classifier with Bagging) model as the most accurate, achieving an outstanding accuracy rate of 99.42% for breast cancer predictions. This research endeavors to provide invaluable insights into breast cancer management, offeItem Comparison of Breast Cancer Prediction using Machine Learning(Daffodil International University, 2025-01-13) Hasan, Md. MehediRecent times, breast cancer has seen a concerning rise, affecting a significant proportion of women. To tackle this pressing issue, extensive research efforts have been dedicated to devising effective methodologies for early detection and prediction. Our proposed approach leverages techniques to predict potential risks also promote recent alert of breast cancer. What sets our approach apart is its practical applicability in real-world scenarios, offering a straightforward method for breast cancer prediction. We harnessed the power of four datasets hosted on the Kaggle platform and integrated various classifiers, including Decision Tree (DT), Random Forest (RF), Logistic Regression (LR), K-Nearest Classifier (KNN), among others, into our model. The results were promising, with the KNN achieving a noteworthy test accuracy of 81.14% for Dataset A, KNN of 97.2% for dataset B, KNN of 98.85% for dataset C and LR of 96.125% for dataset D. Furthermore, Bagging KNN also demonstrated accuracy matching this high standard of 99.42%. To further enhance performance, we implemented a range, including Bagging, Boosting, Stacking and Voting algorithms, optimizing each classifier with the best parameters through hyperparameter tuning. Through our experimental investigation, we not only contributed to the body of knowledge on breast cancer detection and prediction but also identified the KNNB (K-Nearest Classifier with Bagging) model as the most accurate, achieving an outstanding accuracy rate of 99.42% for breast cancer predictions. This research endeavors to provide invaluable insights into breast cancer management, offering a potential solution for early intervention and ultimately improving patient outcomesItem Maternal Health Risk Prediction Based on Health Checkup Using Machine Learning Approaches(Daffodil International University, 2024-07-13) Hasan, Md. MehediMaternal health difficulties are currently one of the most difficult challenges in the world. Every year, many women die during pregnancy and after childbirth, which is a primary source of infant mortality. Maternal risk factors such as the mother's chronic illness, blood pressure, mental health, diet, and other medical care during pregnancy all play important roles. Pregnant women in remote locations confront several obstacles and challenges, including a scarcity of doctors, insufficient expertise, a lack of accessible clinics, infrastructural constraints, and transportation issues. The infant's poor health is mostly due to the mother's pregnancy, rather than any additional issues that may have occurred following childbirth. Using machine learning approaches, the study has predicted the maternal health risk level in previous due to avoid uncertain birth death or any inconvenience of a new born child. A variety of pre-trained advanced machine learning techniques were utilized in the study to find out the sustainable result. ANN, Ridge Classifier, SGD, XGBoost, Cat Boost, Random Forest, XGB, Decision Tree, and more algorithms were implemented. The recommended model was created, trained, and tested on the preprocessed dataset with the help of Hyper Parameter Tuning. The Cat Boost Classifier was the most accurate machine learning system for the study with a score of 97.4%.Item E-Nursery an android base E-Commerce application(Daffodil International University, 2024-07-24) Hasan, Md. MehediE-Nursery” is an Android-based e-commerce platform tailored for the agricultural sector, presenting a virtual nursery and agricultural supply store. E-Nursery refers to the use of electronic commerce websites or applications for purchasing and selling products. It is a user-friendly application. No authentication is required to use this application. Even those who are not familiar with phones can use the application very easily. In this app, plant seeds can buy various flower plants, water plants, indoor plants, soil, plant tubs, organic fertilizers, and various nursery equipment. Managing these things becomes very difficult especially for those who live in cities. There are some applications in the market compatible with the in-nursery application, among them Daraz, Green Dhaka, Plant House BD, and Othoba. Farmers can also know what kind of care a tree needs through this application. Bangladesh has six seasons. In these six seasons, in which season it is better to cultivate which kind of vegetables, you can also get the idea from this application. Those who own a nursery will also be greatly benefited by this nursery android app. In terms of Bangladesh, the prevalence of e-nursery is very low. But there are a number of gardening enthusiasts who have their garden hampered by not having the right thing at hand at the right time. This nursery will mainly provide all gardening materials to the urban people, various big factories of the city, and residential projects.Item Computer vision-based transfer learning techniques for classification of local pigeon species in Bangladesh:(2024-01-25) Hasan, Md. MehediIn the realm of avian conservation, this thesis embarks on a pioneering journey to enhance the classification of pigeon species within Bangladesh. Leveraging the powerful Xception model, we present a breakthrough approach that attains an exceptional testing accuracy of 99.47% and minimal loss of 0.025. Our study encompasses a comprehensive dataset of 7500 images, spanning 15 pigeon species, and employs transfer learning for swift and reliable classification. While the results underscore the efficacy of our approach, the study acknowledges the challenge of subjective criteria in species classification and calls for future exploration into enhancing interpretability. Ethical considerations are central to our findings, advocating transparent communication with conservationists and the establishment of stringent ethical guidelines for responsible technology application in avian conservation. This research, a significant stride at the intersection of technology and ethics, not only contributes to avian conservation but also lays the groundwork for future investigations, paving the way for a sustainable future in avian species management and urban biodiversity preservation.Item Computer vision-based transfer learning techniques for classification of local pigeon species in Bangladesh: A comparative analysis(Daffodil International University, 2024-01-01) Hasan, Md. MehediIn the realm of avian conservation, this thesis embarks on a pioneering journey to enhance the classification of pigeon species within Bangladesh. Leveraging the powerful Xception model, we present a breakthrough approach that attains an exceptional testing accuracy of 99.47% and minimal loss of 0.025. Our study encompasses a comprehensive dataset of 7500 images, spanning 15 pigeon species, and employs transfer learning for swift and reliable classification. While the results underscore the efficacy of our approach, the study acknowledges the challenge of subjective criteria in species classification and calls for future exploration into enhancing interpretability. Ethical considerations are central to our findings, advocating transparent communication with conservationists and the establishment of stringent ethical guidelines for responsible technology application in avian conservation. This research, a significant stride at the intersection of technology and ethics, not only contributes to avian conservation but also lays the groundwork for future investigations, paving the way for a sustainable future in avian species management and urban biodiversity preservation.Item Cellphone:(Daffodil International University, 23-02-12) Hasan, Md. Mehedi; Purno, Sonjay DeyIn the modern world, autonomous technology has largely taken the place of directing structures. An e-commerceewebsite is one thattenables users to buy and sell physicallgoods, services, and digital goods over the internet as opposed to at a physical store. Through an e-commerce website, a business can manage orders, payments, logistics, shipping, anddscustomer service. Our website name is SellPhone. Here any users(buyers) can sell their old phones and users(sellers) can sell their old phones. Our website is a full-stack and responsive website with some good features like searching for category-based products. The front end is built with the javascript framework ReactJS. We designed the website with the CSS framework tailwind. The back end is built with ExpressJS and NodeJs. For the database, we used MongoDB to save all kinds of information. For the user login system, we used Firebase. To secure our APIs we used JWT (JASON WEB TOKEN). Buyers can easily purchase an old used cellphone and pay their payment by using any international card.Item Solar Energy-Based Smart Street Light Control System(Daffodil International University, 23-01-29) Hasan, Md. MehediDistribution stations provide alternating current for most streetlights. As a result, utility companies are required to distribute more power. Currently, however, there is a possibility of using renewable energy sources for streetlights to reduce utility company consumption rates. A streetlight will turn on at 40% of its maximum intensity when there is no object beneath it, according to the program. An Arduino is set to increase the brightness to 100% if any person or vehicle passes a nearby streetlight. Upon expiration of the preset time and if there is no object detected, the intensity gradually decreases to 40%. In the morning, LDR will send a command to Arduino and the streetlight will be turned off. Battery-operated streetlights normally operate with electricity stored in them. When the battery is not sufficiently charged, streetlights will automatically switch to utility power. Keywords: streetlights, LDR, PV cells, brightnessItem SMS Gateway(Daffodil International University, 22-11-02) Hasan, Md. MehediSMS Gateway is a web-based application that is using for sending masking and non-masking message. Send SMS by API, Quick SMS, Variable SMS, and Campaign SMS. There have two types of users, one is postpaid and another is prepaid.