Browse
Recent Submissions
Item Fake news detection in roman urdu language(UMT, Lahore, 2024) Azan Ali and M.AbdullahThe fake news has been determined to exist very much in the modern world which at the same time has the possibility to manage the population and pose a threat to the real means that should help spread the truth. Indeed, in identifying fake news, detecting it is weighty because of written and spoken Roman Urdu in Pakistan and India including spelling, grammar, and no fixed form of a vocabulary, and little labeled data. In this try of detection of fake news in Roman Urdu , approaches which are used in the field of Machine Learning are used where different features are applied i.e. linguistic and semantic. Regarding the natural language processing context planning we can include several feature extraction approaches and some of them are known as token mapping, stemming/lemmatization, and POS tagging. We also use embedding’s, the sentiment analysis and also the topic modeling with a view of grasping the semantics of the text. The used dataset is news articles, scrapped ourselves from different sources , consists of 5500; Machine learning models of different types.ie. Multinomial Naïve Bayes, Decision Tree, Random Forest, XGB Classifier, SVC, Logistic Regression, KNN, Gradient Boosting Classifier, AdaBoost, ANN, and LSTM are trained and tested. Data set is taken from twitter, face-book, Newspapers which is discovered to be highly active, Forums, and from chat GPT. Hence, the Training accuracies, Test accuracies, as well as the precision score are the indicators of the Performance of the developed model. The Best accuracy of 94.34% and 70.79% of training and testing respectively was achieved and was the best model to be integrate with,LSTM model. We conclude that this Research Approach is effective and strong in detecting fake news in Roman Urdu. We also discuss the limitations and further directions of our work.Item Advancing s-box design through chaotic map integration(UMT, Lahore, 2024) Muhammad Bilal, Danial Abbas and Mohsin AliThe security of cryptographic systems is paramount in the digital age, where data protection and privacy are critical concerns. This project explores the enhancement of S-box design by integrating chaotic maps, aiming to improve the cryptographic strength and resilience of encryption algorithms. Chaotic maps, known for their sensitivity to initial conditions and complex dynamic behaviour, offer a promising approach to generating more secure and unpredictable S-boxes. In this study, we delve into the principles of chaotic maps and their application in S-box design. We develop a novel methodology for integrating chaotic maps into the S-box construction process, ensuring enhanced non-linearity, confusion, and diffusion properties. The performance and security of the proposed S-boxes are rigorously evaluated through various cryptographic criteria, including avalanche effect, non-linearity, and resistance to common attacks. Our findings demonstrate that the integration of chaotic maps significantly strengthens the cryptographic properties of S-boxes, providing a robust foundation for secure encryption systems. This research contributes to the field of cryptography by presenting a new direction for S-box design, leveraging the inherent unpredictability of chaotic systems to bolster data security. The outcomes of this project have the potential to influence future cryptographic practices and inspire further innovations in secure communication technologiesItem Plant diseases detection(UMT, Lahore, 2024) Rooha Ilyas, Saleha Maham and Zarmeen AhmedThis projecttaddresses thetsignificant impact of planttdiseases such as powdery mildew and rust on agriculturetin Pakistan by developing a smarttagriculture application for accurate disease detectiontand diagnosis. Utilizing image-basedttechnology, the project leverages thetResNet-50tframework to create atmachine learning model capable of classifying plant leavestas healthy, rusty, or powderytwith high accuracy. Initially intended fortreal-time drone-based detection, the project evolved intota two-phase approachtduetto technical challenges.tPhase one focused on trainingtthe model using the Plant DiseasetRecognition Dataset fromt universal dataset, achievingtan accuracy of 89%. Phase twotinvolved integrating thettrained model into atmobile application that allows userstto capture images oftplant leaves and receivetimmediate diagnostic results with confidencetscores. The application istdesigned to be user-friendlytand efficient, supporting varioustAndroid devices. Future enhancements willtaddress model overfitting, expand dataset diversity, andtexplore network solutionstfor drone integration to improve systemtreliability and utility.Item Fabric defect detection using computer vision(UMT, Lahore, 2024) Muhammad Talha Qadri, Muhammad Faizan and Faizan HaiderCV procedures, tending to a basic need in Pakistan's material industry. We fostered a continuous defect location framework by tackling the force of DL, expecting to improve the quality control processes in material assembling. The center of this examination included the assortment of a thorough dataset straightforwardly from the assembling floors of different material enterprises particularly from Faisalabad and across Pakistan. The dataset, including a different scope of textures and deformity types, was carefully per-handled and named to work with viable model preparation. We utilized both pretrained models such YOLO_V9, InceptionV3, EfficientNet, ResNet-18 and U-Net and exceptionally assembled DL model for defect discovery. The pretrained models were adjusted to suit the particular prerequisites of texture defect location, while the custom models were creatively intended to catch the novel qualities of material deformities. These models were thoroughly prepared and approved utilizing our dataset, guaranteeing power and exactness. Introductory outcomes show a promising limit of the framework to precisely distinguish a wide cluster of texture deserts. These discoveries not just display the capability of CV in quality control yet additionally lay the preparation for additional innovative work in this field. Through this venture, we seek to add to the height of Pakistan's material industry, improving its effectiveness and worldwide seriousness through mechanical developmentItem Animal tracking system(UMT, Lahore, 2024) Muhammad Ammar Murtaza and Muhammad AmirThis study introduces a novel approach for livestock farm tracking utilizing both RFID technology and IoT solutions. The proposed system aims to address the challenge of efficiently monitoring and managing animal movements within farm environments. By integrating RFID readers and tags with a central microcontroller and employing GPS modules alongside LoRaWAN communication, the system enables comprehensive real-time data collection and processing. RFID readers are strategically placed at entry and exit points, and RFID tags are attached to individual animals, facilitating accurate and timely tracking. The GPS module captures geographical coordinates, which are transmitted via LoRaWAN to a receiver unit and then uploaded to the ThingSpeak IoT platform. A custom-built application uses the ThingSpeak API to display the location data on a map, providing a user-friendly interface for real-time monitoring. Through rigorous testing and implementation, the system demonstrates significant improvements in performance and reliability compared to traditional tracking methods. The results indicate enhanced efficiency, accuracy, and adaptability in livestock management, offering promising prospects for optimizing farm operations. In conclusion, this integrated RFID and IoT-based livestock tracking system presents a viable solution to streamline animal monitoring processes, laying the groundwork for more efficient and sustainable agricultural practices.Item Leveraging GPT for embedded C code generation(UMT, Lahore, 2024) Muneeb Ullah, Sayyed Muhammad Hamza Chishti and Ali RazaThe need for dependable and effective embedded systems is increasing, which has led to a move towards creative methods of code generation. In order to generate embedded C code, this work investigates the use of Generative Pre-trained Transformers (GPT). Modern language models like GPT have proven to be exceptionally good at producing and comprehending natural language. Taking use of its strong contextual learning capabilities, this study explores the viability and efficiency of using GPT to automatically generate C code customized for embedded devicesItem Smart building(UMT, Lahore, 2024) Sadia Khurshid, Rimsha Gulzar, Waqas Abdullah and Maleeka ZainebOur SmartuBuilding project aims touaddress the crucial need forusustainable and energy-efficient building management. Byuleveraging advanced sensors and smartutechnologies, we strive touoptimize HVAC and lighting systemsudynamically, reducing electricity costsuand contributing to environmental sustainability.uThe project focuseuon key objectives, includinguenergy efficiency environmentaluimpact reduction, occupantucomfort, data-driven decision-making, automation, remoteumonitoring, and scalability. The stakeholders includinguuniversity administration, facilities management and end-users benefit fromuimproved comfort and energy efficiency. Theuproject's impact extends to the local community, showcasingua commitment to sustainability. Dependenciesuon external systems, such as electricaluand HVAC infrastructure, and referenceudocuments, including related projects and feature comparison, shape the project's framework. Requirements analysisuemphasizes real-time occupancy detection, ambientulight sensing, dynamic systemucontrol, user-friendly interfaces, integration, scalability, and energyumonitoring Actors and useucases, such as Occupant Controluand Facilities Management, outlineuinteractions within the smartubuilding system. The projectualigns with the university'sugoals, contributing to a sustainableuand comfortable environment whileudemonstrating social andueconomic benefits for stakeholders.Item Education-bot(UMT, Lahore, 2024) Bilal Hassan and Pehlaj RajputEDUCATION BOT is an innovative educational tool designed to revolutionize interactive communication and knowledge management in the digital learning landscape. This cutting-edge platform leverages a combination of advanced technologies to create a seamless, collaborative, and intelligent learning environment. At its core, EDUCATION BOT utilizes Javascript to implement a dynamic and user-friendly web interface, allowing for intuitive navigation and interaction. Effective processing and operation of PDF files (a backbone of instructional materials) is made possible by PyPDF2. Text manipulation is being done by using Langchian, which improves the system's capacity to analyze and comprehend complicated pdf material of course content. The implementation of OpenAI's LLM GPT-3.5-turbo into EDUCATIONAL BOT is a crucial component that allow interactive conversation between students and course Content. FAISS (Facebook Artificial Intelligence Similarity Search) is used as a Vector Databse and retrieval, database is searched and the similar documents are retrieved. This document cover details about the technology are used in the Final Year Project, what’s planned for the future, and how the EDUCATIONAL-BOT was developed. It shows how scalable and easy-to-use solutions can really highlight the power of AI and machine learning to change education. By adapting to the evolving needs of both teachers and students, it’s clear that these tools have the potential to make a big impact on the way we learn and teachItem AI-Powered pet care and marketplace(UMT, Lahore, 2024) Hammad Ahmad, M. Aman Gohar, Zarar Naseem and Shoaib AliThe AI-Powered Pet Care and Marketplace endeavors to transform the experience of pet care by leveraging technology to create a seamless and enjoyable process for pet owners, sellers, and care providers. This initiative addresses common challenges faced by pet owners, such as identifying health issues and accessing reliable care, through the integration of a user-friendly online marketplace with advanced AI features. Functioning as a centralized hub, this project offers a comprehensive range of services, or adopting animals. Additionally, the platform provides AI-powered breed identification and health recommendations, enhancing the overall well-being of pets and simplifying decision-making for their owners. In summary, this project seeks to revolutionize pet care by providing a convenient and supportive platform also prioritizes the welfare of pets and their human companions.Item Artificial intelligent recruitment system (AIRS)(UMT, Lahore, 2024) Hamna Rafi and Wasi AbbasThrough Artificial Intelligent Recruitment System (AIRS) we aim to make the hiring process smarter, user-friendly and efficient. Through the AIRS application we intend to enhance the experience of both candidates and HR. Traditional HR methods involve manually screening high volumes of resumes that could lead to human bias, assessing candidate personality by initial interview that is a costly process and communication gap between the candidates and HR due to often unavailability. We are using advanced Large Language Models (LLMs) to automate tasks such as resume screening, candidate ranking, candidate personality assessment through sentiment analysis and generating initial interview questions. Through AIRS we will bridge the communication gap between HR and candidates by providing chatbot assistance that will assist candidates 24/7 a week. TheiMachineiLearningi(ML)iandiNaturaliLanguageiProcessingi(NLP)itasks areiintegrated with an user-friendly web application that provides ease in accessibility. AIRS is based on the concept of Augmented Artificial Intelligence This project aims to bring efficiency and a modern touch to the hiring process making it easier for both employers and candidates.Item Event hub(UMT, Lahore, 2024) Muhammad Bilal, Muhammad Zunair Waheed and Muhammad Ahmad AshfaqIn modern life, people are neck and neck in competition with other people to survive in the corporate environment. Moreover, corporates must keep competing with other corporates to keep their business flowing. All this hustle and bustle has a severe effect on human health. It has become very difficult for people to even take out time for their dear ones. Resultantly people go through a difficult phase in organizing their social gatherings and events. Events for supposition birthday parties and marriages are supposed to be fun and cause happiness for people. Instead, they result in anxiety and headaches for people because of the arrangements they must make by taking out a short amount of time from their busy schedules. Nowadays almost every activity is being done online including shopping, booking rides, ordering food, and other activities. Keeping in touch with the new trends and recognizing the need of the hour the goal of this project is to provide people with a platform that will give them access to a marketplace of event planners. The application is designed after conducting thorough market analysis to understand the needs of both users as it allows the users to customize different functionalities to what suits them the best such as making custom packages, custom posting in the application and more. The user can get in contact with them and organize his different events with their help. The application provides people with a variety of service sellers to whom people will delegate different tasks or responsibilities. The system provides ease of use with a dedicated user- friendly interface for each type of user.Item Laptops hub(UMT, Lahore, 2024) Quratulain Nasrullah and Maham EjazIn today's world, laptops are like our trusted companions that support us at work, study and communication. But let's be honest, physically buying and selling them can be a little annoying. Think about how much effort goes into getting the real deal and finding the perfect laptop that's right for you. Go to the laptop center that has the right solution for you! It is like a laptop matchmaker that connects buyers with reliable sellers in Pakistan and used laptop market. Laptops HUB seeks to explore authenticity and individuality. It uses great machine learning tricks to make your laptop run more smoothly. It's not just about checking vendors; it stands for security, advanced performance and user experience, giving Pakistani laptops a whole new feel..Item JEdify(UMT, Lahore, 2024) M. Umair, M.Waleed and Faraz AhmedOur Project “JEdify” is an innovative educational platform aimed at bridging the gap between students and the IT industry. It seeks to empower students by helping them discover and pursue their true interests, providing comprehensive guidance and skills development to align their capabilities with industry demands. By focusing on personalized learning paths and industry-relevant skills, JEdify aims to transform students into valuable assets, ready to contribute effectively to the evolving IT sector.Item Digital city web portal for Lahore(UMT, Lahore, 2024) Nimra Abid and Mishaal TanveerThe Digital City Website Development project aims to design, develop, and launch a comprehensive website that serves as the central hub of a digital city initiative. This website will offer a plethora of features and functionalities where all the information is centralized to enhance urban living, promote community involvement, and advance sustainability. It is an integrated portal. We will put in place robust security measures and create an easy-to-use user interface for locals, companies, and guests as well as forgein’s within the digital city in order to safeguard user data and maintain the integrity of the website.Item Predictive model for lung cancer(UMT, Lahore, 2024) Saira Bukhari, Azka Noureen, Shiza Waheed and Duaa ZameerThe disease of lung Cancer is one of the biggest health challenges of this modern period. Millions of people around the world are suffering from this disease. It is also the second most common disease in Pakistan. Any effective treatment, which may involve radiation, chemotherapy, and surgery, depends on early detection and its prevention can be done by finding its risk factors and severity stage. Machine learning algorithms provide useful knowledge in a variety of fields by using data patterns to predict and make decision regarding future outcomes. In this study, we aim to use many machine learning algorithms to analyze risk factors of the lung cancer as well as its severity. We get the dataset from one of the international hospitals, which contains data of nearly five hundred healthy people and one thousand patients, which are suffering from lung cancer disease. This dataset includes various male and female patients which are at different stages of cancer. By using random forest algorithm and neural networks, which are finest machine learning algorithms, we find the risk factors of lung cancer disease. Results shows that the coughing of blood, obesity and passive smoker are more severe risk factors with respect to random forest, having weight of 22%, 15% and 12% respectively. While, the results with respect to neural networks shows that alcohol usage has load of 37% while coughing of blood and air pollution have load of 36% and 30%respectively. Also, by using random forest algorithm and neural networks, we detect the severity of this disease among those suffering patients, with high and accurate percentage. Our model train using random forest exhibits 98.6% accuracy, 98.2% precision, 99.0%recall and f1 score is 0.987. While, in case of neural network our model exhibits 97.3%accuracy, 96.7% precision, recall 98.0% and f1-score is 0.972. Random forest achieves high accuracy as compare to neural network because neural network needs larger datasets to achieve optimal results. Our models have significant detection result and, both can be used to make better decisions in healthcare regarding futureItem Computer vision for plant growth management(UMT, Lahore, 2024) Muhammad Abdullah Khan and Muhammad Faizan AliThe idea for our project emerged after witnessing a relative struggling with crop health management last season. In Pakistan, many farmers rely on traditional, manual methods to assess plant health—typically walking through fields and visually inspecting plants. This approach is not only labor-intensive but also prone to missing early signs of disease. To address this, we developed an affordable, IoT-based computer vision system using a Raspberry Pi paired with a basic camera module. The total cost of the setup was approximately PKR 40,000, making it significantly more accessible compared to the high-end equipment used on large-scale farms. The system captures images of crops and analyzes them to detect signs of disease, such as leaf spots or discoloration. Initially, we designed a complex dashboard with graphs and percentages, but field testing revealed that it was too difficult to use for those unfamiliar with digital tools. We redesigned it with a much simpler interface that provides a basic status—"Plant Healthy" or "Plant Sick"—along with brief suggestions for action. The system was tested on a local farm for three weeks in February. Results were mixed: it successfully detected diseases like potato blight early in some cases, but struggled with environmental challenges such as poor lighting and dust accumulation on the lens, which we hadn't initially accounted for. Currently, the detection accuracy is around 75–80%. While not perfect, it still represents an improvement over traditional methods. Work is ongoing to improve the model and address the environmental limitations, with plans for continued development in future..Item Career counselling(UMT, Lahore, 2024) Abu Hurera and Nadish UmairThe project aims to create a comprehensive career counseling platform designed to foster and enhance users' career development through interactive feedback mechanisms and structured guidance. This platform caters to three primary user roles: counselors, teachers, and administrators. For students, the platform offers a range of features including the ability to contact counselors for personalized advice, search for educational institutes, and access dynamically updated job listings fetched through the LinkedIn API. Students can sign up, log in, update their profiles, and stay informed about the latest job opportunities, ensuring they have the tools and information needed to make informed career decisions. Counselors play a crucial role in the platform by creating and maintaining their profiles, offering tailored career advice, and supporting students throughout their career journeys. They provide personalized counseling experiences aimed at guiding students towards achieving their career goals effectively. Administrators oversee the platform's operations, ensuring smooth functionality and a secure user experience. They manage user accounts, including the ability to delete users if necessary, and oversee the overall performance and security of the platform. Educational institutes using the platform can update crucial information such as admission dates and application form content dynamically. This flexibility ensures that students have access to the most current and relevant information, aiding them in their academic and career planning processes.Item Smart fashion assistant(UMT, Lahore, 2024) M.Umair Malik and Aiman AliThe project Smart Fashion Assistant is a flutter-based application that allows users to buy clothing products from a mobile app. Along with this it also allows users to get predictions from the app about the colors of cloth that will suit the user. Basically the Machine Learning module will detect the skin tone of the user and on the basis of skin tone it will suggest the user. For this app, we use Android Studio for flutter development and integration of machine learning, firebase for the backend of the app.Item Pak saaf(UMT, Lahore, 2024) Arbaaz Pervaiz, Ahmed Hassan Butt and Kashaf NoorIn response to the escalating challenges posed by urban waste management, this project introduces a visionary digital platform tailored for the dynamic cityscape of Lahore. This endeavor represents a fusion of cutting-edge technology and environmental responsibility, aiming to revolutionize waste collection and recycling practices. With a user-centric approach, the platform actively engages individuals, transforming waste disposal into a collaborative effort with tangible incentives. Stakeholders, including Customers, Riders, Administrators, Warehouse Staff, Payment Processing Teams, and Banking Institutions, form a cohesive ensemble in orchestrating the waste management process. Real-time incentives, integrated seamlessly into the platform, reward users based on the weight and material of contributed waste, fostering a culture of accountability and awareness. This innovative approach not only motivates individuals to participate actively but also educates them on the importance of sustainable waste management, thereby promoting long-term environmental stewardship. While the website serves as the primary interface for Customers, Riders, and Administrators, the unsung heroes, the Warehouse Staff, operate diligently in the background. Their responsibility extends beyond the digital realm, ensuring the dispatch of waste to recycling plants and contributing to a closed-loop cycle aligning with circular economy principles. They meticulously sort, process, and prepare the waste for its journey towards recycling, playing a crucial role in transforming waste into reusable resources. Their efforts underpin the entire system, ensuring that the digital platform translates into real-world environmental benefits. This project envisions a cleaner, greener Lahore, where technology and community engagement come together to tackle one of the most pressing urban challenges. By turning waste management into a rewarding and educational experience, we aim to inspire a collective shift towards sustainable living, one that other cities might emulate in their quest for environmental resilienceItem Heuristic for game based language learning and mobile application assistance(UMT, Lahore, 2024) Seth Usama, Yousaf Ahmad and Abdul RaffayThe Gamification is a integrations of game designs aspects into educations, especially languages acquisitions, with the goals of improving students’ engagement and learning. This study examines how well a gamified teaching techniques works in game-based language learning (GBLL), with a particular way and emphasis on the relationships between heuristic judgments and cultural settings. Our main goal is to discover a current heuristic and suggest new ones that are more efficiently, culturally sensitive by using (SSM) Soft Systems Methodology in conjunction and collaboration with a systematic literature review study. According to our research work, Gamification improves students’ engagement motivational, and recall of material. However, due to cultural biases and the requirement for adaptive learning algorithms, difficulties persist in developing frameworks that are relevant to all situations. We create a culturally appropriate instructional game model through iterative testing and stakeholder feedback, showing notable gains in language learning results above conventional approaches. This research provides recommendations for educators and game developers to maximize the effectiveness of GBLL and emphasizes the significance of cultural relevance in educational game design
- «
- 1 (current)
- 2
- 3
- »