Browsing by Author "Mohammad Ashrafuzzaman Khan"

Now showing 1 - 8 of 8
  • Thumbnail Image
    Item
    Open Access
    Are Abstracts Really the Abstracts: Alignment in Academic Research Papers and Summarizing Medical Research Paper With Deep Learning
    (North South University, 2023) Tanvir Khan; Mohammad Ashrafuzzaman Khan; 1911481642
    This research project investigates the degree of alignment between academic research and its summarizing abstracts in the medical domain, employing deep learning models for abstractive summarization. Abstractive summarization is a technique that generates summaries of texts using natural language generation rather than extracting sentences from the original texts. However, most existing studies on abstractive summarization focus on news articles or short stories, and few have explored its application to scientific literature, especially medical research papers. We used two state-of-the-art deep learning models, bert2bert and roberta2roberta, which are based on the encoder-decoder architecture and use pre-trained language models as encoders and decoders. We trained our models on a dataset of full-text papers and their abstracts from CORD-19 and other sources, which provided access to a large and diverse collection of medical research papers. We evaluated the performance of our models using ROUGE and METEOR scores, which are commonly used metrics for measuring the quality of summaries. We found that roberta2roberta outperformed bert2bert on both metrics, achieving a ROUGE score of 0.50 and a METEOR score of 0.48. We also compared the generated summaries with the original abstracts and performed a qualitative analysis of their differences and similarities. We found that the abstracts in the papers most often suffer from some common errors and limitations, such as repetition, inconsistency, omission, or exaggeration. We identified some potential benefits and challenges of applying abstractive summarization to scientific literature, such as improving accessibility and readability of research papers, reducing information overload, enhancing scientific communication, promoting interdisciplinary research, ensuring ethical and responsible use of AI, protecting intellectual property rights, preserving data quality and integrity, ensuring data privacy and security, reducing environmental footprint, etc. We concluded that abstracts don’t always reflect the whole paper through abstractive summarization of medical research papers, but further research and development are still required to overcome its challenges and limitations.
  • Thumbnail Image
    Item
    Embargo
    Dark Web E-Commerce Sites
    (North South University, 2020) Abdur Rakib Rahat; Sumaya Yeacin Rimu; Raihan Hossain; Anwar Shadaab; Mohammad Ashrafuzzaman Khan; 1620521042; 1530664042; 1520456042; 1611383042
    The dark Web is a much bigger place than our surface web in terms of the number of websites, so we cannot say the same thing about the e-commerce world of the Dark Web. Most of the e-commerce on the Dark Web is linked with various illegal activities. It can be considered a huge repository of selling drugs, weapons, pornography, personal documents, stolen & brand new electronic parts, and even body parts of humans. Using transactions via anonymous cryptocurrency, the Dark Web assured buyers that they could trade here without exposing their personal information. Privacy is a key component that makes the e-commerce of the Dark Web so popular, and it is increasing rapidly. Our purpose of this research is only to observe the structure of e-commerce websites of the Dark Web & the main causes that bring the interest of buyers to visit Dark Web e-commerce more & more. In order to accomplish that, this research paper represents our attempt to crawl and scrape dark websites.
  • Thumbnail Image
    Item
    Restricted
    Finding and exploring Similarity, relations, sentiments from a text, sentence, corpus using different NLP techniques
    (North South University, 2021) Shahnila; Asif Hassan Rahul; Samil Abdullah; Mohammad Ashrafuzzaman Khan; 1610450042; 1620787042; 1611408642
    Natural language processing (NLP) is the intersection of computer science, linguistics and machine learning to understand language much like humans do. The field focuses on communication between computers and humans in natural language and NLP is all about making computers understand and generate human language. It’s useful to businesses because it breaks down human language making it easier for machines to analyze automatically .With no-code NLP-driven platforms like Monkey Learn popping up NLP tools are becoming more accessible than ever, helping businesses automatically process huge quantities of text data, streamline their operations, reduce costs, improve customer satisfaction, and more. We have worked on extracting relations, similarities, entities, characters by using different natural language processing techniques from different books, news articles, emails etc. and displayed them in graph so people can view them easily and quickly . There were errors and we solved them. we have also applied different methods applicable to check which methods are more preferable.
  • Thumbnail Image
    Item
    Open Access
    Phishing Website Prediction Using Machine Learning And Explainable AI
    (North South University, 2022) Tushar Basak; Mohammad Ashrafuzzaman Khan; 1911858042
    Phishing websites look just like regular websites but are used to hack into other people's servers and steal their information. It is mainly used by hackers who intend to steal user data and identity. These hackers steal the most common user data: our login credentials and credit card numbers. It is difficult for an ordinary human to identify a phishing website from an official and secured website. They can be the target of a phishing attack without even realizing it. The proposed project is on phishing website prediction using machine learning. The objective and project aim is to make predictions about Phishing websites using Machine Learning and Explainable AL. Since it is difficult for users to realize whether the website will hack information, a model is being created to solve the problem. For that, machine learning models will be trained on phishing website features to identify such websites in the future. My Application will guard against user data leakage through the website. It ensures data protection on different levels, such as personal, organizational, and national.
  • Thumbnail Image
    Item
    Open Access
    Schoolaboration- Your All-in-One Academic Solution
    (North South University, 2023) Mabeean Suukyi; Israka Jahir; Sadia Sultana; Mohammad Ashrafuzzaman Khan; 1921348642; 1921312042; 1921746042
    The "Schoolaboration" project represents a significant milestone in the realm of academic management. This innovative platform is designed to address the multifaceted challenges that students and educational institutions face in today's dynamic educational landscape. It offers a holistic approach, integrating various essential features such as note organization, course access, task management, team collaboration, institute news updates, and a website blocking feature to enhance focus. In the current fast-paced academic environment, students often struggle to manage their coursework efficiently. "Schoolaboration" streamlines the academic experience, making it more user-friendly and productive. Moreover, the project not only caters to students' needs but also emphasizes the reduction of environmental impact. It curbs paper usage, conserves resources, and promotes sustainable practices by reducing carbon emissions from travel. "Schoolaboration" is not just an all-in-one educational solution but also an eco-friendly one.With a vision of widespread adoption, "Schoolaboration" is designed to accommodate changes and future enhancements. It aims to expand its compatibility with various devices and operating systems, improve its machine learning algorithms for better task recommendations, and integrate advanced security features.
  • Thumbnail Image
    Item
    Open Access
    Senior Design Project : Bill Monitoring System with Home Automation
    (North South University, 2022) Hasan Abrar; Walid Hossain; Sk. Abdul Walid; Mohammad Ashrafuzzaman Khan; 1632650643; 1722349642; 1712662642
    The world is quickly being automated. Because people have less time to complete tasks, automation is a convenient way to ensure that any item or piece of technology performs as we like. The purpose of this post is to show how to use an Arduino and a Bluetooth module to construct and build a home automation system. A home automation system delivers a simple and trustworthy technology with the Android application. A bluetooth module, an Arduino microcontroller, and an electricity meter are the primary components of the system. The connection route between the Android phone and the Arduino microcontroller is wifi. The complexity of the concepts involved in the home automation system is hidden by grouping them together into a simple, yet comprehensive collection of linked concepts. This simplification is necessary in order to pack as many features as needed on the limited area available on a mobile device's display. This article offers a system that is low-cost, secure, universally accessible, auto-configurable, and remotely operated. The technique described in the research is innovative, and it has been effective at controlling home appliances and creating a vital impact on the user to limit their unnecessary electrical uses.
  • Thumbnail Image
    Item
    Open Access
    Silent Speaker: A Lip-reading Model Using Deep Learning
    (North South University, 2020) Faria Karim Porna; Nazmul Hauqe; Mohammad Ashrafuzzaman Khan; 1620424042; 1530862642
    Silent Speaker is an applied model of human-computer interaction. This model can be applied in various vital areas, such as crime fighting and helping the hearing-impaired. It consists of one subject: Speech Recognition. This project is done to recognize speech without the presence or support of any auditory signal. So far, a lot of research has been done on lip-reading in English, French, Chinese, and many other languages. However, little research has been done to recognize speech from a silent video in the Bengali language. This thesis work provides a new approach to detecting some words in the Bengali language using deep learning. We want to classify some words using the dataset we created for this project. We track the distance between the inner and outer lip extract features for LSTM and create our model that can classify the word. Our final accuracy is 43%, and in the future, we want to increase our dataset size and modify our model to produce more accuracy.
  • Thumbnail Image
    Item
    Open Access
    Third Eye:Student attendance & behavior Monitoring in classroom through facial Recognition by analyzing live video data
    (North South University, 2022) MD Raihan Khan Student; Zannatul Islam Proma; Mohammad Ashrafuzzaman Khan; 1831118042; 1911916642
    We are more drawn to automated systems because of rapid technological advancements. For the majority of issues, we find automated systems to be the answer. Therefore, we came up with an automated system for student behavior monitoring in the classroom. This system captures and makes a summary of student behavior in the classroom. The faculty is in charge of overseeing student attendance, attention, and activities including entering and exiting the classroom in addition to making sure that lessons go well. Manual observation of these could affect the teaching and learning process of the faculty and students and causes a distraction from the main syllabus. The system records the entire session and identifies when the students pay attention in the classroom, and then reports to the faculties. Students’ performance can be recorded, and the data can be used for continuous assessment in the future. There are mainly two objectives of our project: First, to detect faces and recognize them for attendance and to detect students’ attentiveness and behavior in the classroom during lectures.