Browsing by Author "Shahid Bhatti"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Automated Extraction Of Multiple News
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2021) Azmat Ullah , Anum Riaz , Sarah Akbar Khan; FA17-BSE-022 , FA17-BSE-082 , FA17-BSE-100; Shahid Bhatti; LHR TP 7031
    News provides valuable information about the current situation. In the era of information and technology, People use different websites to get different news. These requirements raised the need for the system, which is Automated Extraction of Multiple news (AEMN). It will collect information from multiple online platforms and publish them on a single end-point. It also shows relevant news to the user. Moreover, it classifies news based on a different category, and news will be added to the relevant category. The source of the news is also provided to avoid copyright issues. In the future, AEMN is emerging as everyone wants to see authentic and similar information without searching too much or consuming time. In order to make the system more interesting, a different section will be added, such as auto-classification, similar/relevant news, and most trending news from different platforms without any manual effort. The system will be using Machine Learning algorithms to resolve all manual settings and searching.
  • No Thumbnail Available
    Item
    Skin Disease Identification Project
    (Library Information Services, CUI Lahore, 2023) Huzaifa Imran; /FA20-BSE-008; Shahid Bhatti
    The Skin Disease Identification Project aims to develop an advanced system using Machine Learning to build a user-friendly prototype of a system that with a simple picture can identify skin diseases in machine learning. This will be taken from a cross- platform mobile application identifying a variety of skin diseases with a wide range of conditions, including dermatitis, psoriasis, eczema, acne, and various infections. Diagnosis often requires a time-consuming process of visual examination, medical history review, and lab tests. The proposed system will utilize a vast dataset of dermatological approved and annotated images to train deep learning models that can effectively recognize and classify skin diseases based on color features and decomposed skin image and its melanin levels. The primary goal of this project is to enhance the accuracy or efficiency of skin disease diagnosis. By employing state-of-the-art image recognition and pattern analysis algorithms, the system will be able to rapidly process skin images and provide instant disease identification. This will serve as the basis for helping dermatologists with effective and instant diagnosis, therefore, saving time and reducing the delay in the diagnosis. The system will first identify the features of the images and then train the model based on it using various algorithms mentioned below which after a popular vote will determine the best match of the identification of the project. In conclusion, the Skin Disease Identification Project strives to revolutionize skin disease diagnosis through cutting-edge AI technology. By automating the identification process and providing instantaneous results, this project promises to be a valuable tool for dermatologists and serves as a prototype for more reliable feedback. Figure 1 illustrates the abstract idea.