Browsing by Author "Sarker, Sajib"

Now showing 1 - 3 of 3

A Comparative Analysis of Identifying Multi Class Toxic Comment Classification Utilizing Deep Learning and Machine Learning Technologies
(Comilla University, Jan-2025) Sarker, Sajib
Despite extensive efforts to prevent and manage malicious human behavior, there are still many compelling issues in cyber platforms. When such individuals live in a world where technology is rapidly developing and where various online social media platforms like Twitter, Facebook,Instagram, etc. are widely accessible, those nefarious human acts rise. Because of this, extremely dangerous crimes like toxic comments, which pose a threat to users, groups, and even
A Comparative Analysis of Identifying Multi Class Toxic Comment Classification Utilizing Deep Learning and Machine Learning Technologies
(Comilla University, 1-Jan-2025) Sarker, Sajib
Despite extensive efforts to prevent and manage malicious human behavior, there are still many compelling issues in cyber platforms. When such individuals live in a world where technology is rapidly developing and where various online social media platforms like Twitter, Facebook, Instagram, etc. are widely accessible, those nefarious human acts rise. Because of this, extremely dangerous crimes like toxic comments, which pose a threat to users, groups, and even governments, have increased in frequency. This study aims to identify such harmful comments in Bangla text that are posted on social media sites, classify them using machine learning and deep learning algorithms, and take preventative measures. By combining TF-IDF with various machine learning algorithms, such as Naive Bayes, Decision Tree, Random Forest, AdaBoost Classifier, Stochastic Gradient Descent Classifier, Logistic Regression, KNN, and Support Vector Machine, we were able to detect toxic comments with an accuracy of 76.5%, 78.2%, 75.0%, 75.1%, 76.2%, 77.7%, 70.2%, and 78.5%, respectively. In comparison to previous machine learning techniques, we achieved an accuracy of 89% by employing deep learning algorithms
An advanced machine vision technique for quality control of fabric in the textile industry
(BRAC University, 2025-06) Chowdhury, Rifat Arman; Bhattacharjee, Souharda; Sarker, Sajib; Gani, Asim Ajwad; Roy, Pulak Deb; Alam, Md. Ashraful
The textile industry in Bangladesh, one of the leading sectors of the country’s economy, requires flawless product quality to ensure their worldwide reputation. However, one of the major setbacks they face is the detection of defects in fabric. Traditionally, quality control has relied on manual inspection, which is time-consuming, inefficient, and prone to human error; which incurs significant financial losses to the industry. Fabric defects introduced during manufacturing or processing make visual inspection necessary; however, the repetitive and monotonous nature of this task makes manual detection unreliable. This paper proposes a system to inspect faults on fabric in the production line of textile industries using machine vision techniques to ensure greater precision. We analyzed established deep learning based object detection models such as YOLOv8n, Single Shot MultiBox Detector (SSD) with VGG16 backbone, and Faster R-CNN with ResNet-50, and further developed a novel two-stage pipeline architecture. In our hybrid approach, YOLOv8l is employed for initial defect detection, followed by EfficientNet-B0 for classification. A custom dataset, containing a total of 9,705 images across four defect classes, was developed using pictures taken from textile factory environments to simulate real industrial settings. We also designed a prototype system for industrial automation to integrate our model into practical workflows. Fabric defect detection, automated using machine vision techniques is a growing research focus in the area, offering efficient, fast, and scalable solutions to quality control. While previous methods have been widely used, they often suffer from limitations in adaptability, accuracy, and real-world deployment. Our proposed model not only addresses these drawbacks, but also enables reliable inspection of fabric faults in industrial production. It is especially suitable for developing an economical and customizable system for fault detection. The pipeline achieved a precision of 0.842, recall of 0.819, and F1-score of 0.830—demonstrating both accuracy and industrial applicability.