Khandaker Abrar Nadib

Interaction Based Credibility Analysis of News on Facebook Using Machine
Learning Methodologies

Fake news and its dissemination are growing more popular as social media becomes more pervasive in our daily lives and Facebook users are particularly vulnerable. Numerous studies on detecting fake news have already been published. However, a study focusing on fake news in low resource languages like Bengali propagated through Facebook is rare due to data extraction challenges. In this paper, we work with fake news in the Bengali language and used various supervised machine learning algorithms to classify Facebook posts as fake, real, or satire and found XGBoost to generate the best outputs. Instead of the content, we trained our model with the interaction data of the posts to make it resilient against adversarial attacks. We observed that ensemble methods perform well with such social media metrics.

News Credibility Analysis on Facebook using User Interaction Data

Amid the COVID-19 outbreak, we initially focused on detecting fake health-related news on Bengali-language Facebook content. Recognizing information disparities, we transitioned from a content-based approach to a language-independent, computationally efficient interaction metric based method. The scope was further expanded beyond health related news to general news. We employed Machine Learning methodologies to classify public Facebook posts based on authenticity. The proposed method outperformed existing content-based and NLP-based solutions. Furthermore, our research demonstrated that user reactions or interactions with the system vary significantly based on the content of news articles, offering a useful way to gain valuable insights.

Heuristic Analysis of BKash

We conducted a heuristic analysis of a mobile financial application called bKash. I interviewed several users to identify and analyze various issues that are present within the platform. Identified some key problems with the primary functions of the app and ranked their severity via a heuristic analysis. Provided suitable approaches to resolve these issues and presented recommendations on how to enhance accessibility and multimodality.

University of Utah

Bangladesh University of Engineering and Technology (BUET)

GPA: 3.50/4.00

Last Two Semesters: 3.82/4.00

Major GPA: 3.68/4.00

Graduate Research Fellow

Software Engineer II

Digital Asset Management Team

Technologies: Python, Flask, JavaScript, TypeScript, React.js, MySQL, MongoDB, Alembic, Celery, Elasticsearch

Software Engineer I

Digital Asset Management Team

• Implemented Brand Template features including Download, Export, access, and Task integration.
• Implementing Searching, Filtering, and Navigation in DAM collection folders.
• Implemented various user activity tracking for analytics.
• Implemented various asset features like meta information, relations, bulk operation improvements, and GPT-3.5-turbo model integration to generate smart content.
• Handled user roles and privileges for various features.
• Made improvements to several backend and UI components in terms of performance, and code quality.

Technologies: Python, Flask, JavaScript, TypeScript, React.js, MySQL, MongoDB, Alembic, Celery, Elasticsearch

Software Engineering Intern

Asset Renditions Team

• Worked on implementing and maintaining a feature Asset Rendition.
• Built three services to generate asset renditions using the given specifications.
• Implemented stateless generators to scale horizontally and integrated asynchronous messaging for decoupling and scaling.
• Integrated the Rendition Service with the local development environment for developers.
• Implemented logging schemes to enable debugging by combining multiple services.

Technologies: Python, FastAPI, MySQL, PostgreSQL, Docker, Kubernetes, Message Queue