Hybrid MDB Filtering Tool

Project Overview

The Moderated Discussion Board (MDB) in VU’s LMS is frequently cluttered with non-academic responses such as “good,” “done,” “present,” or phone numbers for WhatsApp groups. These messages reduce efficiency for faculty who must manually review hundreds of posts. This project proposes developing a browser-based tool integrated with the LMS front-end to automatically detect, filter, and optionally reply to non-academic messages. The solution will improve faculty productivity and maintain the MDB’s academic integrity. Simulate MDB data using mock HTML pages or exported static content.

        Collect and label a dataset of sample MDB messages (academic vs. non-academic).

        Implement two filtering approaches: a keyword-based system and an AI-powered classifier then

compare performance.

Users and Roles

Admin:

        Manage global keyword list.

        Update AI models.

        Set default filtering behavior.

Faculty Members:

        Use the tool for filtering MDB messages.

        Manage keyword list locally.

        Review and evaluate AI-classified results.

Functional Requirements

        Dataset Creation Module:

        Collect at least 500–1,000 sample messages (synthetic or crowdsourced), labeled as academic or non-academic.

        Store datasets in CSV or JSON format.

        Keyword-Based Filtering:

        Implement regex-based filtering for known patterns (good, done, present, sir, phone numbers).

        Provide a toggle to enable/disable keyword filtering.

        AI/NLP-Based Classification:

        Use TF-IDF + Logistic Regression or Naïve Bayes for classification (Python + scikit-learn).

        Optionally experiment with BERT or DistilBERT for advanced filtering.

        Display model accuracy (precision, recall, F1-score).

        Comparison Dashboard:

        Provide metrics comparing keyword filtering and AI classification accuracy.

        Allow faculty to review misclassified examples.

        Mock LMS Integration:

        Build a static MDB interface (HTML/JS/CSS) to simulate the LMS environment.

        Inject filtering functionality via a browser extension or userscript.

Page 69 of 167

        Export Feature:

        Export filtered academic queries into CSV or text.

Tools / Languages / Frameworks

Component Front-End Browser Extension Dataset Handling

Technology

JavaScript (ES6+), HTML5, CSS3

Tampermonkey or Chrome Extension API Python (pandas, scikit-learn)

NLP/ML Classificationscikit-learn, spaCy, or TensorFlow.js

          A dual-filtering system: keyword-based for simplicity and ML-based for adaptability.

          Demonstrates AI and front-end skills without requiring LMS backend access.

          A reusable dataset of MDB-like messages for future research or improvements.

Supervisor:

Name: Saima Jamil

Email ID: saima.jamil@vu.edu.pk

MS Teams ID: saima.jamil1988@outlook.com