Assessfy Pvt. Ltd Moderate 5 milestones 50 marks

Toxic Comment Classification System using Deep Learning

Real-world project · AICTE-aligned · AI-graded · Audit-ready certificate

Enroll as Student Mentor this project

Milestones

Available mentors

Enrolled students

Core skills

About this project

Toxic Comment Classification System using Deep Learning

Objective: Build a deep learning system to automatically classify toxic comments in online text.

Context: Toxic online behavior is a growing concern in India, impacting mental health and digital discourse; automated moderation tools are essential for safer communities.

What you'll build: You'll preprocess text data with NLTK, tokenize and embed sequences, and train an LSTM or CNN model using TensorFlow/Keras for classification. The system will be integrated into a simple web UI for real-time comment analysis.

Deliverables: Provide a trained classification model, codebase, evaluation results, and a demo web interface for toxic comment detection.

Milestones

1. Text cleaning

10 marks 6d

Clean the provided toxic comment dataset by removing HTML tags, special characters, and stopwords, and by normalizing text (e.g., lowercasing). Submit a Python script and a sample of the cleaned data. 'Done' means your script runs without errors and the output text is consistently preprocessed for downstream tasks.

2. Tokenization and embedding

10 marks 6d

Tokenize the cleaned comments and convert them into sequences suitable for deep learning. Apply word embeddings (e.g., Word2Vec, GloVe, or embedding layer). Submit your tokenization code, embedding implementation, and a sample of the resulting sequences. 'Done' means your process outputs padded sequences and embeddings ready for model input.

3. Model training

10 marks 6d

Build and train a deep learning model (e.g., LSTM, CNN) on the embedded comment data to classify toxicity. Submit your model architecture code, training script, and training logs. 'Done' means your model completes training without errors and produces a saved model file for later evaluation.

4. Accuracy evaluation

10 marks 6d

Evaluate your trained model on a held-out test set, reporting accuracy and confusion matrix. Submit your evaluation code, test results, and a brief summary of findings. 'Done' means you provide clear accuracy metrics and demonstrate the model's ability to distinguish toxic from non-toxic comments.

5. Web UI integration

10 marks 6d

Integrate your trained model into a simple web UI that allows users to input a comment and receive a toxicity prediction. Submit the UI code, deployment instructions, and screenshots of the interface in action. 'Done' means the UI loads, accepts user input, and displays model predictions correctly.

Open internships using this project -->

Upcoming sessions

Session	Window	Enrolled
Toxic Comment Classification System using Deep Learning	11 Jun 2026 to 10 Jun 2028	0

Skills you'll learn

NLP and DL (LSTM/CNN)Text preprocessingTensorFlow/Keras

Tools used

Python TensorFlow/Keras NLTK

Prerequisites

NLP basics Deep learning foundations

Available mentors

Priyang Kumar

Free

You'll earn — Certificate (PDF)

AICTE-aligned Project Completion Certificate

A formal, audit-ready PDF certificate issued by Assessfy + your institute on successful completion. Includes AICTE credit hours, your evaluator's signature, and a QR code for third-party verification.

AICTE-aligned

Certificate of Project Completion

This is to certify that

has successfully completed the project

Toxic Comment Classification System using Deep Learning

Scan to Verify

Date

Signature

Auto-issued on completion QR-verifiable

You'll earn — Digital Badge

Shareable LinkedIn / Resume Skill Badge

A compact, verifiable Open-Badges-2.0-compliant digital credential. Add to your LinkedIn profile, GitHub README, or resume in one click. Recruiters can validate authenticity via a unique URL.

Moderate

Toxic Comment Classification System u…

Assessfy

Auto-issued on completion One-click LinkedIn add