Assessfy Pvt. Ltd Moderate 5 milestones 100 marks

Multilingual Sentiment Analysis for Indian Customer Reviews

Target year: TE Sem 5-6 (Mini-Project-IIA/IIB) AICTE: 3 credits · ~75 hrs Bloom: Analyze MU CBCS: CSC602/CSC702 Mini-Project 2A/2B Powered by: AI Products Factory

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

Build a sentiment classifier that handles Hindi, English, AND code-mixed Hinglish (e.g. 'Bahut achha product hai, totally recommend'). Fine-tune IndicBERT or MuRIL on a labelled review dataset. Expose via REST API. Use case: Indian e-commerce + restaurant review analytics for SME owners.

Course Learning Outcomes (CLOs):
CLO1: Apply transformer fine-tuning to a multilingual classification task.
CLO2: Handle code-mixed text (Hinglish) tokenization correctly.
CLO3: Build evaluation harness with macro-F1 + per-language breakdown.
CLO4: Deploy a fine-tuned model as a REST API.
CLO5: Reason about bias + failure modes specific to Indian sociolinguistic context.

Industry/societal relevance: Indian SME ecosystem has 60M+ businesses on platforms like Google Maps / Zomato / Amazon. NLP-as-a-service for sentiment / review analytics is high-volume B2B. Companies: Sprinklr India, Rezo.ai, Yellow.ai, Haptik.

Milestones

1. Dataset + Labels

15 marks 7d

Use SAIL 2017 Hinglish + IIT Patna Hindi reviews. Combine + balance classes (positive/negative/neutral).

2. Baseline + IndicBERT FT

25 marks 14d

Baseline = logistic regression on TF-IDF. Fine-tune IndicBERT-base. Compare macro-F1.

3. Hinglish-Specific Eval

20 marks 14d

Slice test set into pure-Hindi, pure-English, Hinglish. Report per-slice F1.

4. REST API + Streamlit Demo

20 marks 18d

FastAPI endpoint /predict?text=... returns sentiment + confidence. Streamlit UI for live testing.

5. Bias Audit + Report

20 marks 22d

Identify 5+ failure cases (sarcasm, mixed-sentiment). 8-page IEEE-format report.

Open internships using this project -->

Open internships using this project 1

Verified companies are actively hiring interns to work on this exact project. Apply directly.

Multilingual Sentiment Analysis for Indian Customer Reviews

AI Products Factory · Remote · 12w

Certificate of Completion 2 of 2 seats

NLP for Indian SME analytics — Hinglish sentiment + intent on real customer reviews. You will fine-tune transformer models (IndicBERT / MuRIL) on code-mixed ...

Apply now

Skills you'll learn

NLPTransformer fine-tuning (IndicBERT / MuRIL)Code-mixed language handlingHuggingFace TransformersTokenization (Devanagari + Latin)Sentiment label encodingF1 + per-class metrics

Tools used

Python 3.11HuggingFace TransformersPyTorchdatasets (IIT-Patna reviews / SAIL 2017 Hinglish)StreamlitFastAPIGPU on ColabGitHub

Prerequisites

Python intermediate; intro to NLP + transformers; basic PyTorch; willingness to read HuggingFace docs

Available mentors

No mentors have signed up for this project yet.

Be the first to mentor

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

Multilingual Sentiment Analysis for Indian Customer Reviews

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

Multilingual Sentiment Analysis for I…

Assessfy

Auto-issued on completion One-click LinkedIn add