Assessfy Pvt. Ltd Moderate 5 milestones 100 marks

AI-Powered Product Recommendation Engine for E-Commerce

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

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

5
Milestones
0
Available mentors
0
Enrolled students
8
Core skills
About this project

Build a hybrid recommendation system for an e-commerce catalogue using collaborative filtering (user-item matrix factorization) and content-based filtering (product embeddings). Expose recommendations via FastAPI; A/B compare the strategies on a sample MovieLens or Amazon Reviews dataset.

Course Learning Outcomes (CLOs):
CLO1: Apply collaborative filtering using matrix factorization (SVD).
CLO2: Build content-based recommendations using sentence embeddings.
CLO3: Design a REST API that serves model predictions.
CLO4: Evaluate recommender quality using Precision@K and NDCG.
CLO5: Reason about hybrid strategies and cold-start problems.

Industry/societal relevance: Recommender systems power Flipkart, Myntra, Swiggy, Hotstar — directly placement-relevant.

Milestones
1. Dataset Prep + EDA
15 marks 5d
Use MovieLens-Small or Amazon Reviews subset. Load into Postgres. Compute stats: #users, #items, sparsity. Justify train/test split strategy.
2. Collaborative Filtering Baseline
20 marks 10d
Implement SVD-based CF via surprise library. Train/test on user-item ratings. Report RMSE + Precision@10.
3. Content-Based Recommender
25 marks 14d
Generate sentence embeddings for product/movie descriptions using sentence-transformers. For each user, recommend top-K nearest items to their historical favorites. Evaluate.
4. REST API + Hybrid Combiner
20 marks 10d
FastAPI endpoint /recommend/{user_id}?strategy=cf|content|hybrid. Hybrid = weighted combo. Document with OpenAPI/Swagger UI.
5. A/B Comparison + Final Report
20 marks 11d
Pick 20 test users. Show top-10 recs from each strategy. Compute Precision@10 + NDCG. Write 2-page report on which strategy wins on which user types + cold-start handling.
Open internships using this project -->
Skills you'll learn
PythonCollaborative FilteringMatrix Factorization (SVD)Content-Based FilteringSentence EmbeddingsREST API designEvaluation Metrics (Precision@KNDCG)
Tools used
Python 3.11scikit-learnsurprise librarysentence-transformers (HuggingFace)FastAPIPostgreSQLDockerGitHub
Prerequisites
Python intermediate; basic linear algebra (vectorsdot productmatrix multiplication); intro to ML; SQL basics
Available mentors

No mentors have signed up for this project yet.

Be the first to mentor
Share
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.

Certificate of Project Completion

This is to certify that

has successfully completed the project

AI-Powered Product Recommendation Engine for E-Commerce

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
AI-Powered Product Recommendation Eng…
Assessfy
Auto-issued on completion One-click LinkedIn add

Similar Projects you might like

Hand-picked by the recommender from your program & skill area.

Relevant Certifications to boost your application

From the Assessfy Certification library — take one and add it to your resume / LinkedIn before applying.

100 marks AICTE 3cr
Sign up & enroll