PopularMajor Project
Machine Learning
AI Resume Screening System
NLP pipeline that ranks candidates against a job description
- Python
- spaCy
- scikit-learn
- FastAPI
- +2
₹4,499₹7,999
4.9/214 reviews
Machine Learning · Major Project
Credit Card Fraud Detection
Imbalanced classification with SMOTE, cost-aware thresholds and PR curves
2-3 weeks build
Overview
Works with the public Kaggle credit card fraud dataset, where fraud is a tiny fraction of transactions. The kit compares class weighting, random undersampling and SMOTE, applied only inside training folds, then chooses a decision threshold from a cost matrix rather than defaulting to 0.5. A FastAPI endpoint scores transactions in real time.
What makes it stand up
- 01SMOTE, undersampling and class weights compared fairly
- 02Resampling applied only inside training folds to avoid leakage
- 03Precision-recall curves and PR-AUC instead of plain accuracy
- 04Threshold chosen from the cost of missed fraud versus false alarms
- 05FastAPI scoring endpoint with SHAP explanations per transaction
Module breakdown
- 01
EDA
Class imbalance, amount and time distributions.
- 02
Resampling
imbalanced-learn pipelines for each strategy.
- 03
Models
Logistic regression, random forest and XGBoost.
- 04
Thresholding
Cost-based threshold search on validation data.
- 05
Scoring API
FastAPI service returning a score and explanation.
After this, you will be able to
- Explain why accuracy is meaningless on this dataset
- Defend where SMOTE belongs in the pipeline
- Demo a transaction being scored and explained live
Technology stack
- Python
- Pandas
- scikit-learn
- XGBoost
- imbalanced-learn
- FastAPI
- Level
- Major Project
- Domain
- Machine Learning
- Build time
- 2-3 weeks
- Difficulty
You receive
- Complete, commented source code
- 60-90 page project report (IEEE format)
- Editable presentation deck
- Architecture, ER and UML diagrams
- Local setup and deployment guide
- Sample dataset or seed data
- Viva question bank with answers
- One 45-minute walkthrough call
- 30 days of doubt-clearing support