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 · Mini Project
Hate Speech Detection
Sorts tweets into hateful, offensive or neutral with an explained model
1-2 weeks build
Overview
A three-way text classifier trained on the public Davidson hate speech and offensive language dataset. Tweets are cleaned, tokenised and turned into TF-IDF features, then logistic regression and linear SVM are compared with class weighting, because hateful posts are a small minority and a naive model simply learns to ignore them.
What makes it stand up
- 01Tweet-aware cleaning for mentions, hashtags, URLs and emoji
- 02TF-IDF word and character n-gram features compared side by side
- 03Class weighting to handle a heavily imbalanced label set
- 04Per-class precision, recall and F1 with a confusion matrix in the report
- 05Streamlit demo that highlights the words driving each prediction
Module breakdown
- 01
Preprocessing
Normalisation, stop-word handling and lemmatisation.
- 02
Features
Word and character TF-IDF vectorisers with tuned vocabularies.
- 03
Models
Logistic regression, linear SVM and naive Bayes baselines.
- 04
Evaluation
Stratified cross-validation and per-class error analysis.
- 05
Demo app
Paste a post and see the label with its top contributing terms.
After this, you will be able to
- Explain why accuracy misleads on imbalanced classes
- Defend the difference between offensive and hateful labels
- Demo live classification with an explanation of each result
Technology stack
- Python
- scikit-learn
- NLTK
- Pandas
- Streamlit
- Level
- Mini Project
- Domain
- Machine Learning
- Build time
- 1-2 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