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 · Capstone
Stock Prediction with Twitter Sentiment
Fuses price features with tweet sentiment to predict next-day direction
5-6 weeks build
Overview
A capstone that scores tweets about a company with a finance-tuned transformer, aggregates them into daily sentiment features, and joins them with lagged price and volume indicators to predict next-day direction. Tweets come from the official X API or a public Kaggle tweet dataset, and an ablation shows whether sentiment adds anything over price alone. Not investment advice.
What makes it stand up
- 01Tweets collected through the official X API or a public Kaggle dataset, never scraped
- 02FinBERT-style sentiment scoring with daily aggregation and decay
- 03Ablation study: price-only versus price-plus-sentiment models
- 04Strict time-ordered splits to avoid look-ahead leakage
- 05Research demo only, with a clear not-investment-advice disclaimer
Module breakdown
- 01
Collection
API client with rate-limit handling and cashtag filtering.
- 02
Sentiment
Cleans tweets and scores them with a pretrained transformer.
- 03
Feature store
Joins sentiment aggregates with lagged technical indicators.
- 04
Modelling
Gradient boosting and an LSTM trained on time-ordered splits.
- 05
Dashboard
React view of sentiment, price and predicted direction per day.
After this, you will be able to
- Explain look-ahead bias and how your splits prevent it
- Defend the ablation that tests whether sentiment actually helps
- Demo the pipeline from raw tweets to a daily prediction
Technology stack
- Python
- Pandas
- Hugging Face Transformers
- PyTorch
- FastAPI
- React
- Level
- Capstone
- Domain
- Machine Learning
- Build time
- 5-6 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