Capstone-Project

This project is the Final Year BTech Project.

We are using svm_dataset_2 (1).ipynb as final model as it has better accuracy.

Key highlights of this project:

We compared the performance of three machine learning classifiers: Support Vector Machine, Logistic Regression and Naive Bayes on a dataset. The model with the best accuracy was chosen to predict if a news was fake or real.
The dataset was created by web scraping news articles from news website. This dataset was updated with new data on daily basis.
We retrained the model on weekly basis so it the remains aligned with current trends, improving predictive accuracy and robustness over time.

Name		Name	Last commit message	Last commit date
Latest commit History 17 Commits
Capstone project backend		Capstone project backend
Capstone project docs		Capstone project docs
Week 1 or Day 1 accuracy 63 fake 51 real		Week 1 or Day 1 accuracy 63 fake 51 real
week 2 or day 5		week 2 or day 5
week 3- 4 May		week 3- 4 May
Final Year Project Report (Black Book).pdf		Final Year Project Report (Black Book).pdf
README.md		README.md
SVM Model.jpg		SVM Model.jpg
accuracy_for graph.py		accuracy_for graph.py
naive_bayes_dataset_1.ipynb		naive_bayes_dataset_1.ipynb
naive_bayes_dataset_2.ipynb		naive_bayes_dataset_2.ipynb
svm_dataset_2 (1).ipynb		svm_dataset_2 (1).ipynb

Provide feedback