Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/prachipatel15/fake-news-detection
In this data science project idea, we will use Python to build a model that can accurately detect whether a piece of news is real or fake.
https://github.com/prachipatel15/fake-news-detection
fake-news jupyter-notebook machine-learning python
Last synced: 2 days ago
JSON representation
In this data science project idea, we will use Python to build a model that can accurately detect whether a piece of news is real or fake.
- Host: GitHub
- URL: https://github.com/prachipatel15/fake-news-detection
- Owner: PrachiPatel15
- Created: 2022-02-01T04:50:00.000Z (almost 3 years ago)
- Default Branch: main
- Last Pushed: 2022-02-01T05:47:28.000Z (almost 3 years ago)
- Last Synced: 2023-05-30T19:34:58.212Z (over 1 year ago)
- Topics: fake-news, jupyter-notebook, machine-learning, python
- Language: Jupyter Notebook
- Homepage: https://github.com/PrachiPatel15/Fake-News-Detection
- Size: 41.6 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Fake-News-Detection: Project Overview
- In this data science project idea, we will use Python to build a model that can accurately detect whether a piece of news is real or fake.
- We've build TfidfVectorizer and used different classifier to classify between "Real" and "Fake" news.
- In this project, for optimization we've used __3__ different machine learning model.
- Using different classifiers, we've reached to optimal solution.
# Code and Resources Used
- ***Python Version:*** 3.8
- ***Packages:*** pandas, sklearn, matplotlib, pipeline, seaborn
- ***Dataset:*** https://www.kaggle.com/umarfarooq45/fake-news-and-true-news
# Data
- In this project, we have 2 different datasets named __"Fake.csv"__ and __"True.csv"__ and both datasets have below attributes:
- title
- text
- subject
- date
- We merged both datasets and added 'Target' column to classify which is true news or fake.
# Data Cleaning
- Raw dataset needed to be cleaned up little.
- Firstly, we dropped 'title' and 'date' columns so new dataset has below attributes.
- text
- subject
- Target
- Then we shuffled the data for exploration.
- And lastly We removed punctuations and stopwords from the 'text' column.
# Basic Data Exploration
- Created wordcloud for fake and true news.
- We checked the most frequent words in both the news and plotted the graph for both.
# Model Building
- I tried three different models and evaluated them using f1-score and confusion matrix.
- LogisticRegression -basline
- Decision Tree Classifier
- Random Forest Classifier
# Model Performance
- LogisticRegression : f1-score = 0.98
- Decision Tree Classifier : f1-score = 0.99
- Random Forest Classifier : f1-score = 0.99
- confusion matrix of Random Forest Classifier:
.png)