项目作者: anthonyc1

项目描述 :
Machine learning models to classify news datasets as real or fake
高级语言: Jupyter Notebook
项目地址: git://github.com/anthonyc1/FakeNewsClassifier.git
创建时间: 2018-11-24T21:40:12Z
项目社区:https://github.com/anthonyc1/FakeNewsClassifier

开源协议:

下载


FakeNewsClassifier

We wanted to apply data science and machine learning models to the topic of fake news. My team was inspired by after finding this article.

Steps we took:

  1. Data Collection
  2. Data Preprocessing
  3. Preprocessing the Text
  4. Text-to-Features Conversion
  5. Model Selection
  6. Classification

For a full breakdown of what we did, check out our report.

Our Kernel on Kaggle

Data Collection Kernel

Data Preparation Kernel

Our Findings

ROC curve for Classifiers

Classifier Breakdown

[Note: this is a partial repo. Not all our work is uploaded.]