24 Hr AI Challenge: Build AI Fake News Detector

AI fake news detector helps detect fake news through binary classification methods. It builds better experiences by controlling the disinformation flow.

It's built on top of various powerful machine learning libraries. The tool works by training a neural network to spot fake articles based on their text content. When you run your own data through the tool, it gives you back a list of articles that it thinks are likely to be fake. You can then train the model further or decide if those results are acceptable or not. In addition to identifying fake news, this model can also be trained to identify real news. This allows you to compare the model's performance across different domains (e.g., politics vs. sports). The following installer and deployment instructions will walk you through the steps of creating an AI fake news detector by using fakenews-detection, jupyter, vscode, and pandas.

We will use fake news detection libraries (having fully modifiable source code) to customize and build a simple classifier that can detect fake news articles. kandi kit provides you with a fully deployable AI Fake News Detector. Source code included so that you can customize it for your requirement. With this kit, you can 1. Use a pre-trained model for detecting fake news. 2. Train the model on your custom dataset. 3. Expose the fake news detection as an API Add-on on examples are also included as given below 1. Use web scraper to automatically make your training dataset. 2. Visualise training and prediction data for useful insights.

Training and Certification - How to build fake news detection using machine learning

Watch this self-guided tutorial on how you can use training data, NLP pipeline, TF-IDF vectorizer, and text classifier to build your own AI fake new detector.

Completed the training? Apply for your Participation Certificate and Achievement Certificate now! Apply for Certification Tag us on social media with a screenshot or video of your working application for a chance to be featured as an Open Source Champion and get a verified badge.

Instructions to Run

Follow the below instructions to run the solution. 1. Locate and open the FakeNewsDetection-starter.ipynb notebook from the Jupyter Notebook browser window. 2. Execute cells in the notebook by selecting Cell --> Run All from Menu bar 3. Once all the cells of the notebook are executed, the prediction result will be written to the file 'fake_news_test_output.csv' Training with your dataset: 1. Add news articles to a csv file under a column name 'news_text'. 2. Add corresponding labels as 'real' or 'fake' denoting whether a news article is real or not. 3. You can refer to the file 'fake_news_train.csv' for an example. 4. Set the variable for training file in the notebook under Variables section. Testing with your dataset: 1. Add news articles to a csv file under a column name 'news_text'. 2. You can refer to the file 'fake_news_test.csv' for an example. 3. Set the variable for testing file in the notebook under Variables section. You can execute the cells of notebook by selecting Cell from the menu bar. For any support, you can reach us at FAQ & Support

Libraries useful for this solution

Development Environment

VSCode and Jupyter Notebook are used for development and debugging. Jupyter Notebook is a web based interactive environment often used for experiments, whereas VSCode is used to get a typical experience of IDE for developers. Jupyter Notebook is used for our development.

Exploratory Data Analysis

For extensive analysis and exploration of data, and to deal with arrays, these libraries are used. They are also used for performing scientific computation and data manipulation.

Text mining

Libraries in this group are used for analysis and processing of unstructured natural language. The data, as in its original form aren't used as it has to go through processing pipeline to become suitable for applying machine learning techniques and algorithms.

Machine Learning

Machine learning libraries and frameworks here are helpful in providing state-of-the-art solutions using Machine learning.

Data Visualization

The patterns and relationships are identified by representing data visually and below libraries are used for generating visual plots of the data.

Troubleshooting

1. If you encounter any error related to MS Visual C++, please install MS Visual Build tools 2. While running batch file, if you encounter Windows protection alert, select More info --> Run anyway. 3. During kit installer, if you encounter Windows security alert, click Allow. 4. If you encounter Memory Error, check if the available memory is sufficient and it is proportion to the size of the data being used. For our dataset, the minimum required memory is 8GB. If your computer doesn't support standard commands from windows 10, you can follow the instructions below to finish the kit installation. 1. Click here to install python 2. Click here to download the repository 3. Extract the zip file and navigate to the directory 'fakenews-detection-main' 4. Open terminal in the extracted directory 'fakenews-detection-main' 5. Install dependencies by executing the command 'pip install -r requirements.txt' 6. Run the command ‘jupyter notebook’ and select the notebook ‘FakeNewsdetection-starter.ipynb’ on the browser window.

Support

For any support, you can reach us at FAQ & Support