Objective:
To identify and flag the authenticity of a given news dataset using machine learning to filter out misleading information on the internet.
Solution:
Analyse the features extracted from normalized text data using models such as decision tree and SVM, and perform document clustering by using an unsupervised hierarchical clustering algorithm such as Agglomerative Clustering.
Result:
The model is able to flag the news with true or fake labels with an accuracy of 66%.
To Do:
- Proper documentation of the project