This mini-project aims to classify the given document (news) regarding to if it is about a protest or not. This mini project consist of 2 main parts.
-
In the first part, news from The Hindu are extracted from their given URL and saved in a csv.
-
You can check extract_texts.ipynb notebook to see how they are extracted.
-
On the second part, first of all, raw data is being cleaned and couple points are analyzed on the cleaned data.
-
Then, two diffrent approach and two different algorithm for each approach are applied and results are shown as "Test Scores".
-
You can check preprocessing_and_models.ipynb to see the details and comments.
For more details on this project, check out the project report!
Due to copyright issues, we are not allowed to share the dataset. More information about the dataset can be found on:
Hürriyetoğlu, A., Yörük, E., Yüret, D., Mutlu, O., Yoltar, Ç ., Gürel, B., and Duruşan, F. (2020). Cross-context news corpus for protest events related knowledge base construction. In Automated Knowledge Base Construction (AKBC), June. URL: https://www.akbc.ws/2020/papers/7NZkNhLCjp