Political tweets Sentiment anayslis for Pro, neutral and anti-govt activities

Citation Author(s):
Ihsan
Ullah Khan
Muhammad U. S.
Khan
Submitted by:
Ihsan Ullah Khan
Last updated:
Thu, 06/20/2024 - 05:36
DOI:
10.21227/rf7t-6g73
Data Format:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

This dataset aims to identify the polarity of tweets—whether they are supportive, oppositional, or neutral—towards the current government. It comprises a total of 26,000 tweets: 15,000 in English and 11,000 in Urdu. These tweets were collected from 80 different political users' accounts to ensure a diverse and comprehensive representation of opinions.

 

The English dataset includes tweets from a wide range of political figures, activists, and commentators, providing insights into the English-speaking political discourse. Similarly, the Urdu dataset captures the perspectives of prominent Urdu-speaking political users, reflecting the nuances of political sentiment in this language.

 

Each tweet in the dataset is labeled with its polarity: support, opposition, or neutral. This labeling facilitates various analyses, such as sentiment analysis, opinion mining, and the study of political communication patterns. Researchers can utilize this dataset to explore how political opinions are expressed and propagated on social media platforms.

 

Overall, this dataset serves as a valuable resource for researchers interested in political sentiment analysis, multilingual text processing, and the study of social media's impact on political dynamics. The inclusion of both English and Urdu tweets enhances its utility for comparative studies across different linguistic and cultural contexts.

Instructions: 

To identify the polarity (support/opposition/neutral) of the tweets against the current government. For this purpose, a BERT based model is proposed to identify the polarity. We conducted our experiments on 13,000 tweets collected from 80 different political users accounts.

Comments

no comment

Submitted by rajan deo on Wed, 02/08/2023 - 07:21

no

Submitted by rajan deo on Wed, 02/08/2023 - 07:24