# Empirical Study on Detecting Controversy in Social Media

**Authors:** Azadeh Nematzadeh, Grace Bang, Xiaomo Liu, Zhiqiang Ma  
**arXiv:** [1909.01093](https://arxiv.org/abs/1909.01093)  
**Date:** 2019-08-25  
**Summary page:** https://www.summarizepaper.com/en/paper/1909.01093v1/

## Key points

- Companies and financial investors recognize the importance of social consciousness in corporate strategies and investment decisions
- Supporting a sustainable economy requires considering social and environmental factors
- Public discussions surrounding incidents and controversies offer insights into companies' social consciousness and operational capability
- Proposed system uses Twitter data to detect and monitor controversial events and assess their impact on market volatility
- System identifies controversial events by clustering tweets with similar 5W terms and sentiment polarities
- Truthfulness of events is validated using credible news links shared within event tweets
- Case study on Starbucks Philadelphia arrests demonstrates effectiveness of the system in detecting and monitoring controversial events
- Incorporating social consciousness into corporate strategies and investment decisions is significant
- Novel approach leveraging Twitter data can identify controversial events and evaluate their impact on market volatility
- Research contributes to addressing challenges associated with assessing a company's social consciousness and environmental sustainability.

## Summary

Companies and financial investors are increasingly recognizing the importance of social consciousness in shaping their corporate strategies and investment decisions. They understand that supporting a sustainable economy for the future requires considering social and environmental factors. Public discussions surrounding incidents and controversies involving companies can offer valuable insights into how well these companies operate with regards to social consciousness, as well as indicate their overall operational capability. To address this issue, the authors propose a system that utilizes Twitter data to detect and monitor controversial events and assess their impact on market volatility. The system identifies controversial events by clustering tweets that share similar 5W terms (who, what, when, where, why) and sentiment polarities. The truthfulness of these events is validated using credible news links shared within the event tweets. The effectiveness of this method is demonstrated through a case study on the Starbucks Philadelphia arrests. By analyzing clustered tweets related to this incident, the authors show how their system can provide valuable functionality in detecting and monitoring controversial events. In conclusion, this empirical study highlights the significance of incorporating social consciousness into corporate strategies and investment decisions. It also introduces a novel approach that leverages Twitter data to identify controversial events and evaluate their impact on market volatility. This research contributes to addressing the challenges associated with assessing a company's social consciousness and environmental sustainability in an increasingly interconnected world.