This is a text mining project over 200k Tweets regarding a reputation crisis in a religious institute.

The study scraped all the 259632 tweets searched by the key word “Joel Osteen” and analyzed the corresponding trends of the crisis by the method of text mining and a bit of machine learning to analyze the sentiments of them.

The project found that while the paracrisis escalated to an overwhelming crisis within one day, it dissolved in a drastic way, given the crisis has been addressed properly and swiftly by the church and Joel Osteen. What is more, the faith-holders of the organization have not been influenced much during and after the crisis recedes. And they have been found actively participated in the discussion regarding the crisis.

Due to confidentiality, the raw dataset is not uploaded here. A random sample of the data can be shared by inquery. Please email me at [email protected] if interested.