Did you read
the long post I made where I took the time to explain some things that are easy to understand? I specifically mentioned things which are commonly observed by people and do not require any kind of data analysis to prove. Most of what I outlined can be studied and confirmed by anyone, without any special knowledge, data, or algorithms. I even listed some examples of observations that you can make yourself to confirm different kinds of censorship and manipulations, which are things that most Twitter users see daily, even if they don't realize it.
Our team uses a proprietary stack developed in-house by people who are smarter than me. It runs crawling and data collection workers on a AWS compute cluster, pushes data to processing nodes via Kafka, before ingesting into one of our sharded ClickHouse database clusters. Further analysis is done using a variety of methods, ranging from simple aggregated metrics that run in the cloud, to machine learning models which use pytorch on dedicated GPU hardware which we have at the office. Team members can also use their own tools for discovery and local analysis.
www.rawstory.com
