In recent years, there has been a growing research interest in social media platforms such as Facebook and Twitter, which have been shown to be influential in reporting news and events from around the world. The conduct of regular, peaceful, and democratic elections is a central aim to a just and fair society. Unfortunately, electoral malpractice and violence continue to persist across a number of countries, preventing the consolidation of the democratic process. In this paper, we develop upon existing information retrieval and text classification approaches to develop effective classifiers that can detect electoral violence incidents from the users’ posts and discussion in Twitter. In particular, we create a large test collection by collecting and monitoring tweets pertaining to the Venezuela parliamentary election in 2015. Using state-of-the-art machine learning approaches, we develop a new text classifier tailored to the detection of electoral malpractice and violence, and demonstrate its effectiveness on the aforementioned 2015 Venezuela election test collection.