This systematic review explores the utilization of crowdsourcing for geoinformation in enhancing awareness and mitigating terrorism-related disasters. Out of 519 studies identified in the database search, 108 were deemed eligible for analysis. We focused on articles employing various forms of crowdsourcing platforms, such as Twitter (now known as X), Facebook, and Telegram, across three distinct phases of terrorism-related disasters: monitoring and detection, onset, and post-incident analysis. Notably, we placed particular emphasis on the integration of Machine Learning (ML) algorithms in studying crowdsourced terrorism geoinformation to assess the current state of research and propose future directions. The findings revealed that Twitter emerged as the predominant crowdsourcing platform for terrorism-related information. Despite the prevalence of natural language processing for data mining, the majority of studies did not incorporate ML algorithms in their analyses. This preference for qualitative research methods can be attributed to the multifaceted nature of terrorism, spanning security, governance, politics, religion, and law. Our advocacy is for increased studies from the domains of geography, earth observation, and big data. Simultaneously, we encourage advancements in existing ML algorithms to enhance the accurate real-time detection of planned and onset terrorism disasters.