The proliferation of online social media platforms has led to an increase in various types of content, including online hate. This trend poses substantial risks by amplifying harmful ideologies, inciting violence, and perpetuating discrimination. In response to this growing concern, Machine Learning (ML) has emerged as powerful tools for the automatic analysis of online hate. Researchers from diverse fields, including the Social Sciences and Information Science, are increasingly turning to ML for solutions. However, researchers are facing fundamental challenges in accessing essential resources, such as datasets, ML models, and analysis tools. In this paper, we present Integrative Cyberinfrastructure for Online Abuse Research (ICOAR), a system that automates the process of collecting, analyzing, and visualizing online abuse data. ICOAR pipeline begins with automated data collection from various social media platforms, followed by integration of state-of-the-art ML models to streamline the detection, categorization, and analysis of online hate. ICOAR also features customizable tools for data visualizations, such as network and temporal analysis, catering to a range of research needs and expertise levels.

PDF