ISD Glossary
ISD’s Glossary provides clear definitions of key terms related to extremism, hate, information operations and tech accountability. This page is continually reviewed to ensure accuracy.
Salafism
a form of Sunni Islam whose followers advocate a return to the practices of the first three generations of Muslims (the salaf or ‘ancestors’) who lived immediately after the prophet Mohammed. Within Salafism, there are different currents, which differ significantly in their interpretations of the holy scriptures of Islam and their implications for political action. Salafis are often subdivided into quietist Salafists, who reject political activism, political Salafis, who are actively engaged in transforming society according to their ideological ideas, and Salafi-jihadists, who use violence to implement a Salafi interpretation of Islamic law
Seeds
Seeds act as pre-defined starting points for extensive data collection. Lists of Seeds may be made up of keywords, internet addresses (URLs) or social media channels (accounts).
Shallowfake
Shallowfake (sometimes referred to as ‘Cheapfake’) refers to media that has been altered or manipulated in a relatively simple way (as opposed to ‘deepfakes’ which involve more sophisticated techniques like AI and deep learning).
Spree Posting
The same individual Facebook account posting identical content from Link Source entities to two or more Community groups or pages in under one minute.
Systematic search methods
Systematic search methods use technology to extract large amounts of data and metadata directly from online platforms. Data might include, for example, the content of online text, connections between online accounts and metadata, such as the time or geographical location of posts. Many social media platforms also make data easier to access by providing APIs that allow researchers to directly access various forms of data from platforms without needing to build their own code from scratch. The development of AI-based approaches has also allowed for ever more sophisticated analysis methods; for example, natural language processing (NLP) is increasingly used to detect trends, sentiments and entities mentioned across vast quantities of online text.
