Data Mining in Networks
D. Jensen (2002). Data mining in networks. Invited talk to the Roundtable on Social and Behavior Sciences and Terrorism. National Research Council, Division of Behavioral and Social Sciences and Education, Committee on Law and Justice. Washington, DC. December 11.
- Data mining is the process of constructing predictive models from large and complex databases. Counter-terrorism data present unique challenges for data mining algorithms, including the relational and heterogeneous structure of the data, the fragmentary nature of the data, and the presence of relational autocorrelation. New research is addressing challenges, but applications are still 3-5 years away. Many design options exist with widely differing uses and impacts. Current popular understanding of data mining for counter-terrorism is suffering from a few widely disseminated myths, including the necessity of a single large database and a vast new program of data collection.
- A web-accessible version of this talk is available.