Designing data collection forms for unexpected data: The case of Suspicious Activity Reports for financial fraud detection

SARs (Suspicious Activities Reports) are the forms that financial institutions use to report any suspected activity. Literature shows that the user interface will affect the data provided by users, because of many psychological reasons like different biases and cognitive shortcuts. Data collected is typically analyzed with machine learning algorithms. Different user interfaces provide different levels of quality for this data, which will influence the outcome(s) of the machine learning techniques.

This research will help Verafin design better user interface(s) for their input form for reporting suspicious activities, which should then provide them with more complete data for analysis. For the purposes of this research, ‘more complete data’ is achieved by collected more diverse and granular data, achieved through the improved input forms. Diverse data refers broader diversity of information collected while granular data refers to the depth of a specific information set.

Faculty Supervisor:

Jeffrey Parsons

Student:

Partner:

NASDAQ Canada Inc

Discipline:

Business

Sector:

Information and Communications Technology; Finance and Insurance

University:

Memorial University of Newfoundland

Program: