A Gomsl Analysis Of Semi-Automated Data Entry

We used GOMSL (Goals, Operators, Methods, and Selection Rules Language) to per-form computational workflow analyses of two different data entry applications: a fully manual web form client used to update an enterprise-wide knowledge base (already in operational use), and an alternative prototype client that uses content extraction to semi-automate data entry. Our goal was to explore conditions that affect the speed of manual vs. semi-automated data entry, and to quantify expected difference in relative system efficiency across these conditions. We developed GOMSL models for major functionality in both systems and used GLEAN (GOMS Language Evaluation and Analysis) to simulate user interactions with representative data. Based on the results of these simulations, we quantified workflow costs, explored how costs vary across ranges of parameters, and developed overall estimates of relative system efficiency.