2.4. Employing a Parametric Model for Analytic Provenance
When exploring a dataset using tools like those presented in this special issue, analysts leave behind not only a trail of interactions but also a trail of insights, decisions, and discoveries.
This trail of thinking, or “analytic provenance,” is often lost, because few systems have means to effectively record analyst interactions. The approach presented here by Chen et al. includes a new language and model to capture the analytic process for later exploration and reuse.
Their approach is evaluated using a number of openly available datasets, and it shows how a user can refer to their own analysis history to inform future analysis directions. As the need for scalable analytic provenance continues to grow, work in this direction will serve as a basis for capturing analytics in many scientific domains.
2.4. Employing a Parametric Model for Analytic Provenance
When exploring a dataset using tools like those presented in this special issue, analysts leave behind not only a trail of interactions but also a trail of insights, decisions, and discoveries.
This trail of thinking, or “analytic provenance,” is often lost, because few systems have means to effectively record analyst interactions. The approach presented here by Chen et al. includes a new language and model to capture the analytic process for later exploration and reuse.
Their approach is evaluated using a number of openly available datasets, and it shows how a user can refer to their own analysis history to inform future analysis directions. As the need for scalable analytic provenance continues to grow, work in this direction will serve as a basis for capturing analytics in many scientific domains.
การแปล กรุณารอสักครู่..
