Towards an Empirical Evaluation of Imperative and Declarative Process Mining

TY - GEN

T1 - Towards an Empirical Evaluation of Imperative and Declarative Process Mining

AU - Back, Christoffer Olling

AU - Debois, Søren

AU - Slaats, Tijs

PY - 2018

Y1 - 2018

N2 - Process modelling notations fall in two broad categories: declarative notations, which specify the rules governing a process; and imperative notations, which specify the flows admitted by a process. We outline an empirical approach to addressing the question of whether certain process logs are better suited for mining to imperative than declarative notations. We plan to attack this question by applying a flagship imperative and declarative miner to a standard collection of process logs, then evaluate the quality of the output models w.r.t. the standard model metrics of precision and generalisation. This approach requires perfect fitness of the output model, which substantially narrows the field of available miners; possible candidates include Inductive Miner and MINERful. With the metrics in hand, we propose to statistically evaluate the hypotheses that (1) one miner consistently outperforms the other on one of the metrics, and (2) there exist subsets of logs more suitable for imperative respectively declarative mining.

AB - Process modelling notations fall in two broad categories: declarative notations, which specify the rules governing a process; and imperative notations, which specify the flows admitted by a process. We outline an empirical approach to addressing the question of whether certain process logs are better suited for mining to imperative than declarative notations. We plan to attack this question by applying a flagship imperative and declarative miner to a standard collection of process logs, then evaluate the quality of the output models w.r.t. the standard model metrics of precision and generalisation. This approach requires perfect fitness of the output model, which substantially narrows the field of available miners; possible candidates include Inductive Miner and MINERful. With the metrics in hand, we propose to statistically evaluate the hypotheses that (1) one miner consistently outperforms the other on one of the metrics, and (2) there exist subsets of logs more suitable for imperative respectively declarative mining.

U2 - 10.1007/978-3-030-01391-2_24

DO - 10.1007/978-3-030-01391-2_24

M3 - Article in proceedings

SN - 978-3-030-01390-5

T3 - Lecture Notes in Computer Science

SP - 191

EP - 198

BT - Advances in Conceptual Modelling

PB - Springer

T2 - 37th International Conference on Conceptual Modeling

Y2 - 22 October 2018 through 25 October 2018

ER -