Larity Level Agglomerative Timestamp Facts Bag of Activitivies Clustering Rules Discovery Conditional Probability Event Activity Alignment Occasion log Incorrect Procedure Model Incomplete Troubles Duplicated K-gram Model Infrequent Procedure MiningStatisticalTraceFilteringTechniques AutomaticManually ConformanceLaplace Smoothing Chaotic Activities Classification Pre-processing Entropy Embedded Supervised Understanding Patterns Distance Classifier Guidelines Generic Bayesian Apromore Euclidean Levenshtei RapidMiner TimeCleanser ProM Tools Automaton Metrics Structure Graph SequenceFigure 5. Summary of diverse closely GYKI 52466 web associated terms and their relations inside the information preprocessing domain in procedure mining.For the duration of the literature review, a content material study was performed. Within this study, we identified and classified the common and relevant traits identified in the surveyed papers. Table two outlines a general view plus a summary in the most significant characteristics (C1–techniques, C2–tools, C3–representation schemes, C4–imperfection sorts, C5–related tasks, and C6–types of information), which are described in greater detail in the subsequent sections.Table two. Most important qualities inside the reviewed studies.ID Characteristic Tactics Tools Representation schemes Imperfection sorts Description Two primary households of tactics: (1) transformation approaches and (two) detection and visualization procedures ProM, Disco, RapidProM, Celonis, Apromore, RapidMiner, Java application, preprocessing framework Sequences of events/traces or vectors, graphs, automatons Form-based occasion capture, inadvertent time travel, unanchored event, scattered occasion, elusive case, scattered case, collateral events, polluted label, distorted label, synonymous labels, homonymous label, timestamp granularity, unusual temporal ordering Two types: occasion abstraction and alignment Occasion label, timestamp, ID, cost, resource, added occasion payload[C1] [C2] [C3] [C4] [C5] [C6]Related tasks Types of information3.2. C1. Approaches Is there a way of grouping occasion log preprocessing strategies Distinct criteria could possibly lead to distinct taxonomies of information preprocessing methods inside the context of procedure mining. In the surveyed works, we organize the current occasion log preprocessing approaches, in two primary groups: transformation procedures and detection isualization methods. The principle classification criterion will be the approach followed by the preprocessing methods to clean the data, which involves identification, isolation, and reparation of errors. Figure 6 schematically shows a feasible taxonomy for the surveyed works. The proposed taxonomy organizes the diversity of existing preprocessing procedures and aids determine traits that they may have in common. Our grouping also serves to determine in which information top quality issues that specific forms of procedures are more appropriate to work with. The first ML-SA1 Biological Activity category consists of strategies that execute transfor-Appl. Sci. 2021, 11,eight ofmations in the occasion log in order to appropriate the imperfect behaviors (missing, irrelevant, duplicate data, and so forth.), prior to applying a method mining algorithm. The second category is comprised of techniques to detect or diagnose imperfections in an occasion log. Whilst the second category of techniques only detect prospective issues associated to data quality inside the occasion log, the methods within the 1st category directly right the imperfections found inside the event log.Filtering-Based Transformation methods Event log preprocess.