Erality with the articles retrieved from the selected digital libraries, in the second stage a

Erality with the articles retrieved from the selected digital libraries, in the second stage a search and choice technique was applied in the inclusion/exclusion criteria, to determine which articles could be included in the final review.Appl. Sci. 2021, 11,five of3.1.1. Inclusion Criteria Below will be the inclusion criteria that have been utilized to pick the works analyzed and discussed within this review. 1. two. 3. 4. Research functions written in English. Study performs published in journals, conferences, or theses. Functions published from at the least 2005. Performs explicitly addressing the preprocessing or cleaning of occasion logs.Table 1. Query details and results (number or retrieved papers) in the chosen digital libraries.Search String (“filtering” OR “cleaning” OR “repairing” OR “clustering” OR “refinement” OR “preprocessing”) “event log” “process mining” (“filtering” OR “cleaning” OR “repairing” OR “clustering” OR “refinement” OR “preprocessing”) “trace” “process mining” (“ordered” OR “aligning”) “event log” “process mining” (“anomalous detection” OR “infrequent behavior” OR “noisy” OR “imperfection”) “event log” “process mining” (“anomalous detection” OR “infrequent behavior” OR “noisy” OR “imperfection”) “trace” “process mining” IEEE Xplore 110 Springer BMS-986094 Cancer Hyperlink 32,440 Science Direct63,13941,3852133.1.2. Exclusion Criteria The following are the criteria to discard analysis performs which are not of interest for this revision. 1. two. 3. Performs which might be not associated to method mining. Operates that don’t concentrate on particular domains in the field of procedure mining (sector, manufacturing); that’s, ad hoc techniques to get a offered domain. Operates that don’t involve evaluation and experimental results.After filtering by the queried subject and removing the duplicated retrieved papers, a total of 95 papers had been obtained and analyzed, thinking about the inclusion criteria. On this set, and just after applying the exclusion criteria, the result was a set of 70 papers. All of these are GS-626510 manufacturer incorporated in our qualitative evaluation. Figure 2 shows the distribution on the selected works (in ), from 2006 to 2020, primarily based on the year of publication (during a period of 3 years). Even though Figure two reveals a slow expanding tendency, it really is tough to decide if that tendency is going to be kept. As shown in Figure 3, 25 with the chosen works had been published in journals, 70 were reported at international conferences, plus the rest are theses. Performs formerly reported (2006008) have had great influence within the neighborhood, as Figure 4 reveals a total of about 2400 cites. Moreover, most current functions (2017020) reveal greater than 350 cites. Figure 5 shows a network of closely associated terms that describe the different addressed topics of event log preprocessing. This network was formed from the relationships identified amongst ideas incorporated inside the abstract of your surveyed functions. The colour intensity within the network nodes refers to much more common or abstract terms that contain other additional precise terms, all connected to describing the handling of preprocessing in approach mining. This network of terms might be beneficial for scientists to understand and organize the diversity of existing preprocessing approaches.Appl. Sci. 2021, 11,six ofPapersFigure 2. Publications, each and every 3 years of the surveyed papers.Figure three. Percentage of publications connected with distinctive surveyed sources in this survey.Figure 4. Citations, every single three years in the surveyed papers.Appl. Sci. 2021, 11,7 ofMissing Outlier Order Anomaly Log Profiles Granu.