Reeglikaeve tööriistaga RuM
Files
Date
2019
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Protsessikaeve on äriprotsesside juhtimise (BPM) valdkonda kuuluv andmeteaduse tehnika, mille lähtepunktiks on protsessi täitmise logide analüüsimine. Hetkel kasutatakse protsessikaeve jaoks põhiliselt kahte töölauarakendust: Disco ja ProM. Disco sobib protseduuriliseks protsesside tuvastamiseks ja on intuitiivse kasutusliidesega, aga ei toeta deklaratiivseid mudeleid. ProM toetab mitmeid erinevaid deklaratiivse protsessikaeve tehnikaid, aga on ebaintuitiivsema kasutusliidesega kui Disco. Antud magistritöö eesmärk on välja pakkuda uus deklaratiivsele protsessianalüütikale suunatud tööriist RuM, mis pakub ühtlasi intuitiivset kasutajakogemust. RuM pakub sobiva keskkonna protsesside väljaselgitamise, vastavuskontrolli ja logide genereerimise meetodite jaoks. Kõik RuM poolt toetatud meetodid tuginevad deklaratiivsetel mudelitel ja on kasutaja poolt seadistatavad. Meetodite tulemusi on võimalik lihtsustamise eesmärgil järeltöödelda ja failisüsteemi salvestada. RuM töölauarakenduse hindamiseks on teostatud kasutatavuse testimine, milles osales viis deklaratiivse protsessianalüütika valdkonna eksperti.
Process Mining is a data science practice in Business Process Management (BPM). It is based on the analysis of process execution logs. Nowadays, ProM and Disco are two mainstream desktop applications for process mining. Disco can be used for procedural process discovery with intuitive user experience. However, it does not support declarative models. ProM gives support for a wealth of techniques for declarative process mining. However, its user interface is not so intuitive as the one of Disco. In the light of this, this Master’s Thesis is focused on proposing RuM, an application for declarative process analysis with an intuitive user experience.The application presents an environment for a set of process discovery, conformance checking and log generation methods based on declarative models. The methods can be configured with various options and the results can be post-processed to simplify and store them in the file system. The tool has been evaluated through a usability test involving 5 experts on declarative process analysis.
Process Mining is a data science practice in Business Process Management (BPM). It is based on the analysis of process execution logs. Nowadays, ProM and Disco are two mainstream desktop applications for process mining. Disco can be used for procedural process discovery with intuitive user experience. However, it does not support declarative models. ProM gives support for a wealth of techniques for declarative process mining. However, its user interface is not so intuitive as the one of Disco. In the light of this, this Master’s Thesis is focused on proposing RuM, an application for declarative process analysis with an intuitive user experience.The application presents an environment for a set of process discovery, conformance checking and log generation methods based on declarative models. The methods can be configured with various options and the results can be post-processed to simplify and store them in the file system. The tool has been evaluated through a usability test involving 5 experts on declarative process analysis.