Observing process discovery technique using alpha and heuristic miner - case study on sepsis event log

Therawan, Prapta Arya (2023) Observing process discovery technique using alpha and heuristic miner - case study on sepsis event log. Bachelor thesis, Universitas Pelita Harapan.

	Text (Title) Title.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (110kB)
	Text (Abstract) Abstract.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (299kB)
	Text (ToC) ToC.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (501kB)
	Text (Chapter1) Chapter1.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (519kB)
	Text (Chapter2) Chapter2.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB)
	Text (Chapter3) Chapter3.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (536kB)
	Text (Chapter4) Chapter4.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (2MB)
	Text (Chapter5) Chapter5.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (247kB)
	Text (Bibliography) Bibliography.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (238kB)
	Text (Appendices) Appendices.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (7MB)

Abstract

This thesis aims to exploring two commonly used Process Discovery techniques in Process Mining: Alpha Miner and Heuristic Miner, applied to the Sepsis Cases – Event Log dataset. The primary objective is to compare and analyze the significance of technique’s parameter used to process event data logs. The problem definition revolves around the necessity for a more in-depth comprehension of Process Discovery techniques and their parameter’s significance to the output of those techniques. This research is based on theoretical framework focuses on process mining steps, specifically on Process Discovery stage within the context of event data logs. The selected techniques are Alpha Miner and Heuristic Miner, both of which are implemented to extract valuable information from event logs. The anticipated outcome of this research will be a comprehensive understanding of the significance of parameter values associated with each technique. This understanding will contribute to the broader knowledge of the impact of the parameter variations on the outcomes of both Alpha Miner and Heuristic Miner.In conclusion, the version used for Alpha Miner affected the behavior and the structure of the Petri Nets. On the other hand, values used in Dependency and Frequency threshold significantly affects the nodes and the arcs that represents relation between nodes of the Causal Graph. In Dependency Testing, a parameter value of 0 or 0.1 yields the most intricate Causal Graph, contrasting with the simplest graph at parameter value 1. Conversely, in Frequency Testing, a parameter value of 0 results in the most complex graph, while 1 produces the simplest. / Skripsi ini bertujuan untuk melakukan eksplorasi dua teknik Process Discovery yang umum digunakan dalam Process Mining: Alpha Miner dan Heuristic Miner, yang diterapkan pada dataset Sepsis Cases – Event Log. Tujuan utama adalah membandingkan dan menganalisis signifikansi parameter teknik yang digunakan untu memproses event log. Definisi masalah berkaitan dengan kebutuhan akan pemahaman yang lebih mendalam mengenai teknik Process Discovery dan signifikansi parameter mereka terhadap hasil teknik tersebut. Penelitian ini didasarkan pada kerangka teoritis yang berfokus pada langkah-langkah process mining, khususnya pada tahap Process Discovery dalam konteks sebuah event log. Teknik yang dipilih adalah Alpha Miner dan Heuristic Miner, keduanya diimplementasikan untuk mengekstrak informasi dari event log. Hasil yang diantisipasi dari penelitian ini adalah pemahaman menyeluruh mengenai signifikansi nilai parameter yang terkait dengan masing-masing teknik. Pemahaman ini akan berkontribusi pada pengetahuan lebih luas mengenai dampak variasi parameter terhadap hasil dari Alpha Miner dan Heuristic Miner. Sebagai kesimpulan, versi yang digunakan untuk Alpha Miner memengaruhi perilaku dan struktur dari Petri Nets. Di sisi lain, nilai yang digunakan dalam parameter Dependency and Frequency secara signifikan mempengaruhi node dan arc yang mewakili hubungan antar node dari Causal Graph. Dalam Dependency Testing, nilai parameter 0 atau 0.1 menghasilkan Causal Graph paling rumit, berbeda dengan grafik paling sederhana pada nilai parameter 1. Sebaliknya, dalam Frequency Testing, nilai parameter 0 menghasilkan grafik paling kompleks, sementara 1 menghasilkan yang paling sederhana.

Actions (login required)

View Item