Process Mining：Data science in action 學習筆記

視訊地址：https://www.bilibili.com/video/BV1xB4y1w7RF

P2 1-1:

4 V’s of Big Data

• VOLUME 量大
• VELOCITY 資料增長速度
• VARIETY 種類多
• VERACITY 真實性

Is on the interplay between event data and processes, and process models.

Process mining use cases

P3 1-2:

Three types of mining :

• Process discovery 流程發現
• Performance checking 一致性檢查
• Enhancement 增強

It is not just about collecting data , it's about analyzing processes.
That is one of the key things.

Event Log Table ：

Every line, corresponds to an event .

Three types of relationships between models and event data:

• Play-Out

• Play-In

• Replay

Play-out is about the classical use of process models not involing any event data.

Play-in corresponds to discovery.

Play Out:

Single execution for single trace:

Many execution for many traces:

P4 1-3:

Process Discovery

For example:

We can see a sentence and compare that to atrace in an event log, and we can think of the language as a process model.

P5 1-4:

decision tree: