**3. The data science process**

According to Saltz, most data science processes focus on the tasks that need to be completed in data science such as the techniques to acquire and analyze data [2]. Saltz analyzed different data science approaches and found that most outlined the steps as data acquisition, cleansing, transformation, integration, modeling, analysis, and deployment [2]. The data science approaches are task-oriented, and no real evolution of the process had occurred since the cross-industry standard process for data mining (CRISP-DM) was introduced in the 1990s [2].

#### **3.1 CRISP-DM**

The most commonly used data mining process is CRISP-DM which is a process that conceptually described the stages used in data mining. Originally created to support data mining projects, it has been adapted by data scientists. There are six stages in the CRISP-DM process which are presented sequentially, but iteration is expected:

