Chapter 4. Data Classification
Data classification means the classification of datasets into pre-defined classes. The aim of data classification is to identify datasets subject to common rules and operating methods. Classification can be utilised, for example:
- In data use policies, such as “data that is classified as something other than restricted can be primarily used in the forming of a City-level situation picture.”
- To improve the usability of data. By defining common metadata practices: the user can determine the original source, updates to and usability of data based on its metadata.
- In data architecture, such as “sensor data is stored primarily in a City-wide cloud environment.”
The processing of data with the help of data classes directly benefits data management as follows:
- Data processing becomes more efficient when it is steered with processes without the need to review the content of the data multiple times.
- Data management errors are reduced with the help of automated and systematic processing.
- Data security is increased, as the right to access data is determined based on its class instead of a separate process or discretion, for example.
- The work of the data utilisation working group becomes easier when the classes of data available for utilisation are unambiguously defined.
- Communication related to data utilisation becomes easier and precedents are easy to find based on standardised data classification, which increases transparency in the utilisation of data.
The utilisation of data in accordance with the City’s data vision necessitates data classification and the defining of associated rules. Information structures and systems must also support these types of metadata in a consistent manner. Since understanding data classification is vital to understanding the Data Strategy as a whole, the City’s current data classification model is presented in the following section. The classification model is divided into content-based classification and usage-based classification. Both of these perspectives are described in the following two subsections. Processing rules are examined in greater detail in Section 4.3.