Primary and secondary sources often contain information that signifies a complex real world. Relationships between entities are a key part of the real world. So if we want to be able to ask questions where the answers involve more than simply counting data, or finding data, we need to structure our sources using an ontology.
For example, the Old Bailey Proceedings Online contains the accounts of nearly 200,000 criminal trials held between 1674 and 1913. Using taxonomies to tag entities such as the names of defendants, the categories of crimes they committed, and the sentences they received will enable us to find and count how many people were tried for crimes, how many types of crimes there were, and how many types of sentences were given. However, an ontology will enable us to relate defendants to the specific types of crime they are accused of and the sentences they received. We can then ask more sophisticated questions such as:
- What types of crimes receive what types of sentences?
- Are there types of crimes or sentences that tend to be gender, age or class specific (assuming defendants are given attributes such as their gender, age and occupation)?
- Which types of crime are conducted by individuals and which are conducted by more than one individual?
- Does the severity of sentencing increase or decrease for the same type of crime over time?
- Based on what the data tells us about how defendants are sentenced for a specific type of crime, can we predict the sentence that John Doe will receive for having been found guilty of committing the same crime?
URL for this page
https://www.dhi.ac.uk/books/ontology-guide/understanding-what-data-means