Structured Data

The terms ’unstructured’, ‘semi-structured’, and ‘structured’ often come up when discussing about data analytics. Structured data is data that has been transformed into an understandable format, usually by following a nicely defined data model. The data model is able to map raw data into pre-defined fields. SQL databases, with all the rows and columns, is a great example of structured data. Structured data is oftentimes quantifiable. Semi-structured data is somewhere between structured data and unstructured data. It has some consistent characteristics, but it doesn’t have enough rigidity to fit into a relational data model. We can attach organizational properties to semi-structured data to make it more manageable. Unstructured data is data that shows up in its raw, unprocessed form. Qualitative data represents a good portion of the unstructured data. These data typically is hard to organize into pre-defined data models. NoSQL databases are more suited to manage unstructured data. In the context of the DIKW model, structured data & unstructured data need to come together in order to be extracted into useful information. In the case of a healthcare clinic, both quantitative data and qualitative data can come in handy. The unstructured, qualitative data is equally important for diagnosing as quantitative health measurements, if not more. Not everything about a person’s health condition is quantifiable, otherwise doctor’s jobs would become extinct. A doctor’s brain is like a supercomputer, it automatically processes any unstructured data, and then turns it into descriptive information, then extracts it into useable knowledge, and eventually into forward-looking wisdom.