DatasetType

Summary

Enumeration of dataset types.

Description

Describes the different structures of data within a given dataset. A dataset can have multiple types of data, or even a single type of data but still match multiple types, for example sensor data could also be timeseries or labeled image data could also be considered categorical.

Metadata

https://spdx.org/rdf/3.0.1/terms/Dataset/DatasetType

Name	DatasetType

Entries

• audio: data is audio based, such as a collection of music from the 80s.
• categorical: data that is classified into a discrete number of categories, such as the eye color of a population of people.
• graph: data is in the form of a graph where entries are somehow related to each other through edges, such a social network of friends.
• image: data is a collection of images such as pictures of animals.
• noAssertion: data type is not known.
• numeric: data consists only of numeric entries.
• other: data is of a type not included in this list.
• sensor: data is recorded from a physical sensor, such as a thermometer reading or biometric device.
• structured: data is stored in tabular format or retrieved from a relational database.
• syntactic: data describes the syntax or semantics of a language or text, such as a parse tree used for natural language processing.
• text: data consists of unstructured text, such as a book, Wikipedia article (without images), or transcript.
• timeseries: data is recorded in an ordered sequence of timestamped entries, such as the price of a stock over the course of a day.
• timestamp: data is recorded with a timestamp for each entry, but not necessarily ordered or at specific intervals, such as when a taxi ride starts and ends.
• video: data is video based, such as a collection of movie clips featuring Tom Hanks.