In many taxonomies, topics and concepts are mixed and matched. This structure is beneficial in some instances, and in others, it may prove to be challenging.
Mixing and matching topics and concepts in taxonomies intended for relating interests of users for recommendation systems or searching for products, services, or content, do not create challenges. For taxonomies that are used as the basis for ontologies or machine learning and other kinds of reasoning, conflating topics and concepts can be problematic.
In most instances, the vagueness between topic and concept are found nearer the top level or top terms. Often, top terms are generally used as topical buckets and are not indexed. These terms serve as navigational starting points. In a strict sense, this does not exclude top terms from the requirements for taxonomy relationships.
However, it is ubiquitous for top terms to be used as topics and not concepts. This could prove challenging when modeling concepts and developing an ontological structure to define the semantic relationship between concepts.
In sum, taxonomies relating to the interests of users can be organized with fewer problems. However, in taxonomies used for reasoning or as the basis for ontologies, a more considered approach to concept-relationship modeling is needed.
Click here to read the complete article published by the Synaptica.
Please give your feedback on this article or share a similar story for publishing by clicking here.
