Relating taxonomies with regulations

Increasingly, taxonomies are being developed for a wide variety of industrial domains and specific applications within those domains. These industry or application specific taxonomies attempt to represent the vocabularies commonly used by the practitioners. These formal representations have the potential to automate information retrieval, facilitate interoperability and improve decision making. Decisions made must comply with existing government regulations and codes of practices, which are not always known to the industry practitioners. Although regulations and codes are now in digital forms and are often available online, it remains difficult to search for relevant regulatory information that are applicable to particular decisions. As industry practitioners, unlike legal practitioners, are familiar with one or more industry-specific taxonomies but not necessarily regulatory organization systems, it would be desirable to relate regulations with existing industry-specific taxonomies. The mapping from a single taxonomy to a single regulation is a trivial keyword matching task. In this paper, we examine techniques to map a single taxonomy to multiple regulations, as well as to map multiple taxonomies to a single regulation. Those techniques include cosine similarity, Jaccard coefficient and market-basket analysis. These techniques provide a metric that measures the similarity between concepts from different taxonomies. Preliminary evaluations of the three metrics are performed using examples from the building industry. These examples illustrate the potential regulatory benefits from the mapping between various taxonomies and regulations.