This is an archive of an inactive wiki and cannot be modified.

The Semantic Web Environmental Directory (SWED)

Contact e-mail: a.j.miles #


General purpose and services to the end user

The Semantic Web Environmental Directory (SWED) is a web-based directory of organisations and projects in the United Kingdom, whose business has something to do with the natural environment - an online "Who's who in the environment".

SWED is a proof of concept demonstration, aiming to create a sustainable and scalable web-based system for building and maintaining Community Information Resources (portals). The system is intended to overcome many of the limitations and problems with more traditional and current approaches.

Key goals of the SWED project include:

These goals are made possible by using a Semantic Web based approach. This approach means that the directory member organisations publish, own and control their own information. This information is then harvested and collated by SWED to produce a directory Web site.

Essentially the SWED directory provides a view of the data, that is brought together from across the Web. This 'self publishing' or data harvesting approach means that others can also harvest and collate the information and provide different views (e.g. a specialist directory containing only a small subset of the organisations) and enrich the information by adding their own additional information (e.g. what specialist services or resources organisations provide).

Functionality examples

The main purpose of SWED is to enable visitors to the site to find organisations and projects matching particular characteristics. For example, a visitor might be looking for projects whose topic of interest is the welfare of captive animals; or organisations that are not-for-profit; or organisations whose activities include forestry and whose operational area includes the south west of England.

The SWED interface provides search and browse functionality. When browsing the directory, the user is initially presented with six groups of options, entitled, "Topic of interest", "Organisation type", "Activity", "Project type", "Operational area" and "Name". Each of these groups corresponds to a particular characteristic of an environmental organisation or project. By selecting an option from within one of these groups (e.g. by selecting "Archaelogy" from within the "Topic of interest" group) the user creates a "filter", which is used to present a list of results corresponding to only those organisations or projects that match for the given characteristic (This type of functionality is sometimes known as "faceted browsing").

Special strategies involved in the processing of user actions

Once presented with a set of results, the user may "refine" or "narrow down" their search in one of two ways:

(1) If there are more specific options for the current filter then one of these may be chosen, which updates the filter and causes a new, smaller, set of results to be displayed. E.g. if the user begins browsing by selecting "Animal Welfare" from the "Topic of interest" group, the user may then choose one of the more specific options "Captive animals (Welfare of)" or "Wild Animals (Welfare of)" from within the same group.

(2) Another option from within any of the groups may be added to the current filter. E.g. A user might begin browsing by selecting "Not for Profit" from within the "Organisation type" group, and then select "Campaigning and Lobbying" from within the "Activity" group. Or E.g. A user might begin browsing by selecting "Built Environment" from the "Topic of interest" group and then select "Biodiversity" also from the "Topic of interest" group.

Note that options may be iteratively refined, added or removed from the current filter within a browsing session, provided that the action will yield at least one result.

Note also that results are displayed if they match the current filter, or if they match any of the possible specialisations of the current filter. For example, an organisation that describes its topic of interest as "Fossil Fuels" will appear in the results for a filter on the option "Energy" in the "Topic of interest" group. Whether a result matches directly or indirectly does not affect the ranking of results.

Integration between vocabulary-linked functions and other application functions

Browsing may be combined with free text searching. I.e. at any time a user may enter a free text query, and search for textual matches in the entire directory or only within the currently selected results.

Additional references



There are 5 distinct controlled vocabularies used within the SWED system, one vocabulary for describing each of the "Topic of interest", "Organisation type", "Activity", "Project type" and "Operational area" characteristics. (The "Name" characteristic is uncontrolled).

Vocabulary extract

Some examples of the hierarchical structure in the "Topic of interest" vocabulary are given below, with indentations indicating a hierarchical relationship.

   Biodiversity surveys
   Trade Unions
   Fossil Fuels
   Biotechnology and Genetic Engineering
     Genetically Modified Food

Structure explanation

The controlled vocabularies for the first four of these characteristics all conform to the following structural characteristics.

Each of the meaningful units (i.e. concepts) of a vocabulary is identified by a URI reference (e.g.>), has a single human-readable label in English (e.g. "Developing World") and may have a scope note or general comment which is a string of text written in English.

Hierarchical relationships may be asserted between meaningful units, and any one unit may have more than one hierarchy parent (i.e. the structure may be poly-hierarchical). Hierarchical relationships do not uniformly correspond to a single logical relationship, such as class/instance or class subsumption.

Machine-readable representation of the vocabulary

The RDF representation of the vocabularies, as currently used by the SWED portal, may be obtained from the SWED Technical Resources page,