ACM has published a de facto standard taxonomy for classifying and indexing computing literature and researchers’ areas of expertise since the 1960s. The CCS underwent a major overhaul in 1982 with substantive updates in 1998 and 2012.
The 2012 CCS was created by group of 120 ACM volunteers, a third of them ACM Fellows, who collaborated with ACM Staff and with Semedica, a Division of Silverchair.
The Update Project was led by Professor Zvi Kedem of NYU who served as its Editor-in-Chief, working closely with Bernard Rous, ACM Director of Publications.
The 2012 ACM Computing Classification System has been developed as a poly-hierarchical ontology that can be utilized in semantic web applications. It relies on a semantic vocabulary as the single source of categories and concepts that reflect the state of the art of the computing discipline and is receptive to structural change as it evolves in the future.
ACM has provided tools to facilitate the application of 2012 CCS categories to forthcoming papers.
URL OF KNOWLEDGE MODEL:
The full CCS classification tree is freely available for educational and research purposes in these downloadable formats: SKOS (xml), Word, and HTML. In the ACM Digital Library, the CCS is presented in a visual display format that facilitates navigation and feedback. The full CCS classification tree is also viewable as a flat file in the Digital Library.
- Owner: ACM
- Developer: ACM-Semedica
- Technical Contact: Bernard Rous (rous [at] @hq.acm.org)
- License Contact: Deborah Cotton (cotton [at] hq.acm.org)
Various libraries, companies, and publishers such as Springer, IEEE, and Emerald have made use of the ACM CCS.
HOW IS THIS KM APPLIED?
- ACM articles are generally indexed manually. Auto-tagging software is being evaluated.
- Index terms are applied by ACM authors and by professional indexers.
- A map of the 1998 CCS to the 2012 version has been built and automatically run against all articles in the ACM Digital Library. Both the 1998 and 2012 sets of concepts are available on Citation Pages of all indexed articles at this time.
- In displays of Article Citation Page under “Index Terms” tab. (See: 1145/1963190.1963191)
- In tag cloud displays of topics covered by specific publications (See: http://0-dl.acm.org.libcat.lafayette.edu/pub.cfm?id=J401) or Special Interest Groups (See: http://0-dl.acm.org.libcat.lafayette.edu/sig.cfm?id=SP923) or Institutional Profile pages (See: http://0-dl.acm.org.libcat.lafayette.edu/inst_page.cfm?id=60022148)
- CCS subjects are included in the index for Simple Search
- CCS subjects are directly searchable in Advanced Search. See left bottom of page: http://0-dl.acm.org.libcat.lafayette.edu/advsearch.cfm
- CCS subjects are themselves clickable to return papers indexed by those concepts
- CCS subjects are currently displayed on Author Profiles pages under Subject Areas. (See: http://0-dl.acm.org.libcat.lafayette.edu/author_page.cfm?id=81100246710)
DESCRIPTION OF FUTURE/POTENTIAL USE CASES OF THE KM (NOT YET REALIZED)
ACM is building a community and people-oriented search where the primary objects returned are experts, their attributes, and their contextual relationships. Published works will become attributes of the author (rather than the primary object of bibliographic search where authors are attributes of the published.) One clear use of the CCS in this new facility is the direct and immediate ability to discover people who are expert in one of the defined subject areas; to order them by their impact in that area; and display their working relationships.
Additionally, the 2012 CCS is only partially deployed in the ACM Digital Library today. Sections of the ACM DL still rely on the 1998 version of the CCS.
WHAT ARE THE MAIN GOALS FOR USING THIS KM?
To enable efficient and precise discovery and exploration of topics
The ACM CCS is a hierarchical taxonomy. It is designed to provide a cognitive map of the computing space from the most general subject areas to the most specific topics.
RATIONALE FOR KM VS OTHER MEANS OF SEARCHING AND BROWSING?
When speaking of taxonomies in computer science circles, the question is often asked “Why bother? Taxonomies are antiquated; Google renders them unnecessary; and the ACM CCS is not used by anyone other than authors who are required to index their ACM articles with it, much to their irritation.”
There are certainly camps within the Information Retrieval community on this issue; one tends to dismiss the usefulness of taxonomies in today’s world while another sees them as powerful and with growing application. In the scientific, technical, and medical (STM) publishing domain, the taxonomic approach to semantic classification is booming — with publishers using taxonomy to allow users to cross-cut content topically, increasing application usage.
CCS searches in the ACM Digital Library have been a relatively small percentage of total searches. Yet some part of the user community finds the CCS very useful in search. Despite the fact that direct searching on CCS subject categories is not highly visible (being found rather cryptically at the bottom left of the Advanced Search page (http://0-portal.acm.org.libcat.lafayette.edu/advsearch.cfm), the annual number of CCS searches launched in the ACM DL is still about half a million. Adjustments in the Digital Library user interface to promote the CCS as a retrieval tool should multiply this number many fold.
Many scholarly publishers in the scientific, technical, and medical fields are making use of taxonomic classification to create topic collections and virtual journals that dynamically rebuild as new content is added.
ACM efforts to derive topical visualizations from our full-text index proved inferior to those derived from taxonomic terms. Using author-supplied keywords (which are not selected from a controlled vocabulary) was somewhat better but also proved inferior.
Google-type searching appears best suited to directed searches where the user knows exactly what he is looking for. Google supplies almost total recall and the user supplies the precision. For more general subject exploration and discovery purposes, the searches do not work quite as well. And the page-ranking algorithm itself skews results by defining relevance in terms of popularity. Finally, it should be noted that Google is well aware of how its indexes are enhanced by structured, fielded data leading to improved precision in searches; that is why Google Scholar at least has tried to make arrangements with all the publishers whose sites it crawls to provide specific standard meta-tagging. Most publishers, including ACM, have complied in their indexing agreements with Google Scholar.
Lastly, a robust up-to-date taxonomy provides a cognitive map of the discipline. This in itself can be useful in understanding what computer science is all about; where a specific area of concentration fits within the broader discipline; and in development of curricula.
IS THE KM BEING ACTIVELY DEVELOPED?
Yes. The ACM CCS itself is evolving along with its deployment.
For commercial use, please write cotton [at] hq.acm.org