Developing Corporate Taxonomies for Knowledge Auditability: A Framework for Good Practices
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
30 Knowl. Org. 35(2008)No.1
R. S. Sharma, Sch. Foo, M. A. Morales-Arroyo. Developing Corporate Taxonomies for Knowledge Auditability
Developing Corporate Taxonomies
for Knowledge Auditability:
A Framework for Good Practices†
Ravi S. Sharma,* Schubert Foo**, and Miguel Morales-Arroyo***
Nanyang Technological University, Wee Kim Wee School of Communication
& Information, 31 Nanyang Link, Singapore 637718,
*, **, ***
Ravi S. Sharma is Associate Professor at the Wee Kim Wee School of Communication and Information
at the Nanyang Technological University. He was concurrently founding Director (India Strategy) at
NTU for a 2-year term. Ravi had earlier been Asean Communications Industry Principal at IBM
Global Services and Director of the Multimedia Competency Centre of Deutsche Telekom Asia.
Ravi’s teaching, consulting and research interests are in multimedia applications, services and strate-
gies. Ravi received his PhD in engineering from the University of Waterloo and is a Chartered Engi-
neer (UK) and a Senior Member of the IEEE. He co-authored a book Knowledge management: tools
and techniques with Schubert Foo and Alton Chua.
Schubert Foo is Professor and Associate Dean, College of Humanities, Arts and Social Sciences, Nan-
yang Technological University, Singapore. He received his B.Sc.(Hons), M.B.A. and Ph.D. from the
University of Strathclyde, UK. He has published in the areas of multimedia technology, Internet tech-
nology, multilingual information retrieval, digital libraries and knowledge management. He co-
authored a book Knowledge management: tools and techniques with Ravi Sharma and Alton Chua, and
co-edited a book Social information retrieval systems: emerging technologies and applications for search-
ing the Web effectively with Dion Goh in 2007.
Miguel A. Morales-Arroyo received his Interdisciplinary PhD as a Fullbright Scholar from the Univer-
sity of North Texas. His master’s degree in Systems Engineering and undergraduate studies in Engi-
neering were from the National University of Mexico. He is currently a Research Fellow working the
area of business models for Interactive Digital Media at the Nanyang Technological University in Sin-
gapore. His research interests are related to information management studies, where information,
knowledge, and technology are essential to the decision-making process.
†
Acknowledgements: The findings reported in this article are part of the on-going efforts of an in-
formal, irreverent group within the Wee Kim Wee School of Communication & Information at the
Nanyang Technological University in Singapore known as the knowledge research factory. The au-
thors are grateful to the funding agency within NTU as well as to their graduate students–Varun Jain,
Ekundayo Samuel, Vironica, Mervyn Chia and Angela Wu–for the active engagement. Many thanks are also due to the partici-
pating vendors for their useful insights and feedback. The authors gratefully acknowledge the guidance of the editorial staff
and the anonymous reviewers in helping to improve the quality of the submission.
Sharma, Ravi S., Foo, Sch., and Morales-Arroyo, Miguel A. Developing Corporate Taxonomies for Knowledge Auditability.
A Framework for Good Practices. Knowledge Organization, 35(1), 30-46. 40 references.
ABSTRACT: The organisation of knowledge for exploitation and re-use in the modern enterprise is often a most perplexing
challenge. The entire knowledge management life-cycle (for example – create, capture, organize, store, search, and transfer) is
https://doi.org/10.5771/0943-7444-2008-1-30
Generiert durch IP '46.4.80.155', am 20.01.2021, 04:51:15.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.Knowl. Org. 35(2008)No.1 31
R. S. Sharma, Sch. Foo, M. A. Morales-Arroyo. Developing Corporate Taxonomies for Knowledge Auditability
impacted by the organisation of intellectual capital into a corporate taxonomy or at the least a knowledge map (often incor-
rectly used interchangeably). Determining the extent to which such an objective is achieved is the focus of what is known as a
knowledge audit. In this practice-oriented article, the authors review the fundamentals of creating a taxonomy, the use of meta-
data in a necessary process known as classification and the role of expertise locators where the knowledge is not explicit but re-
sides within experts in the form of tacit knowledge. The authors conclude with a framework for developing a corporate taxon-
omy and how such a project may be executed. The conceptual contribution of this article is the postulation that corporate tax-
onomies that are designed to facilitate knowledge audits lead to greater organizational impact.
1. Organising corporate knowledge The White Paper concluded that knowledge workers
need unified, universal access to all information, but
The organisation of knowledge resources is hardly a they only need that subset of the information base
novel undertaking. Many in the western world (cf. that actually solves the problem at hand and the req-
Woods 2004) attribute to Aristotle the first attempt uisite expertise or skill that would apply it. It identi-
at information organization, to the Swedish scientist fied the search costs of tediously retrieving required
Linnaeus the first system for categorising the natural information, the cost of repeated knowledge creation
world, and to the Melvil Dewey the first library cata- (i.e. not re-using existing knowledge components),
loguing scheme of medical and scientific knowledge. and the opportunity cost of not applying known
However, the ancient civilizations within China, In- knowledge as “the high cost of not finding informa-
dia and the Mid-East in fact organized their knowl- tion.”
edge, particularly related to philosophy, government Knowledge in the modern organization hence has
and medicine, carefully for the purpose of transfer elements of variety and is incongruous and heteroge-
and re-use. The ancient libraries of Alexandria (circa. neous in nature in terms of creation, storage and re-
2000 B.C.E.) which comprised world-class methods use. In academic research as well as trade forums, the
in their time for collection, storage and retrieval, we- understanding of what constitutes knowledge is often
re predicated on the realization that a system for or- debated because of the multidimensionality associ-
ganizing knowledge was the key to understanding an ated with it. Only when the knowledge is captured
existing body of knowledge and its repeated exploi- and organised into proper formats can it be made ac-
tation. Today’s knowledge-driven economy demands cessible and put to further use. In effect, capturing
such a strategy more than ever. From public archives knowledge is of little use if it is not stored in such a
and libraries to corporate repositories and individual way that it can be understood, indexed, accessed eas-
collections, knowledge that is not organized is often ily, cross-referenced, searched, linked, and generally
rendered worthless by the sheer velocity of business manipulated for maximum benefit of all members of
decisions that need to be made (Nonaka and Takeu- an enterprise. Hence the organisation of knowledge
chi 1995; Davenport and Prusak 1998; Senge 1999). plays a critical role throughout the knowledge cycle.
In other words, the entire study of organising know- One aspect of an organisation’s intellectual capital
ledge into a systematic classification of hierarchical is collective knowledge, which can be viewed in
categories, which may be labelled and subsequently terms of information within the context of the or-
searched, is all about fulfilling the cliché “knowing ganization. It first involves the process of acquisition
what we know”. A corporate taxonomy is the inter- from personal knowledge and existing organisational
face for all such activity. information resources, then sharing and subse-
An IDC White Paper by Feldman and Sherman quently action-taking by knowledge workers—
(2001) examined industry practices and indeed con- resulting in new information being added back to the
firmed the worst fears of practitioners: organisational memory. The collective knowledge of
an organization is diffused through several processes
Intranet technology, content and knowledge of knowledge acquisition, sharing, and action initi-
management systems, corporate portals, and ated as a consequence of new knowledge being cre-
workflow solutions have all generally improved ated. This flow is illustrated in Figure 1.
the lot of the knowledge worker. These tech- Nonaka and Takeuchi (1995) concluded that “The
nologies have improved access to information, Knowledge Creating Company” cannot create knowl-
but they have also created an information del- edge on its own without the initiative of the individual
uge that makes relevant information more dif- and the interactions that take place between individu-
ficult to find. als and groups. The effective design of the organiza-
https://doi.org/10.5771/0943-7444-2008-1-30
Generiert durch IP '46.4.80.155', am 20.01.2021, 04:51:15.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.32 Knowl. Org. 35(2008)No.1
R. S. Sharma, Sch. Foo, M. A. Morales-Arroyo. Developing Corporate Taxonomies for Knowledge Auditability
Figure 1: Knowledge processes in organisations.
tion makes it possible for the knowledge content of not the first to popularise these terms, their work
many of these interactions to be captured. Thus, per- nonetheless pointed to the stark differences in which
sonal and collective cycles of knowledge creation and the two types of knowledge are created, captured,
use are inter-related and as we shall see later in this ar- organised, stored, searched and transferred for re-
ticle, corporate taxonomies serve as useful intermedi- use. Hansen et al. (1999) concluded that whilst ex-
aries. plicit knowledge sharing, which they called codifica-
Figure 1 also shows some of these fundamental tion, comprises the most frequent in terms knowl-
knowledge processes within an organization. While edge transactions, most value was derived from tacit
there are several models for knowledge cycles, for knowledge sharing, which they called personalisa-
example Birkinshaw and Sheehan (2002)–capture, tion. This was primarily because tacit knowledge was
store, transfer; Feldman and Sherman (2001)–create, less common, more difficult to replicate and there-
distribute, manage, retrieve, apply; Mohanty and fore served as a competitive advantage. They further
Chand (2005)–create, capture, organise, store, use; suggested that superior IT infrastructure such as web
and the figure above is nevertheless a reasonable syn- portals and wireless LANs were only suited for codi-
thesis described in Foo et al. (2007). Knowledge fication; personalisation requiring opportunities for
workers create intellectual capital in the course of both traditional as well as IT-based communication
their work, or more precisely, as a result of their and collaboration.
work, and the extent to which this may be captured Several scholars have addressed the major chal-
is a measure of the organisation’s standard proce- lenges facing the sharing of knowledge within an or-
dures or structural capital. The knowledge infra- ganization. Nonaka and Takeuchi (1995) have also
structure within the organisation also drives how suggested the requirement of a shared workspace that
value in the form of re-exploitable knowledge is or- will facilitate knowledge sharing across the organisa-
ganised and stored. The competitive advantage of the tion. Zack (1999), in an empirical study of knowledge
organisation lies in the speed and precision with strategies in over 20 knowledge intensive firms, con-
which its knowledge assets are searched and trans- cluded that the misalignment of business and knowl-
ferred as and when opportunities arise. edge strategies was a fairly common cause for poor
At yet another layer of abstraction, as these proc- performance–for example, businesses that did not
esses continue in perpetuity within an organisation, know when value was being created and did not or-
two types of knowledge flow through–explicit, ganise themselves to continually exploit this value.
which is codifiable, and implicit or tacit, which is not Gupta and Govinderajan (2000) performed a more
codifiable but resides as expertise in the minds of extensive field investigation of knowledge flows
knowledge workers. The explicit vs. tacit knowledge within multinational corporations and found that the
dichotomy is often held as being analogous to struc- mismatch between the source and target of knowl-
tured (databases, web portals) vs. unstructured edge flows—for example, if knowledge was not trans-
knowledge (e-mail, blogs). However, this analogy is ferred appropriately—led to severe ineffectiveness.
misplaced as both explicit as well as tacit knowledge Hence the fundamental motivation for building a
may be structured or unstructured. In other words, corporate taxonomy is the realization that knowledge
explicit knowledge can be well articulated, especially is of little value unless it can be shared and re-used (as
in the written form, while tacit knowledge might be opposed to re-discovered) when opportunities for
much less so. While Nonaka and his co-authors were exploitation arise. There is considerable agreement
https://doi.org/10.5771/0943-7444-2008-1-30
Generiert durch IP '46.4.80.155', am 20.01.2021, 04:51:15.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.Knowl. Org. 35(2008)No.1 33
R. S. Sharma, Sch. Foo, M. A. Morales-Arroyo. Developing Corporate Taxonomies for Knowledge Auditability
that such taxonomy is the basis for interactions and Hence a corporate taxonomy may be viewed as a con-
organizational learning (Cheung et al. 2005; Daven- ceptual map, an information access tool, and a com-
port and Prusak 1998; Gilchrist and Kibby 2000; Gil- munications and training device at the same time,
christ 2001; Nonaka and Takeuchi 1995; Potter 2001). providing history, expertise and inside information
In the realm of digital resources, Noy and Mc- that can assist every business activity. Naturally, as
Guiness (2001) suggest that taxonomies are particu- with any other information access tool, a taxonomy
larly useful: (1) to share common understanding of has to serve special requirements and purposes before
the structure of information among people or soft- it is developed and exploited. Other classical perspec-
ware agents; (2) to enable reuse of domain knowl- tives widely accepted in the literature include:
edge; (3) to make domain assumptions explicit; (4)
to separate domain knowledge from the operational 1. A taxonomy is a creation of structure and labels
knowledge; (5) to analyze domain knowledge. to aid location of relevant information. A closer
The remainder of this article describes the build- definition might be the arrangement and labelling
ing blocks of a corporate taxonomy and presents a of metadata to allow primary data or information
synthesis of best practices for developing it so that it to be systematically managed and manipulated
may be continually updated so that it stays useful. (Gilchrist and Kibby 2000).
The development of good and relevant taxonomy 2. A taxonomy is a hierarchical presentation of in-
coupled with a supportive knowledge management formation that represents a specific knowledge
platform and environment is an integral aspect of domain. It includes several sub-topics that can
remaining competitive in the face of the continual contain two types of relations, namely, hierarchi-
deluge of (particularly, digital) knowledge. cal relations where one category is viewed as being
above another category, and non-hierarchical rela-
2. Principles of corporate taxonomies tions using links that indicates that a certain cate-
gory is related to another category. Applications
The word taxonomy is derived from Greek words are the navigation tools available to help users find
(taxis + nomos)—taxis is arrangement and nomos information (Graef 2001).
law—and can be conjugated to mean “the science of
classification.” The Swedish scientist Carl Linnaeus Taxonomies are often referred to as conceptual
(1707-1778) was perhaps the first to use the idea of knowledge maps in knowledge management. How-
taxonomy to classify the natural world. From its ori- ever, they have some distinguishing characteristics in
gins in the classification of living things, the idea of the sense that: (1) they support structure, content
taxonomy now has universal applications in group- and applications (navigational tools); (2) they are
ing knowledge so that it can be systematically devel- customised to reflect the language, culture and goals
oped, stored and re-used. In the information sci- of particular organisations; (3) they are often created
ences, the study of corporate taxonomies has been a using a combination of human effort and specialised
subject of considerable and longstanding interest software; they may refer to disparate information re-
among researchers (Cheung et al. 2005; Geisler sources such as e-mail, memoranda, documents,
2006; Gruber 1993; Noy and McGuinness 2001; books, part of books, reports and URLs; (4) they are
Saeeh and Chaudhry 2002) as well as practitioners usually created by multidisciplinary teams; and (5)
(Conway and Sligar 2002; Delphi 2002; Ernst & they are part of a process so that they are constantly
Young; Gilchrist and Kibby, 2000; Gilchrist 2001; refined and updated.
Greif 2001; Lehman 2003; Pepper 2000; Potter 2001; In either case, corporate taxonomies and knowl-
Woods 2004). edge maps may be considered the fundamental basis
A modernist definition of a corporate taxonomy for knowledge sharing in the organization and spe-
may be found in Lehman (2003 emphasis original): cific processes such as create, capture, organize,
store, search and transfer) in the organization. They
A taxonomy is a subject map to an organiza- provide a common understanding within the organi-
tion’s content. [It] reflects the organization’s zation that link to the knowledge cycle.
purpose or industry, the functions and respon- Corporate taxonomies are dynamic and need con-
sibilities of the persons or groups who need to stant refinement and update because organizations
access the content, and the purposes / reasons need to adapt to a changing environment (competi-
for accessing the content. tion, threats, etc.), which forces them to modify
https://doi.org/10.5771/0943-7444-2008-1-30
Generiert durch IP '46.4.80.155', am 20.01.2021, 04:51:15.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.34 Knowl. Org. 35(2008)No.1
R. S. Sharma, Sch. Foo, M. A. Morales-Arroyo. Developing Corporate Taxonomies for Knowledge Auditability
their knowledge flows. Grey (1999) suggests posing It may be viewed as the knowledge management
the following key questions to the knowledge work- equivalent of the requirements determination phase
ers within an organization in order to ascertain the undertaken during traditional systems analysis and
major knowledge flows: (1) What type of knowledge design.
is needed? (2) Who provides it and how does it ar- During the course of a knowledge audit, there is a
rive? (3) How is it improved and re-used? (4) What critical first step which leads to the creation of a
happens to new knowledge that is created? (5) What knowledge map--a visual representation of an organi-
prevents the organisation from doing more, better, sation’s knowledge. Technically, a knowledge map is a
faster? (6) How can knowledge flows (therefore) be logical abstraction of a corporate taxonomy, which
improved? includes implementation details such as how knowl-
This is in effect what is known as a knowledge audit edge assets are to be captured and indexed. A knowl-
(Cheung et al. 2007; Liebowitz et al. 2002; NLH edge map, at first cut, reveals possible answers to the
2005), which involves identifying what knowledge is key questions of a knowledge audit (outlined above).
needed, what knowledge already exists, where the gaps There are two recommended approaches to know-
lie, who needs the knowledge, and how it will be used. ledge mapping (NLH 2005): (1) map knowledge re-
Hylton (2002) more formally states that the knowl- sources and assets, showing what knowledge exists
edge audit (K-Audit) is an assessment of how the sum in the organisation and where it can be found; and
of explicit as well as tacit knowledge within an organi- (2) include knowledge flows, showing how that
sation is exploited throughout the knowledge-cycle knowledge moves around the organisation from
and the people and business processes add to such source to target. In both cases, the key is a diagram-
knowledge. More specifically (Hylton 2002, 2): matic schemata of corporate knowledge of the ex-
plicit as well as tacit nature and an accompanying re-
The knowledge audit process involves a thor- alisation of the value-added during the course of the
ough investigation, examination and analysis of knowledge flows. This may be derived from well-
the entire ‘life-cycle’ of corporate knowledge: known techniques such as process maps, class dia-
what knowledge exists and where it is, where grams, use cases and organisation charts.
and how it is being created and who owns it. It Building a knowledge map looks deceptively sim-
measures and assesses the level of efficiency of ple but perhaps requires more effort and resources
knowledge flow. From knowledge creation and than any other phase of developing a corporate tax-
capture, to storage and access, to use and dis- onomy. It is a profound, soul-search that involves
semination, to knowledge sharing and even the highest level of strategic management and do-
knowledge disposal, when the organisation is main expertise to make judgments on fundamental
no longer in need of particular elements of ex- business and knowledge strategies. One technique
plicit or codified knowledge. With respect to for deriving a knowledge map involves the use of the
people, the K-Audit measures the efficiency of so-called Boston Box suggested by Drew (1999).
transfer of tacit knowledge skills, when particu- Figure 2 shows four quadrants of the Boston Box for
lar skills or expertise is no longer needed. analysis of a complete coverage of an organisation’s
Figure 2: Building a knowledge map. Source: Drew (1999, 134)
https://doi.org/10.5771/0943-7444-2008-1-30
Generiert durch IP '46.4.80.155', am 20.01.2021, 04:51:15.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.Knowl. Org. 35(2008)No.1 35
R. S. Sharma, Sch. Foo, M. A. Morales-Arroyo. Developing Corporate Taxonomies for Knowledge Auditability
knowledge capital. Quadrant 1 asks what the core As a guide to content, a taxonomy has multiple entry
competencies of the organization are. Quadrant 3 points (such as business functions or product types),
addresses the unexploited seepage in its knowledge and will have the same element (lowest level class).
capital repository. Quadrant 2 takes the organiza- The consensus on what characterizes useful elements
tional learning impetus which seeks to position the of corporate taxonomies is the following:
organization to execute its strategic plans for
growth. Quadrant 4 refers to the blind spot of hid- 1. Elements are precise and do not overlap—the
den opportunities and threats that may not be (as closer to proper named elements at the lowest
yet) apparent within the organisation’s leadership. level, the better.
Daunting as this analysis may seem, it does not rep- 2. Elements are independent of the type of content,
resent a paradigm shift. The point being made in this and the organization structure (be they digital or
article is that the organisation of knowledge in the multilingual or distributed data).
form of a corporate taxonomy carries with it criteria 3. Elements reflect the access needs and expectations
for evaluating possible gaps as well as leaks that need of every constituency inside or outside the or-
to be plugged. Drew (op. cit.) had captured some of ganization; and,
these issues for some time now and the knowledge 4. Industry standards (such as UN/SPSC for prod-
management community has since developed an en- ucts and services. IBSN and ISSN for published
tire repertoire of tools for each of these quadrants documents) are recognized and applied whenever
(cf. Foo et al. 2007 for a textbook coverage of many possible.
of these tools).
What then makes a corporate taxonomy effective, To conclude, it is clear that corporate taxonomies
extendable and practical? Table 1 below, which is a have indispensable roles in the organization of busi-
compilation from Gilchrist (2001), Graef (2001), ness knowledge. The bottom-line for a good taxon-
Lehman (2003) and Woods (2004), offers eight per- omy is whether or not the knowledge sharing proc-
spectives or families of taxonomic elements, which ess is facilitated. There are methods and tools which
apply to an organization, although more perspectives help verify and validate that such sharing is indeed
do not necessarily translate to greater business effec- taking place. In the next section of this article, some
tiveness. of these building blocks are described.
Industry Segments - Marketing / Positioning / Com- 3. Building blocks for corporate taxonomies
petitive Intelligence Perspective; Industry Segments may
overlap with Products and Services. Some of the fundamental challenges on how knowl-
Organizational Functions - the organization breakdown edge (explicit as well as tacit) may be incorporated
of a business or organization by function or responsibil-
into a corporate taxonomy are addressed by a variety
ity
of techniques drawn from the domains of computer
Business Relationships - the intensities and types of
and information science. These include the concepts
other companies or organizations a business deals with;
including customers, vendors, regulators, associations, of directories of domain expertise; classification and
partners etc. clustering; indexing, tagging and the use of meta-
Business Issues & Events - economic, legal, M&A, regu- data. Classification is the technique used to organise
latory, environmental, labour, safety, other government a body of knowledge assets that reside within an or-
interfaces, etc. ganisation. It is supported by meta-data which are
Products & Services - products sold; MRO materials; used as keywords or descriptors for indexing, storing
indirect services, direct materials & services purchased. and searching knowledge assets.
Technologies - applicable to the industry or industries in The word “metadata” is derived from Greek and
which the firm participates. Basic or applied sciences are Latin words (Greek: Meta + Latin: Data). Since Me-
also included as appropriate. ta means along with, next or after, metadata is data
Geography – referring to location, particularly region or about data itself; it contains information about other
jurisdiction.
nuggets of information or knowledge. Metadata is
Document or Record Types - this perspective provides
documentation about documents and objects; they
valuable reduction of results based upon the document’s
purpose and its connection to the information need.
describe resources, indicate where they are located,
and outline what is required in order to use them
Table 1: Perspectives of taxonomic elements. successfully. In creating corporate taxonomies, a
https://doi.org/10.5771/0943-7444-2008-1-30
Generiert durch IP '46.4.80.155', am 20.01.2021, 04:51:15.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.36 Knowl. Org. 35(2008)No.1
R. S. Sharma, Sch. Foo, M. A. Morales-Arroyo. Developing Corporate Taxonomies for Knowledge Auditability
practitioner makes use of metadata to describe do- taxonomy (usually a visual representation) but is al-
cuments and other resources thereby enabling a ri- ways described in terms of a method or scheme that
cher means of defining the context of the resource groups a set of entities such that elements within a
and to provide more information access points to group (or class or cluster) are more similar to each
support information query and retrieval operations. other than elements in different groups. Clustering
This is a technique known as “tagging” in contempo- is the technical term used when these groups (or clu-
rary parlance and is very relevant to the idea of de- sters) are non-overlapping. In both cases, the organi-
scribing knowledge assets (whether codified or re- sation may be hierarchical and multi-faceted. The
siding within experts) and cataloguing them for stor- proliferation of Internet content and the require-
age and search. In this section, we discuss some of ment for optimised search engines has brought this
these. ancient science into yet another domain that sustains
its relevance. Hlava and ven Eman (1999) suggest
3.1 Classification, clustering and cataloguing that most cataloguing schemes, many of which
within the English speaking world originated from
A recent IDC study (Gantz et al. 2007) estimated the UK and US library communities, use one of 3
that the “the digital universe” equals approximately methods for classification: original text or idea; ex-
three million times the information in all the books isting vocabulary or topic; or a combination. Given
ever written, or the equivalent of 12 stacks of books, the volume of content to be organised, today much
each extending more than 93 million miles from the of this has to be automated (and continually refined
earth to the sun. The amount of information created manually) using idea extraction, keyword counts, or
and copied in 2010 will surge more than six fold, adaptive algorithms that discern document context.
from 161 to 988 exabytes, a compound annual A classification typically also results in what is
growth rate of 57%, nearly 70% of which will be gen- known as an ontology. Ontology is the term refer-
erated by individuals. Considering the exponential ring to the shared understanding of some domains of
growth of information and knowledge, particularly in interest, which is often conceived as a set of classes
the digital and Internet domains, it makes sense to (concepts), relations, functions, axioms and in-
classify content in some order so that search and re- stances. In the knowledge representation commu-
trieval becomes manageable (Feldman and Sherman nity, the highly cited definition is adopted from
2001). Gruber (1993, 199):
The study of classification is hence re-emerging
from a hiatus after the pioneering work of Rangana- An ontology is a formal, explicit specification of
than, the reknowned classificationist, who drew much a shared conceptualisation. Conceptualisation
inspiration from the work of Dewey and other pio- refers to an abstract model of phenomena in the
neers in order to formulate rules on how documents world by having identified the relevant concepts
might be classified so that they could be retrieved of those phenomena. Explicit means that the
with sufficient specificity when needed (cf. Rangana- type of concepts used, and the constraints on
than’s web repository for a retrospective). Several their use are explicitly defined. Formal refers to
contemporary studies have shown that using either the fact that the ontology should be machine
natural language, such as keyword searching, for de- readable. Shared reflects that ontology should
fined metadata fields, or a controlled vocabulary of capture consensual knowledge accepted by the
subject headings and browsing thesauri, result in su- communities.
perior knowledge retrieval and re-use when the
knowledge base is classified or clustered for effective Noy and McGuiness (2001, 1) offer a more IT-centric
search (Delphi 2002; Feldman and Sherman 2001; definition: “An ontology defines a common vocabu-
Stratify 1997; Williamson 1997). This would be par- lary for researchers who need to share information in
ticularly the case when end user tagging is enabled a domain. It includes machine-interpretable defini-
with the use of controlled vocabularies and multi- tions of basic concepts in the domain and relations
faceted taxonomies are constructed to facilitate the among them.”
search effort. It should be noted that classification and cata-
The idea of classification is frequently inter- loguing are active, dynamic activities that are never
changeably used with the term “taxonomy” but is complete, much like arranging the folders of one’s
semantically different—a classification may lead to a desktop. Hence, there is a continuous process of ap-
https://doi.org/10.5771/0943-7444-2008-1-30
Generiert durch IP '46.4.80.155', am 20.01.2021, 04:51:15.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.Knowl. Org. 35(2008)No.1 37
R. S. Sharma, Sch. Foo, M. A. Morales-Arroyo. Developing Corporate Taxonomies for Knowledge Auditability
pending, updating, pruning, and so on, to keep it re- order to assess the needs of two different kinds of in-
levant and useful. A classification is hierarchical and dustries, say the healthcare and manufacturing indus-
multi-faceted in order to support multiple perspec- tries, the tool should be flexible enough to cater to
tives such as user profiles, applications or data mod- both business contexts and vocabularies – categorise
els. When recognised that small corporate taxono- medical treatment and prescriptions as well as raw
mies comprise up to a thousand knowledge resource materials, equipment and facilities. Another desirable
items and large ones greater than twenty thousand property of such tools is scalability. It must become
(Woods 2004), it is easy to understand why an more accurate when the organisation gets larger and
automatic processor of metadata (often times extrac- not see drastic reductions in accuracy when dealing
tor) and classification rules is necessary. This is dis- with increasing knowledge resources. Last but not
cussed next. least, if the tool is easy to use and adopted very
quickly, it does not require extensive hours of train-
3.2. Automatic classifiers and other tools ing for employees in order for adoption to take place.
The past decade has seen a proliferation of tools
Automated classifiers typically use various proprie- for developing corporate taxonomies, many of which
tary clustering methods for analyzing the content of exhibit some or all of the following functionalities:
each knowledge asset and creating concept folders
that contain related items; organize the concept 1. Classification of digital resources, documents, da-
folders hierarchically based on their interrelation- tabases, directories, etc.
ships, and sort each document into one or more con- 2. Tagging of knowledge resources during content
cept folders that describe the document in whole or creation.
in part (cf. Stratify 2006). With codified knowledge, 3. Site navigation and creating categories for discov-
such clustering is typically based on a statistical ery of information through browsing.
analysis of all words within a document and extract- 4. Identification and retrieval including cross search-
ing keywords or context. Clustered documents are ing of different resource bases.
placed within concept folders that contain docu- 5. Personalisation and delivery of information.
ments that are “close” or similar, to each other ac- 6. Visualisation of knowledge maps.
cording to a chosen similarity measure which at-
tempts to match the term lists or subject headings One of the leading and authoritative web site on
(controlled or otherwise) of the documents to a clu- search engines, SearchTools.com (http://www.search
ster. During search and retrieval, a centroid or repre- tools.com/info/classifiers-tools.html) provides a list
sentative from each cluster is matched with the re- of valuable and up-to-date resources on “Tools for
quirements of the query, and the entire cluster is re- Taxonomies, Browse-able Directories, and Classify-
trieved if held similar in the expectation that they are ing Documents into Categories.” Vendors specialise
similar and hence must be equally relevant. In this in different subsets of the above functionalities and
manner the search space is reduced. there is really no ubiquitous solution that may be
Functionalities aside, classifiers and taxonomy adopted by the organisation developing a corporate
builders exhibit a commonality of traits. For exam- taxonomy. The following are some of the leading
ple, most tools should be able to measure the depth vendors of taxonomy tools which support the key
of a person’s experience on a particular topic, based functionalities and together possess a dominant
on relevant prior experiences, roles in projects, peer market-share.
recognition and other measurements and track rele-
vant end user activity, identifying those individuals Company Tool URL
who may be best suited to address the task. They Autonomy Expertise www.autonomy.com
should also incorporate automatic learning functions Finder
where the program becomes more accurate with con-
Convera SAAS Seman- www.convera.com
tinuing usage, hence removing the need to rely on tic Search
manual user feedback.
They typically allow for customisation such that it Cadenza Knowledge- www.cadenzainc.com
LEAD
is flexible enough to accommodate the different
knowledge management needs of different environ- Divine Athena and www.divine.com
ments in which it will be deployed. To illustrate, in MindAlign
https://doi.org/10.5771/0943-7444-2008-1-30
Generiert durch IP '46.4.80.155', am 20.01.2021, 04:51:15.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.38 Knowl. Org. 35(2008)No.1
R. S. Sharma, Sch. Foo, M. A. Morales-Arroyo. Developing Corporate Taxonomies for Knowledge Auditability
Company Tool URL taxonomy design, greater manual control may
Entrieva Semio Taxon- www.entrieva.com
be preferred.
(formerly omy and Tag-
Semio) ger Although these tools serve the common purpose of
Groove Groove Virtual www.groove.net
organising knowledge and support for information
Networks Office seeking and discovery, they all have their own subtle
(now part characteristics. The question is not about “natural
of Micro- language or controlled vocabulary,” rather the solu-
soft) tion is “natural language and controlled vocabular-
Hum- Hummingbird www.hummingbird.com ies” to complement each other for information seek-
mingbird Knowledge ing. Several of these vendors have provided a range
Server of organisation tools such as taxonomies, subject di-
Kamoon Kamoon www.kamon.com rectories, subject hierarchies, topic maps and knowl-
Connect edge mapping tools in order to assist in the effective
IBM (Lo- Intelligent www.lotus.com management of content in the organisation’s Intra-
tus) Miner, Discov- net, Internet or knowledge portals. These tools pro-
ery Server and vide hierarchical or decision tree structures with
K-station considerable variation in complexity and sophistica-
Microsoft Sharepoint www.microsoft.com tion.
It is also apparent that many of these tools relate
MITi Readware www.readware.com
Knowledge to industry-specific taxonomies. The tool should be
Workshop such that it should preferably be able to “under-
stand” user requests and derive similar requests in
Quiver QKS Classifier www.quiver.com
order to return more relevant results. “Knowing
Sopheon Organik www.sopheon.com what you know” is the central objective of a corpo-
Stratify Stratify Dis- www.stratify.com rate taxonomy and hence taxonomy building tools.
covery System In order to select appropriate taxonomy tools that
Verity Verity Knowl- www.verity.com meet specific features or functions, there needs to be
(now part edge Organiser a basis for comparison. Table 5 is a list of functional-
of Auton- & Ultraseek ities and types of features synthesised from the lit-
omy Advanced erature. A matrix of the availability of these func-
Group) Classifer
tionalities and features within some of the leading
XBRL XBRL Taxon- www.xbrlsolutions.com taxonomy tools is given in Foo et al. (2007) with the
omy Builder objective of rapid prototyping (and subsequent evo-
Table 4: Taxonomy tools in action. lution) of an organisation’s corporate taxonomy.
An examination of the approaches taken in many of Classification Methods: Rule-based, Training Sets, Sta-
tistical Clustering, Manual
these tools confirms that vendors disagree over the
choice of techniques and approaches to the classifi- Classification Technologies: Linguistic Analysis, Neural
Network, Bayesian Analysis, Pattern Analysis/ Match-
cation of information and knowledge. The approach
ing, K-Nearest Neighbours, Support Vector Machine,
that suits any individual organisation will naturally XML Technology, and others
depend on a mixture of the business requirement, Approaches to Taxonomy Building: Manual, Automatic,
the type, volume and volatility of the information to Hybrid
be managed, the skills available in-house and the Visualization Tools Used: Tree/Node, Map, Star, Folder,
time and resources to be expended. But, as Woods None
(2004, 7) suggests: Depth of The Taxonomy: > 3 levels
Taxonomy Maintenance: Add/Create, Modify/Rename,
The greater the volatility of the information Delete, Reorganisation/Re-categorization, View/Print
and the categories to be used, and the less the Cross-referencing Support:
in-house experience available, the more attrac- Import/Export Taxonomy:
tive an automated solution will be. For organi- Import/Export Formats Support: Text file, XML for-
sations with extensive experience of in-house mat, RDBS, Excel file, Others
https://doi.org/10.5771/0943-7444-2008-1-30
Generiert durch IP '46.4.80.155', am 20.01.2021, 04:51:15.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.Knowl. Org. 35(2008)No.1 39
R. S. Sharma, Sch. Foo, M. A. Morales-Arroyo. Developing Corporate Taxonomies for Knowledge Auditability
Document Formats Support: HTML, MS Office docu- organisation operates dictates the more suitable ap-
ment, ASCII/text file, Adobe PDF, E-mail, and Others proach. In closing, the selection of an appropriate
Personalization: Personalized View, Alert- tool is perhaps a matter of efficiency to the trained
ing/Subscribing and experienced organisation but effectiveness as
Product integration: Search Tools, Administration well to many taking the first steps towards a corpo-
Tools, Portals, Legacy Applications (e.g. CRM) rate taxonomy. The final section will note that the
Industry-specific Taxonomy: Business, News, Medi- use of appropriate (and k-auditable) tools is a salient
cine/Pharmaceutical, Legal, Military, Biotech, Technol-
point in the continual evolution of the corporate ta-
ogy, Insurance, Government, Any industry
xonomy in support of a learning organisation.
Access Points to the Information: Browse Categories,
Keywords, Concepts & Categories Searching, Top-
ics/Related Topics Navigation, Navigate Alphabetically, 3.3 Expertise locators
Enter Queries, and others
Multilingual Support: Where knowledge is not explicit, pointers to their
Product Platforms: Window NT/2000, Linux/Unix, Sun (tacit) sources are needed. To this end, expertise lo-
Solaris system, and others cators, euphemistically known as electronic yellow
Table 5: Functionalities of taxonomy builders and classifiers. pages or skills directories, have emerged in the cor-
porate world as means to identify sources of specific
The availability of a particular function and existence expertise and skills. These can be classified or cate-
of a specific feature makes a significant impact on gorized to be included in corporate taxonomies.
the development of a corporate taxonomy. Besides Grey (1999) suggests that corporate yellow pages of-
the obvious requirements fit in terms of Product ten incorporated into taxonomies give the highest
Platform and Integration, GUI Design, Access return on investment and it is easy to understand
Points, Import / Export and Multilingual Support, why. The nature of knowledge work is such that
and so on, there are other nuanced considerations. most professionals turn to their peers as the first re-
For example, the easy part of taxonomy implementa- sort (Davenport and Prusak 1998).
tion is the actual assignment of rules, either from a Expertise locators are basically lists of the exper-
written “cookbook” for human classifiers or soft- tise of individuals, usually very highly regarded sub-
ware. There are basically two approaches to software ject matter experts, in an organisation. Expertise re-
implementation: those packages that accept and exe- fers to special skills or knowledge of subject embed-
cute rules, and those packages that use statistical ded in individuals. Hence, an expertise locator be-
techniques (“content like this”) to construct their comes a de facto directory of individuals in an organi-
own rules. While the statistical vendors can fairly sation with their contact details, designation, and
claim that their products avoid the “rigor” of classi- name, along with details about their knowledge, skill
fication definition, they also miss the precision of sets and experiences. They identify experts in an or-
rules and the result vagueness that accompanies va- ganisation that are authorities in specific knowledge
gue rules. But this may well be the “lesser of the two domains. A subject matter expert may have different
evils” in instances where there is a voluminous mass levels of expertise in different topics and all this goes
of legacy documents and knowledge items. Hence a into the listing. Such expertise is typically self-
“middle way” is the possible use of a statistical or reported (on the basis of job responsibilities or quali-
learning type classification approach after having fications) but sometimes discerned through formal
classified a large and varied group of document- accreditation or using social network analysis (SNA).
records, with a manually defined rule-based tech- SNA is a complex yet highly rewarding activity in
nique. In either case, the resulting classification is k- knowledge organisation. For example, by observation
auditable and may be changed and improved with of email trails or posts on discussion groups, it be-
time. comes apparent who junior team members turn to for
Ongoing maintenance of classification rules is an- various types of advice during projects or within a
other significant but tractable activity, either in- community of practice–this is where the re-useable
house or from third party specialists. Rule-based tacit knowledge within the organization lies.
classifier software will require very low maintenance. The National Electronic Library for Health
Statistical classifier software needs regular re- (2003), part of the National Health Services of the
calibration to address new or modified classification United Kingdom (www.nhs.uk) which manages a
rules. Again, the dynamic environment in which the large number of diverse medical facilities and special-
https://doi.org/10.5771/0943-7444-2008-1-30
Generiert durch IP '46.4.80.155', am 20.01.2021, 04:51:15.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.40 Knowl. Org. 35(2008)No.1
R. S. Sharma, Sch. Foo, M. A. Morales-Arroyo. Developing Corporate Taxonomies for Knowledge Auditability
ists, suggests the following tags as meta-data for the ing complexity of tasks and work that requires the
identification of (tacit) expertise within an organisa- combined skills of various experts often situated
tion: across the globe. The specific benefits of expertise
locators are that they are easily implemented; help in
Name; Job title; Department or team; Brief job locating the knowledge source; allow people to in-
description – current and past; Relevant profes- teract and learn efficiently by providing a platform;
sional qualifications; Uploaded CV in standard save time and effort in not “reinventing the wheel.”
format; Areas of knowledge and expertise se- As the organisation’s base of information about
lected from a pre-defined list of subjects and its subject matter experts grows, so will its ability to
terms with self-reported rankings - “extensive”, harness its knowledge capital. With the information
“working knowledge”, “basic”; Main areas of about an individual's browsing, searching, and post-
interest; Key contacts – both internal and ex- ing habits; metadata from the content that has been
ternal; Membership of communities of practice created; the projects worked on; community in-
and other knowledge networks; Personal pro- volvement; and, preferences and patterns in the data
file; Photograph; Contact information. accessed within the corporate intranet—the Enter-
prise Knowledge Portal (EKP) can manage and de-
Among other critical success factors for the design, scribe the knowledge worker's experience base in the
development and maintenance of such an expertise same way it shows the explicit contents of the re-
listing, the NELH lists currency, auditability and pository. With the aforementioned understanding of
evolution as the most vital. the building blocks of corporate taxonomies, we
Conway and Sligar (2002) remark that capturing conclude this article with a framework for building
such information within the organization may be a taxonomies in the final section.
little more controversial than doing it in an external
Internet environment. The corporate culture will li- 4. Framework for taxonomy building
kely influence or constrain the reaction to and accep- and knowledge-auditability
tance of gathering, mining, manipulating, storing,
and making use of what may be regarded as “per- To recap, corporate knowledge taxonomies play a
sonal” information. Nevertheless, this approach can critical role in knowledge management and the ex-
be one of the best ways of connecting colleagues and ploitation of corporate intellectual capital with direct
team members and for the corporation to begin to impact on the organisation’s performance and
learn more about the tacit knowledge residing within growth. They provide a platform that assists em-
its people. ployees to seek knowledge residing within the or-
Expertise locators are useful for large organisa- ganisation and sometimes beyond, facilitate collabo-
tions which are spread geographically. While they are ration and interactions, and support the iterative
technologically simple to develop, it can effectively process that may be necessary to develop new
assist organisations to “know what they know”. As knowledge. At the onset, the corporate taxonomy
the knowledge community increasingly recognises can provide a “knowledge map” to enable navigation
that the best channel for knowledge sharing is com- of and access to the intellectual capital of the organi-
munication between co-workers and the most effec- sation. Corporate taxonomies work at the level of
tive networking protocol is collaboration, an exper- information management by connecting people to
tise locator provides the identification service to documents, and at the knowledge level by connect-
generate the connection and to support relationships ing people to people (Gilchrist and Kibby 2001). It
building between individuals. It provides a means to is critical that the design of such taxonomy must al-
access a key component of the organisation’s intel- low itself to be audited so that the impact of knowl-
lectual assets. Through such connectivity, it supports edge management may be assessed.
learning, growing and capacity development. Of Based on the discussion in the previous section,
course, finding the right person is a means to an end, we now present a framework comprising four major
not an end itself, and the organisation culture and in- steps required to build and maintain a corporate tax-
centives must support knowledge creation, capture, onomy. These are:
transference and re-use.
Expertise locators are also useful to locate rele- 1. Conduct a knowledge audit that results in a first-
vant knowledge sets particularly due to the increas- cut knowledge map;
https://doi.org/10.5771/0943-7444-2008-1-30
Generiert durch IP '46.4.80.155', am 20.01.2021, 04:51:15.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.Knowl. Org. 35(2008)No.1 41
R. S. Sharma, Sch. Foo, M. A. Morales-Arroyo. Developing Corporate Taxonomies for Knowledge Auditability
2. Perform a more-intensive requirement analysis driven framework for first understanding and then
that formalises tagging, storage and search; developing corporate taxonomies for effective ex-
3. Select the appropriate set of tools that facilitates ploitation of an organisation’s valuable knowledge
the creation of the taxonomy and the classifica- resources. The net result of such integration is a dy-
tion process, and, namic and relevant corporate taxonomy–what Gru-
4. Refine and update the taxonomy as the organisa- ber (1993) calls a “portable ontology.”
tional context changes. Figure 4 is such a framework which incorporates
the key concepts discussed with respect to developing
corporate taxonomies. The framework also maps
classical knowledge flows (create, capture, organise,
store, search, transfer and re-use) and inventories
(documents, expertise directories, learning communi-
ties) with the design of the corporate taxonomy using
automated builders and knowledge mobilisation
(search and re-use). In such a scenario, it is obvious
that neither the knowledge flows nor the inventories
would be static. Hence it becomes crucial that a
methodology for creating knowledge taxonomies
adequately supports the notion of continual growth
and consequently, auditability. From this framework,
Figure 3: Presenting knowledge-audits as ontologies. Source: we have derived the four major steps that need to be
Perez-Soltero et al. (2006, 47). undertaken to build corporate taxonomies with the
design objective of knowledge auditability.
The nexus between a knowledge audit and creating a It is worth repeating that although many practi-
corporate taxonomy is indeed a symmetric one. tioners (cf. Hylton 2002; Lehman 2003; Pepper
Perez-Soltero et al. (2006) have presented a method- 2000) refer to corporate taxonomies and knowledge
ology which results in a knowledge inventory com- maps interchangeably, it should be clear by now to
prising knowledge maps and knowledge flows that the discerning reader that they are indeed distinct in
identify inefficiencies reflected in duplication of ef- the level of detail they carry. At its simplest, a taxon-
forts, knowledge gaps, knowledge barriers and omy is a rule-driven hierarchical organisation of ca-
knowledge-bottlenecks. They show the feasibility of tegories used for classification purposes with the ap-
using ontologies as representational schemas in order propriate subject headings and descriptors. However,
to formally present the results of a knowledge audit such a simple definition hides the many challenges to
which address the problems of knowledge leakage be faced in building and maintaining an effective and
and additionally the benefits of re-using valuable usable taxonomy for the organisation (Woods 2004).
knowledge. Figure 3 is an abstraction of their meth- Corporate taxonomies are particularly used by the
odology. Whilst such an approach makes sense from various enterprise information systems to permit in-
the point of schematic representation of the results of stant access to appropriate information, where there
a knowledge audit for the purpose of communicating are voluminous data, and information needs to be
with stakeholders, we argue that the opposite is even managed carefully. Knowledge maps are at best visual
more critical. Cheung et al. (2007) in fact adopt such aids that help the search and retrieval process. They
a methodology for knowledge audits but do not for- are the result of what has been described in an earlier
mally specify the link between building a knowledge section as a knowledge audit – the technical details
taxonomy and auditing the repository of knowledge of which are beyond the scope of this article.
within an organisation. The design of a corporate Nevertheless, Step One of developing a corporate
taxonomy must necessarily take into account the ease taxonomy is therefore to conduct a knowledge audit
of auditing knowledge inventories, flows, leakages which clearly identifies the creation, capture, organi-
and gaps, and must facilitate the continual growth of zation, storage, search, transfer and re-use of knowl-
the knowledge or learning organisation. edge in the critical business processes of the organi-
Synthesising the numerous approaches from re- zation. Conceptually, this is indeed the most com-
search and practice on taxonomies, classification and plex step as the design is not easily amenable to vali-
ontologies, we have developed a knowledge-cycle dation. Grey (1999) suggests that the key to devel-
https://doi.org/10.5771/0943-7444-2008-1-30
Generiert durch IP '46.4.80.155', am 20.01.2021, 04:51:15.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.You can also read
